System and method for assessing disease load and progression

Abstract

Systems and methods are presented that provide semi-automatic and / or automatic analysis of medical image data to determine and / or communicate metrics that provide risk and / or disease pictures of a patient. Techniques described herein include systems and methods for analyzing medical image data to assess quantitative metrics that provide snapshots of a patient's disease load at a particular time, and / or for analyzing images taken over time to produce a longitudinal data set, the longitudinal data set provides a picture of how the patient's risk and / or disease evolves over time during monitoring and / or treatment response. The metrics calculated by the image analysis tools described herein can itself be used as quantitative metrics of disease load and / or can be linked to clinical endpoints attempting to measure and / or rank patient outcomes. Thus, the image analysis techniques of the present disclosure can be used to notify clinical decisions, evaluate therapeutic efficacy, and predict patient response.

Description

Cross-references to related applications

[0001] This application claims priority to and the benefit of U.S. Provisional Application No. 63 / 350,211 filed on June 8, 2022, U.S. Provisional Application No. 63 / 458,031 filed on April 7, 2023, and U.S. Provisional Application No. 63 / 461,486 filed on April 24, 2023, the contents of each of which are incorporated herein by reference in their entirety. Technical Field

[0002] The present invention generally relates to systems and methods for generating, analyzing and / or presenting medical image data. More specifically, in certain embodiments, the present invention relates to systems and methods for automatically analyzing medical images to identify and / or characterize cancerous lesions and / or prognosis or risk of an individual. Background Art

[0003] Nuclear medicine imaging involves the use of radioactively labeled compounds, called radiopharmaceuticals. Radiopharmaceuticals are administered to patients and accumulate in various areas of the body in a manner that depends on and therefore indicates the physiological and / or biochemical properties of the tissue therein (e.g., those properties affected by the presence and / or condition of a disease such as cancer). For example, certain radiopharmaceuticals after administration to a patient accumulate in areas of abnormal bone formation associated with malignant bone lesions, which indicates cancer metastasis. Other radiopharmaceuticals can bind to specific receptors, enzymes, and proteins in the body that change during the evolution of the disease. After administration to the patient, these molecules circulate in the blood until they find their intended target. The bound radiopharmaceutical remains at the site of the disease, while the remainder of the dose is cleared from the body.

[0004] Nuclear medicine imaging techniques retrieve images by detecting radiation emitted from the radioactive portion of the radiopharmaceutical. The accumulated radiopharmaceutical acts as a beacon, making it possible to obtain images depicting the location and concentration of the disease using commonly available nuclear medicine modalities. Examples of nuclear medicine imaging modalities include bone scan imaging (also known as scintigraphy), single-photon emission computerized tomography (SPECT), and positron emission tomography (PET). Bone scan, SPECT, and PET imaging systems are found in most hospitals around the world. The choice of a particular imaging modality depends on and / or indicates the particular radiopharmaceutical used. For example, a radiopharmaceutical that is ionized with technetium-99m ( 99m Compounds labeled with Tc are compatible with bone scan imaging and SPECT imaging, while PET imaging usually uses fluorinated compounds labeled with 18F.99m Tc Methylene Diphosphonate ( 99m Tc MDP) is a popular radiopharmaceutical used in bone scan imaging to detect metastatic cancer. Targeted radiolabeled prostate specific membrane antigen (PSMA) compounds, such as 99m Tc-labeled 1404 and PyL TM (also known as [18F]DCFPyL) can be used for SPECT and PET imaging, respectively, and offer the possibility of highly specific prostate cancer detection.

[0005] Therefore, nuclear medicine imaging is a valuable technology that provides doctors with information that can be used to determine the presence and extent of disease in a patient. Doctors can use this information to provide patients with a recommended course of treatment and to track the progression of the disease.

[0006] For example, an oncologist may use nuclear medicine images from a patient study as input to an assessment of whether the patient has a particular disease, such as prostate cancer; what stage of the disease is evident; what is the recommended course of treatment, if any; whether surgical intervention is indicated; and the likely prognosis. An oncologist may use a radiologist report in this assessment. A radiologist report is a technical assessment of a nuclear medicine image prepared by a radiologist for a physician who requires an imaging study, and includes, for example, the type of study performed, the clinical history, comparisons between images, the techniques used to perform the study, the radiologist's observations and findings, and the overall impression and recommendations that the radiologist may make based on the results of the imaging study. The signed radiologist report is sent to the physician, the study is ordered for the physician's review, and a discussion of the results and recommendations for treatment is then conducted between the physician and the patient.

[0007] Thus, the method involves having a radiologist perform an imaging study on a patient, analyzing the resulting images, generating a radiologist report, forwarding the report to the requesting physician, having the physician formulate an evaluation and treatment recommendation, and having the physician communicate the results, recommendations, and risks to the patient. The method may also involve repeating the imaging study due to an inconclusive result or ordering other tests based on the initial result. If the imaging study shows that the patient has a particular disease or condition (e.g., cancer), the physician discusses different treatment options (including surgery) and the risks of doing nothing or taking a cautious wait or active surveillance approach rather than surgery.

[0008] Therefore, methods of reviewing and analyzing multiple patient images over time play a key role in the diagnosis and treatment of cancer. There is a clear need for improved tools that facilitate and improve the accuracy of image review and analysis for cancer diagnosis and treatment. Improving the toolkit used by physicians, radiologists and other healthcare professionals in this way provides significant improvements in standard of care and patient experience. Summary of the invention

[0009] Systems and methods are presented herein that provide semi-automatic and / or automatic analysis of medical image data to determine and / or communicate metrics of risk and / or disease conditions for patients. The techniques described herein include systems and methods for analyzing medical image data to assess quantitative metrics that provide snapshots of a patient's disease load at a particular time, and / or systems and methods for analyzing images taken over time to generate longitudinal data sets that provide the evolution of a patient's risk and / or disease over time during monitoring and / or treatment response. The metrics calculated by the image analysis tools described herein can themselves be used as quantitative metrics of disease load and / or can be related to clinical endpoints that attempt to measure and / or stratify patient outcomes. Therefore, the image analysis techniques disclosed herein can be used to provide information for making clinical decisions, evaluating treatment efficacy, and predicting patient responses.

[0010] In certain embodiments, the value of a patient index that quantifies disease burden is calculated by analyzing a 3D nuclear medicine image of an individual to identify and quantify sub-regions (referred to as hot spots) that indicate the presence of potential cancerous lesions. Multiple quantitative metrics of individual hot spots can be calculated to reflect the severity and / or size of the potential lesions they represent. These individual hot spot quantitative metrics can then be aggregated to calculate values ​​for various patient indexes that provide a measure of disease burden and / or risk for the individual as a whole and / or within a specific tissue region or lesion subclass.

[0011] In one aspect, the present invention relates to a method for automatically processing a 3D image of an individual to determine the value of one or more patient indicators that measure the individual's (e.g., overall) disease burden and / or risk, the method comprising: (a) receiving, by a processor of a computing device, a 3D functional image of the individual obtained using a functional imaging modality; (b) segmenting, by the processor, a plurality of 3D hotspot volumes within the 3D functional image, each 3D hotspot volume corresponding to a local region having elevated intensity relative to its surroundings and representing a potential cancerous lesion within the individual, thereby obtaining a set of 3D hotspot volumes; (c) calculating, by the processor, for each specific metric in one or more individual hotspot quantification metrics, a value of a specific individual hotspot quantification metric for each 3D hotspot volume of the set; and (d) determining, by the processor, the value of one or more patient indicators, wherein each of at least a portion of the patient indicators is associated with one or more specific individual hotspot quantification metrics, and is based on at least a portion (e.g., substantially all; e.g., a specific subset) of the values ​​of the one or more specific individual hotspot quantification metrics calculated for the set of 3D hotspot volumes.

[0012] In some embodiments, at least one specific patient indicator of one or more patient indicator values ​​is associated with a single specific individual hotspot quantification metric and is calculated (e.g., mean, median, mode, sum, etc.) based on substantially all values ​​(e.g., all; e.g., excluding only statistical outliers) of the specific individual hotspot quantification metric calculated for the set of 3D hotspot volumes.

[0013] In some embodiments, the single specific individual hotspot quantification metric is an individual hotspot intensity metric that quantifies intensity within a 3D hotspot volume (eg, calculated from intensities of voxels of the 3D hotspot volume for the individual 3D hotspot volume).

[0014] In some embodiments, the individual hotspot intensity metric is an average hotspot intensity (eg, calculated for an individual 3D hotspot volume as an average of the intensities of the voxels within the 3D hotspot volume).

[0015] In certain embodiments, a specific patient index is calculated as the sum of substantially all values ​​of the individual hotspot intensity metrics calculated for the set of 3D hotspot volumes.

[0016] In some embodiments, the single specific individual hotspot quantification metric is the lesion volume (eg, calculated for a specific 3D hotspot volume as the sum of the volumes of the individual voxels within the specific 3D hotspot volume).

[0017] In certain embodiments, the patient-specific index (the value of) is calculated as the sum of substantially all lesion volume values ​​calculated for the set of 3D hotspot volumes (eg, such that the patient-specific index value provides a measure of the total lesion volume within an individual).

[0018] In some embodiments, a particular one of the one or more overall patient indices is associated with two or more particular individual hotspot quantification metrics and is calculated based on substantially all values ​​of the two or more particular individual hotspot quantification metrics calculated for the set of 3D hotspot volumes (e.g., a weighted sum, a weighted average, etc.).

[0019] In certain embodiments, the two or more specific individual hotspot quantification metrics include (i) an individual hotspot intensity metric and (ii) a lesion volume.

[0020] In certain embodiments, the individual hotspot intensity metric is an individual lesion indicator that maps values ​​of hotspot intensity to values ​​on a standardized scale.

[0021] In some embodiments, a specific patient indicator (the value of which) is calculated as the sum of intensity-weighted lesion (e.g., hotspot) volumes by: for each of substantially all 3D hotspot volumes, weighting the value of the lesion volume by the value of a separate hotspot intensity metric (e.g., calculating the product of the lesion volume value and the value of the separate hotspot intensity metric), thereby calculating multiple intensity-weighted lesion volumes; and calculating the sum of substantially all intensity-weighted lesion volumes as the value of the specific patient indicator.

[0022] In some embodiments, the one or more individual hotspot quantification metrics include one or more individual hotspot intensity metrics that quantify intensity within a 3D hotspot volume (eg, calculated from intensities of voxels of the 3D hotspot volume for the individual 3D hotspot volumes).

[0023] In some embodiments, one or more individual hotspot quantification metrics include one or more members selected from a group consisting of: average hotspot intensity (e.g., for a particular 3D hotspot volume, calculated as the average of the intensities of the stereo pixels within the particular 3D hotspot volume); maximum hotspot intensity (e.g., for a particular 3D hotspot volume, calculated as the maximum value of the intensities of the stereo pixels within the particular 3D hotspot volume); and median hotspot intensity (e.g., for a 3D hotspot volume, calculated as the median of the intensities of the stereo pixels within the 3D hotspot volume).

[0024] In some embodiments, one or more separate hotspot intensity metrics include a peak intensity of a 3D hotspot volume [e.g., where for a particular 3D hotspot volume, the value of the peak intensity is calculated by: (i) identifying a maximum intensity stereo pixel within the particular 3D hotspot volume; (ii) identifying stereo pixels within a sub-region around the maximum intensity stereo pixel (e.g., including stereo pixels within a particular distance threshold of the maximum intensity stereo pixel) and stereo pixels within the particular 3D hotspot; and (iii) calculating an average of the intensities of the stereo pixels within the sub-region as the corresponding peak intensity].

[0025] In certain embodiments, the one or more individual hotspot intensity metrics include an individual lesion index that maps values ​​of hotspot intensity to values ​​on a standardized scale.

[0026] In certain embodiments, the method includes: identifying, by a processor, one or more 3D reference volumes (reference volumes) in a 3D functional image, each corresponding to a specific reference tissue region; determining, by the processor, one or more reference intensity values, each of which is associated with a specific 3D reference volume of the one or more 3D reference volumes and corresponds to a measure of intensity within the specific 3D reference volume; and in step (c), for each 3D hotspot volume in the set, determining, by the processor, a corresponding value of a specific individual hotspot intensity measure (e.g., average hotspot intensity, median hotspot intensity, maximum hotspot intensity, etc.); and determining, by the processor, a corresponding value of an individual lesion indicator based on the corresponding value of the specific individual hotspot intensity measure and one or more reference intensity values.

[0027] In some embodiments, the method includes: mapping each of one or more reference intensity values ​​to a corresponding reference index value on a scale; and determining, for each 3D hotspot volume, a corresponding value of an individual lesion index using the reference intensity value and the corresponding reference index value to interpolate the corresponding individual lesion index value on the scale based on the corresponding value of the specific individual hotspot intensity metric.

[0028] In certain embodiments, the reference tissue region comprises one or more members selected from the group consisting of: liver, aorta, and parotid gland.

[0029] In some embodiments, the first reference intensity value (i) is a blood reference intensity value associated with a reference volume corresponding to a portion of the aorta, and (ii) is mapped to a first reference index value; the second reference intensity value (i) is a liver reference intensity value associated with a reference volume corresponding to a liver, and (ii) is mapped to a second reference index value; and the second reference intensity value is greater than the first reference intensity value and the second reference index value is greater than the first reference index value.

[0030] In some embodiments, the reference intensity values ​​include a maximum reference intensity value mapped to a maximum reference index value, and wherein a 3D hotspot volume having a corresponding value of a particular individual hotspot intensity metric greater than the maximum reference intensity value is assigned an individual lesion index value equal to the maximum reference index value.

[0031] In certain embodiments, the method includes: identifying one or more subsets within a set of 3D hotspot volumes, each of which is associated with a specific tissue region and / or lesion classification; and for each specific subset of the one or more subsets, calculating a corresponding value of one or more specific patient indicators using the value of a separate hotspot quantification metric calculated for the 3D hotspot volume within the specific subset.

[0032] In certain embodiments, the one or more subsets are associated with a specific region of the one or more tissue regions and the method includes identifying, for each specific tissue region, a subset of 3D hot spot volumes that are located within the volume of interest corresponding to the specific tissue region.

[0033] In certain embodiments, the one or more tissue regions comprise one or more members selected from the group consisting of: a skeletal region comprising one or more bones of the individual, a lymphatic region, and a prostate region.

[0034] In certain embodiments, each of the one or more subsets is associated with a particular type of one or more lesion subtypes [e.g., according to a lesion classification scheme (e.g., miTNM classification)], and the method includes determining a corresponding lesion subtype for each 3D hotspot volume and assigning the 3D hotspot volumes to the one or more subsets according to their corresponding lesion subtypes.

[0035] In certain embodiments, the method includes using at least a portion of the values ​​of one or more patient indicators as input to a prognostic model (e.g., a statistical model, such as a regression; such as a classification model, whereby a patient is assigned to a particular category based on a comparison of one or more patient indicator values ​​to one or more thresholds; such as a machine learning model, in which the values ​​of one or more patient indicators are received as input) that produces as output an expected value and / or range (e.g., category) indicative of possible values ​​for a particular patient outcome (e.g., time, such as in months, representing expected survival, time to progression, time to radiographic progression, etc.).

[0036] In certain embodiments, the method comprises using at least a portion of the values ​​of one or more patient indicators as input to a predictive model (e.g., a statistical model, such as regression; such as a classification model, whereby a patient is assigned to a particular category based on a comparison of one or more patient indicator values ​​to one or more thresholds; such as a machine learning model, in which the values ​​of one or more patient indicators are received as input), which generates a prediction model for one or more treatment options (e.g., Abiraterone, Enzalutamide, Apalutamide, Darolutamide, Sipuleucel-T, Ra223, Docetaxel, Carbazitaxel, Pembrolizumab, Olaparib, Rucaparib, 177The invention also provides an output of an eligibility score for each of a plurality of therapeutic options and / or classes of therapeutic agents [e.g., androgen biosynthesis inhibitors (e.g., abiraterone), androgen receptor inhibitors (e.g., enzalutamide, apalutamide, darutamide), cellular immunotherapy (e.g., siplucet-T), internal radiation therapy (Ra223), anti-tumor drugs (e.g., docetaxel, cabazitaxel), immune checkpoint inhibitors (pembrolizumab), PARP inhibitors (e.g., olaparib, rucaparib), PSMA binders], wherein the eligibility score for a particular treatment option and / or class of therapeutic agents indicates a prediction of whether the patient will benefit from a particular treatment and / or class of therapeutic agents.

[0037] In certain embodiments, the method includes (eg, automatically) generating a report [eg, an electronic file, eg, within a graphical user interface (eg, for user verification / sign-off)] including at least a portion of the values ​​of one or more patient indicators.

[0038] In certain embodiments, the method includes using one or more machine learning modules [e.g., one or more neural networks (e.g., one or more convolutional neural networks)] to perform one or more functions selected from the group consisting of: detecting a plurality of hotspots, wherein each of at least a portion of a plurality of 3D hotspot volumes corresponds to a specific detected hotspot and is generated by segmenting the specific detected hotspot; segmenting at least a portion of a plurality of 3D hotspot volumes; and classifying at least a portion of the 3D hotspot volumes (e.g., determining the likelihood that each 3D hotspot volume represents a potential cancerous lesion).

[0039] In some embodiments, the 3D functional image comprises a PET or SPECT image obtained after administration of an agent to a subject. In some embodiments, the agent comprises a PSMA binding agent. In some embodiments, the agent comprises 18 F. In certain embodiments, the agent comprises [18F]DCFPyL. In certain embodiments, the agent comprises PSMA-11. In certain embodiments, the agent comprises one or more selected from 99m Tc, 68 Ga, 177 Lu, 225 Ac, 111 In, 123 I. 124 I and 131 I is a member of a group composed of.

[0040] In another aspect, the present invention relates to a method for automatically analyzing a time series of medical images of an individual [e.g., three-dimensional images, e.g., nuclear medicine images (e.g., bone scan (scintigraphy), PET and / or SPECT), e.g., anatomical images (e.g., CT, X-ray, MRI), e.g., combined nuclear medicine and anatomical images (e.g., overlapped)], the method comprising: (a) receiving and / or accessing, by a processor of a computing device, the time series of medical images of the individual; and (b) identifying, by the processor, a plurality of hot spots within each of the medical images and determining, by the processor, one, two or all three of the following (i), (ii) and (iii): (i) a change in the number of identified lesions, (ii) a change in the overall volume of the identified lesions (e.g., each identified lesion is a portion of a volume of the identified lesions); ), and (iii) change in the total volume weighted by PSMA (e.g., lesion index) (e.g., the sum of the product of the lesion index and the lesion volume for all lesions in the region of interest) [e.g., wherein the change identified in step (b) is used to identify (1) disease status [e.g., progression, regression, or no change], (2) make treatment management decisions [e.g., activity monitoring, prostatectomy, anti-androgen therapy, prednisone, radiation, radiotherapy, radioPSMA therapy, or chemotherapy], or (3) treatment efficacy (e.g., where the individual has started treatment or has continued treatment with a drug or other therapy according to an initial set of images in the time series of medical images)] [e.g., wherein step (b) comprises use of a machine learning module / model].

[0041] In another aspect, the present invention relates to a method for analyzing a plurality of medical images of an individual (e.g., to assess the individual's disease status and / or progression), the method comprising: (a) receiving and / or accessing, by a processor of a computing device, a plurality of medical images of the individual, and obtaining, by the processor, a plurality of 3D hotspot maps; maps), each of which corresponds to a specific medical image (of the plurality of medical images) and identifies one or more hot spots within the specific medical image (e.g., representing possible underlying physical lesions within the individual); (b) for each specific image (medical image) of the plurality of medical images, determining, by a processor, a corresponding 3D anatomical segmentation map using a machine learning module [e.g., a deep learning network (e.g., a convolutional neural network (CNN))], which identifies a set of organ regions within the specific medical image [e.g., representing soft tissue and / or bone structures within the individual (e.g., one or more cervical vertebrae; thoracic vertebrae; lumbar vertebrae; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and mandible)], thereby generating a plurality of 3D anatomical segmentation maps; and (c) determining, by a processor, using (i) the plurality of 3D hot spots maps and (ii) the plurality of 3D anatomical segmentation maps, one or more lesion correspondences, each (lesion correspondence) identifying two organ regions within different medical images. one or more corresponding hot spots and determined (e.g., by a processor) to represent the same potential body lesion within an individual; and (d) determining, by a processor, based on the identification of the multiple 3D hot spot maps and the one or more lesion correspondences, one or more metrics {e.g., one or more hot spot quantification metrics and / or changes therein [e.g., quantifying the characteristics of individual hot spots and / or the potential body lesions they represent (e.g., over time / between multiple medical images), such as changes in volume, radiopharmaceutical uptake, shape, etc.]; such as patient indicators (e.g., measuring the overall disease burden and / or condition and / or risk of an individual) and / or changes therein; such as values ​​for classifying a patient (e.g., belonging to and / or suffering from a specific disease condition, progression, etc. category), such as prognostic metrics [e.g., indicating and / or quantifying the likelihood (e.g., overall survival) of one or more clinical outcomes (e.g., disease condition, progression, likely survival, treatment efficacy, etc.); such as predictive metrics (e.g., indicating predicted response to therapy and / or other clinical outcomes)}.

[0042] In some embodiments, the plurality of medical images include one or more anatomical images (eg, CT, X-ray, MRI, ultrasound, etc.).

[0043] In certain embodiments, the plurality of medical images include one or more nuclear medicine images [e.g., a bone scan (scintigraphy) (e.g., obtained after administration of a radiopharmaceutical such as 99mTc-MDP to a subject), PET (e.g., obtained after administration of a radiopharmaceutical such as [18F]DCFPyL, [68Ga]PSMA-11, [18F]PSMA-1007, rhPSMA-7.3(18F), [18F]-JK-PSMA-7 to a subject), or SPECT (e.g., obtained after administration of a radiopharmaceutical such as a 99mTc-labeled PSMA binder to a subject)].

[0044] In certain embodiments, the plurality of medical images comprises one or more composite images, each comprising an anatomical and nuclear medicine pair (e.g., overlapped / co-registered with one another; e.g., acquired at substantially the same time for an individual) (e.g., one or more PET / CT images).

[0045] In certain embodiments, the plurality of medical images is or comprises a time series of medical images, each medical image of the time series being associated with and having been acquired at a different specific time.

[0046] In certain embodiments, the time series of medical images includes a first medical image acquired before administration (e.g., one or more cycles) of a particular therapeutic agent [e.g., a PSMA binder (e.g., PSMA-617; e.g., PSMA I&T); e.g., a radiopharmaceutical; e.g., a PSMA binder labeled with a radionuclide (e.g., 177Lu-PSMA-617; e.g., 177Lu-PSMA I&T)] to a subject, and a second medical image acquired after administration (e.g., one or more cycles) of the particular therapeutic agent to the subject.

[0047] In certain embodiments, the method comprises classifying the individual as a responder and / or a non-responder to a particular therapeutic agent based on the values ​​of the one or more metrics determined in step (d).

[0048] In some embodiments, step (a) comprises generating each hotspot map by (e.g., automatically) segmenting at least a portion of the corresponding medical image (e.g., a sub-image thereof, such as a nuclear medicine image) (e.g., using a second hotspot segmentation, machine learning module [e.g., wherein the hotspot segmentation machine module comprises a deep learning network (e.g., a convolutional neural network (CNN))]).

[0049] In certain embodiments, each heat map includes, for each of at least a portion of the hotspots identified therein, one or more markers (eg, miTNM classification markers) identifying one or more assigned anatomical regions and / or lesion subtypes.

[0050] In some embodiments, the plurality of heat maps include (i) a first heat map corresponding to a first medical image (e.g., and identifying a first set of one or more hot spots therein), and (ii) a second heat map corresponding to a second medical image (e.g., and identifying a second set of one or more hot spots therein); the plurality of 3D anatomical segmentation maps include (i) a first 3D anatomical segmentation map identifying a set of organ regions within the first medical image, and (ii) a second 3D anatomical segmentation map identifying a set of organ regions within the second medical image; and step (c) includes using the first 3D anatomical segmentation map and the second 3D anatomical segmentation map to register (i) the first heat map with (ii) the second heat map (e.g., using the set of organ regions and / or one or more subsets thereof as landmarks within the first and second 3D anatomical segmentation maps to determine one or more registration fields (e.g., full 3D registration fields; e.g., pointwise registration and co-registering the first and second heat maps using one or more determined registration fields).

[0051] In certain embodiments, step (c) comprises: determining, for a group of two or more hotspots, each hotspot being a member of a different hotspot map and identified in different medical images, values ​​of one or more lesion correspondence metrics (e.g., volumetric overlap; e.g., centroid distance; e.g., lesion type match); and determining, based on the values ​​of the one or more lesion correspondence metrics, the two or more hotspots of the group to represent the same specific underlying body lesion, thereby including the two or more hotspots of the group in one of the one or more lesion correspondences.

[0052] In certain embodiments, step (d) comprises determining one, two, or all three of (i), (ii), and (iii): (i) a change in the number of identified lesions, (ii) a change in the overall volume of the identified lesions (e.g., a change in the sum of the volumes of the individual identified lesions), and (iii) a change in the total volume weighted by PSMA (e.g., a lesion index) (e.g., the sum of the product of the lesion index and the lesion volume for all lesions in the region of interest) [e.g., where the change identified in step (b) is used to identify (1) disease status [e.g., progression, regression, or no change], (2) make treatment management decisions [e.g., activity monitoring, prostatectomy, anti-androgen therapy, prednisone, radiation, radiotherapy, radioPSMA therapy, or chemotherapy], or (3) treatment efficacy (e.g., where the individual has started treatment or has continued treatment with a drug or other therapy according to an initial set of images in the time series of medical images)].

[0053] In certain embodiments, the method comprises determining (e.g., based on the values ​​of one or more metrics; e.g., in step (d)) the value of one or more prognostic metrics that are indicative of disease status / progression and / or treatment [e.g., determining the expected overall survival (OS) of the individual (e.g., predicted number of months)].

[0054] In certain embodiments, the method comprises using the values ​​of one or more metrics (e.g., change in tumor volume, SUV mean, SUV maximum, PSMA score, number of new lesions, number of disappeared lesions, total number of follow-up lesions) as input to a prognostic model (e.g., a statistical model, such as regression; such as a classification model, whereby patients are assigned to a particular category based on comparison of one or more patient indicator values ​​to one or more thresholds; such as a machine learning model, in which the values ​​of one or more patient indicators are received as input), which produces as output an expected value and / or range (e.g., category) indicating possible values ​​for a particular patient outcome (e.g., time, such as in months, representing expected survival, time to progression, time to radiographic progression, etc.).

[0055] In certain embodiments, the method includes using the values ​​of one or more metrics (e.g., change in tumor volume, SUV mean, SUV maximum, PSMA score, number of new lesions, number of disappeared lesions, total number of follow-up lesions) as input to a response model (e.g., a statistical model, such as regression; such as a classification model, whereby patients are assigned to a particular category based on comparison of one or more patient indicator values ​​to one or more thresholds; such as a machine learning model, in which values ​​of one or more patient indicators are received as input), which produces as output a classification (e.g., a binary classification) indicating the patient's response to treatment.

[0056] In certain embodiments, the method comprises using the values ​​of one or more metrics (e.g., change in tumor volume, SUV mean, SUV maximum, PSMA score, number of new lesions, number of disappeared lesions, total number of follow-up lesions) as input to a predictive model (e.g., a statistical model, such as regression; such as a classification model, whereby a patient is assigned to a particular category based on a comparison of one or more patient indicator values ​​to one or more thresholds; such as a machine learning model, in which the values ​​of one or more patient indicators are received as input), which generates a prediction model for one or more treatment options (e.g., abiraterone, enzalutamide, apalutamide, darutamide, sipralex-T, Ra223, docetaxel, cabazitaxel, pembrolizumab, olaparib, sirolimus, leuprorelin, lipasone ... The eligibility score for each of the following treatment options and / or classes of therapeutic agents [e.g., androgen biosynthesis inhibitors (e.g., abiraterone), androgen receptor inhibitors (e.g., enzalutamide, apalutamide, darutamide), cellular immunotherapy (e.g., siplucamine-T), internal radiation therapy (Ra223), anti-neoplastic agents (e.g., docetaxel, cabazitaxel), immune checkpoint inhibitors (pembrolizumab), PARP inhibitors (e.g., olaparib, rucaparib), PSMA binders] is provided as an output, wherein the eligibility score for a particular treatment option and / or class of therapeutic agents indicates a prediction of whether a patient will benefit from a particular treatment and / or class of therapeutic agents.

[0057] In another aspect, the present invention relates to a method for analyzing multiple medical images of an individual, the method comprising: (a) obtaining (e.g., receiving and / or accessing, and / or generating) a first 3D heat map of the individual by a processor of a computing device; (b) obtaining (e.g., receiving and / or accessing, and / or generating) a first 3D anatomical segmentation map associated with the first 3D heat map by the processor; (c) obtaining (e.g., receiving and / or accessing, and / or generating) a second 3D heat map of the individual by the processor; (d) obtaining (e.g., receiving and / or accessing, and / or generating) a second 3D heat map of the individual by the processor; and (e.g., receiving and / or accessing, and / or generating) a second 3D heat map of the individual by the processor. (e) determining, by a processor, a registration field (e.g., a full 3D registration field; e.g., point-by-point registration) using / based on the first 3D anatomical segmented map and the second 3D anatomical segmented map; (f) registering, by a processor, the first 3D hotspot map with the second 3D hotspot map using the determined registration field, thereby generating a co-registered pair of 3D hotspot maps; (g) determining, by a processor, identifications of one or more corresponding lesions using the co-registered pair of 3D hotspot maps; and (h) storing and / or providing, by a processor, identifications of one or more corresponding lesions for display and / or further processing.

[0058] In another aspect, the present invention relates to a method for analyzing multiple medical images of an individual (e.g., to assess the individual's disease condition and / or progression), the method comprising: (a) receiving and / or accessing, by a processor of a computing device, multiple medical images of the individual; (b) for each specific image (medical image) of the multiple medical images, determining, by the processor, a corresponding 3D anatomical segmentation map that identifies a set of organ regions within the specific medical image [e.g., representing soft tissue and / or bone structures (e.g., one or more cervical vertebrae; thoracic vertebrae; ; lumbar spine; left and right hip bones, sacrum and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain and mandible)], thereby generating a plurality of 3D anatomical segmented maps; (c) by a processor, using the plurality of 3D anatomical segmented maps to determine one or more registration fields (e.g., full 3D registration fields; e.g., point-by-point registration) and apply the one or more registration fields to register the plurality of medical images, thereby generating a plurality of registered medical images; (d) for each particular image in the plurality of registered medical images, by the processor, determining a corresponding registered 3D hotspot map, which identifies a particular registered medical image within the particular registered medical image (e) determining, by a processor, using the plurality of 3D registered heat maps, one or more hot spots (e.g., representing possible underlying physical lesions within the individual), thereby generating a plurality of registered 3D heat maps; (e) determining, by a processor, using the plurality of 3D registered heat maps, one or more lesion correspondences, each (lesion correspondence) identifying two or more corresponding hot spots within different medical images and determined (e.g., by the processor) to represent the same potential physical lesion within the individual; and (f) determining, by the processor, based on the plurality of 3D heat maps and the identification of the one or more lesion correspondences, one or more metrics {e.g., one or more hot spot quantification metrics and / or changes therein [e.g., quantification of individual hot spots and / or the potential physical lesions they represent] characteristics of a body lesion (e.g., over time / between multiple medical images), such as changes in volume, radiopharmaceutical uptake, shape, etc.]; such as patient indicators (e.g., measuring an individual's overall disease burden and / or condition and / or risk) and / or changes thereto; such as values ​​for classifying a patient (e.g., belonging to and / or suffering from a particular disease condition, progression, etc. category), such as prognostic metrics [e.g., indicating and / or quantifying the likelihood (e.g., overall survival) of one or more clinical outcomes (e.g., disease condition, progression, likely survival, therapeutic efficacy, etc.); such as predictive metrics (e.g., indicating predicted response to therapy and / or other clinical outcomes)}.

[0059] In another aspect, the present invention relates to a method for analyzing multiple medical images of an individual, the method comprising: (a) obtaining (e.g., receiving and / or accessing, and / or generating) a first 3D anatomical image (e.g., CT, X-ray, MRI, etc.) and a first 3D functional image [e.g., a nuclear medicine image (e.g., PET, SPECT, etc.)] of the individual by a processor of a computing device; (b) obtaining (e.g., receiving and / or accessing, and / or generating) a second 3D anatomical image and a second 3D functional image of the individual by the processor; (c) obtaining (e.g., receiving and / or accessing, and / or generating) a first 3D anatomical segmentation map by the processor based on (e.g., using) the first 3D anatomical image; (d) obtaining (e.g., receiving and / or accessing, and / or generating) a second 3D anatomical segmentation map by the processor based on (e.g., using) the second 3D anatomical image; (e.g., receiving and / or accessing, and / or generating) a second 3D anatomical segmentation map by the processor based on (e.g., using) the second 3D anatomical image; (f) by the processor, registering (aligning) a second 3D functional image with the first 3D functional image using the registration field, thereby producing a registered version of the second 3D functional image; (g) by the processor, obtaining a first 3D hotspot map associated with the first functional image; (h) by the processor, determining a second 3D hotspot map using the registered version of the second 3D functional image, thereby registering the second 3D hotspot map with the first 3D hotspot map; (i) by the processor, determining one or more lesion correspondences using the first 3D hotspot map and the second 3D hotspot map registered therewith; and (j) by the processor, storing and / or providing one or more lesion correspondences for display and / or further processing.

[0060] In another aspect, the present invention relates to a method for evaluating the efficacy of an intervention, the method comprising: (a) for each specific individual in a test population (e.g., comprising multiple individuals, such as those enrolled in a clinical trial) who presents a specific disease (e.g., prostate cancer (e.g., metastatic castration resistant prostate cancer)) and / or is at risk of a specific disease, performing a method as described in any of the aforementioned technical solutions to obtain multiple medical images of a specific patient, wherein the multiple medical images of the specific patient include a time series of medical images obtained over a time period spanning (e.g., before, during and / or after) the tested intervention, and one or more risk indicators include one or more endpoints indicating patient response to the tested intervention, thereby determining multiple values ​​for each of the one or more endpoints in the test population; and (b) determining the efficacy of the tested intervention based on the values ​​of the one or more endpoints in the test population.

[0061] In another aspect, the present invention relates to a method for treating an individual having a particular disease (e.g., prostate cancer (e.g., metastatic castration-resistant prostate cancer)) and / or at risk of a particular disease, the method comprising: a first cycle of administering a therapeutic agent to the individual; and a second cycle of administering a therapeutic agent to the individual based on the individual having been imaged (e.g., before and / or during and / or after the first cycle of the therapeutic agent) and identified as a responder to the therapeutic agent using, for example, the methods described in the above paragraphs (e.g., paragraphs

[0011] -

[0060] ), in any of the aspects and embodiments described herein (e.g., based on the values ​​of one or more risk indicators determined using, for example, the methods described in the above paragraphs (e.g., paragraphs

[0011] -

[0060] ), in any of the aspects and embodiments described herein, the individual has been identified / classified as a responder).

[0062] In another aspect, the present invention relates to a method for treating an individual having a particular disease (e.g., prostate cancer (e.g., metastatic castration-resistant prostate cancer)) and / or at risk of a particular disease, the method comprising: administering cycles of a first therapeutic agent to the individual; and administering cycles of a second therapeutic agent to the individual based on the individual having been imaged (e.g., before and / or during and / or after the cycles of the first therapeutic agent) and identified as a non-responder to the first therapeutic agent using any of the aspects and embodiments described herein, e.g., the methods described in the above paragraphs (e.g., paragraphs

[0011] -

[0060] ) (e.g., the individual has been identified / classified as a non-responder based on the values ​​of one or more risk indicators determined using any of the aspects and embodiments described herein, e.g., the methods described in the above paragraphs (e.g., paragraphs

[0011] -

[0060] )) (e.g., thereby subjecting the individual to potentially more effective therapy).

[0063] In another aspect, the present invention relates to a method for treating an individual having a particular disease (e.g., prostate cancer (e.g., metastatic castration-resistant prostate cancer)) and / or at risk of a particular disease, the method comprising: administering a cycle of a therapeutic agent to the individual; and interrupting the administration of the therapeutic agent to the individual based on the individual having been imaged (e.g., before and / or during and / or after the first cycle of the therapeutic agent) and identified as a non-responder to the therapeutic agent using any of the aspects and embodiments described herein, e.g., the methods described in the above paragraphs (e.g., paragraphs

[0011] -

[0060] ) (e.g., based on the values ​​of one or more risk indicators determined using any of the aspects and embodiments described herein, e.g., the methods described in the above paragraphs (e.g., paragraphs

[0011] -

[0060] ), the individual has been identified / classified as a non-responder) (e.g., thereby allowing the individual to receive a potentially more effective therapy).

[0064] In another aspect, the present invention relates to a method for automatically or semi-automatically whole-body assessing an individual with metastatic prostate cancer [e.g., metastatic castration-resistant prostate cancer (mCRPC) or metastatic hormone-sensitive prostate cancer (mHSPC)] to assess disease progression and / or treatment efficacy, the method comprising: (a) receiving, by a processor of a computing device, a first positron emission tomography (PET) image of the individual that targets prostate-specific membrane antigen (PSMA) (a first PSMA-PET image) and a first 3D anatomical image of the individual [e.g., a computed tomography (CT) image; e.g., a magnetic resonance image (MRI)], wherein the first 3D anatomical image of the individual is acquired simultaneously with, immediately after, or immediately before (e.g., on the same date as) the first PSMA PET image, such that the first 3D anatomical image and the first PSMA PET image correspond to a first date, and wherein the images depict a sufficiently large area of ​​the individual's body to cover areas of the body to which metastatic prostate cancer has spread (e.g., a full torso image or a whole-body image covering multiple organs) {e.g., wherein the PSMA-PET image is acquired using F-18piflufolastat PSMA (i.e., 2-(3-{1-carboxy-5-[(6-[18F]fluoro-pyridine-3-carbonyl)amino]-pentyl}ureido)-glutaric acid, also known as [18F]F-DCFPyL) or Ga-68PSMA-11 or other radiolabeled prostate specific membrane antigen inhibitor imaging agent}; (b) receiving, by a processor, a second PSMA-PET image of the individual and a second 3D anatomical image of the individual, both of which were obtained on a second date after the first date; and (c) using, by the processor, the first and second PSMA-PET images of the individual and a second 3D anatomical image of the individual. automatically determining a registration field (e.g., a full 3D registration field; e.g., point-by-point registration) based on automatically identified landmarks within the second 3D anatomical image (e.g., the identified region represents one or more of the cervical spine; the thoracic spine; the lumbar spine; the left and right hip bones, the sacrum, and the coccyx; the left ribs and the left scapula; the right ribs and the right scapula; the left femur; the right femur; the skull, the brain, and the mandible), and aligning the first and second PSMA-PET images using the determined registration field, by the processor [e.g., after segmenting the CT and / or PSMA-PET images to identify the regions of interest] and (d) using the first and second PSMA-PET images thus aligned to automatically detect (e.g., stage and / or quantify) a change in disease from the first date to the second date (e.g., progression or remission) [e.g., automatically identifying and / or identifying as such (e.g., tagging, labelling) one or both of the following (i) and (ii): (i) a change in the number of lesions {e.g., one or more new lesions (e.g., organ-specific lesions) or elimination of one or more lesions (e.g., organ-specific)}, and (ii) a change in tumor size {e.g., an increase in tumor size (PSMA-VOL increase / decrease), e.g., total tumor size, or a decrease in tumor size (PSMA-VOL decrease)} {e.g., a change in the volume of each of one or more specific lesions, or a change in the overall volume of a specific type of lesion (e.g., organ-specific tumor), or a change in the total volume of identified lesions}].

[0065] In certain embodiments, the method comprises one or more members selected from the group consisting of: lesion location assignment, tumor staging, nodal staging, distant cancer metastasis staging, assessment of intraprostatic lesions, and determination of PSMA expression score.

[0066] In certain embodiments, the subject has been administered therapy {e.g., hormone therapy, chemotherapy, and / or radiation therapy, e.g., androgen ablation therapy, e.g., Lu-containing compound, e.g., Lu-PSMA radioligand therapy, e.g., Lu-PSMA-617, e.g., Lu 177 vipivotide tetraxetan (Pluvicto), e.g., cabazitaxel} for treatment of metastatic prostate cancer one or more times from a first date to a second date (after obtaining the first image and before obtaining the second image), such that the method is used to assess the efficacy of the treatment.

[0067] In certain embodiments, the method further comprises obtaining one or more additional PSMA PET images and the 3D anatomical image of the subject after the second date, aligning the other PSMA PET images using the corresponding 3D anatomical images, and using the aligned other PSMA PET images to assess disease progression and / or treatment efficacy.

[0068] In certain embodiments, the method further comprises determining and presenting, by the processor, a predicted PSMA-PET image depicting predicted progression (or remission) of the disease until a future date (e.g., a future date that is later than the second date or any other subsequent date on which the PSMA-PET image was obtained) based at least in part on the detected change in disease from the first date to the second date.

[0069] In another aspect, the present invention relates to a method for quantifying and reporting disease (e.g., tumor) burden in a patient having cancer and / or at risk of cancer, the method comprising: (a) obtaining, by a processor of a computing device, a medical image of the patient; (b) detecting, by the processor, one or more (e.g., multiple) hot spots within the medical image, each hot spot within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein stereo pixels of the 3D hot spot volume have elevated intensities relative to their environment (e.g., and / or are otherwise indicative of or increased radiopharmaceutical absorption)] and representing possible underlying body lesions within the individual; and (c) for each particular lesion category of a plurality of lesion categories representing particular tissue regions and / or lesion subtypes: detecting, by processing (d) presenting, by the processor, a graphical representation of the patient indicator values ​​calculated for each of the plurality of lesion categories (e.g., a summary table listing each lesion category and the patient indicator values ​​calculated for each lesion category), thereby providing a user with a graphical report summarizing the tumor burden within the particular tissue region and / or associated with the particular lesion subtype.

[0070] In certain embodiments, the plurality of lesion categories include one or more of the following: (i) a local tumor category (e.g., a "T" or "miT" category) that identifies potential lesions and / or portions thereof located within one or more local tumor-related tissue regions associated with and / or adjacent to a local (e.g., primary) tumor site within a patient and represented by a corresponding subset of hotspots [e.g., where the cancer is prostate cancer and the one or more local tumor-related tissue regions include the prostate and optionally one or more adjacent structures (e.g., seminal vesicles, external sphincter, rectum, bladder, levator muscles, and / or pelvic wall); for example, where the cancer is breast cancer and the one or more local tumor-related tissue regions include prostate and / or adjacent structures (e.g., seminal vesicles, external sphincter, rectum, bladder, levator muscles, and / or pelvic wall); for example, where the cancer is breast cancer and the one or more local tumor-related tissue regions include prostate and / or adjacent structures (e.g., the one or more local tumor-associated tissue regions include breast; e.g., where the cancer is colorectal cancer and the one or more local tumor-associated tissue regions include colon; e.g., where the cancer is lung cancer and the one or more local tumor-associated tissue regions include lung]; (ii) a regional nodal category (e.g., an "N" or "miN" category) that identifies potential lesions located in local lymph nodes adjacent to and / or proximal to the original (e.g., primary) tumor site and represented by a corresponding subset of hotspots [e.g., where the cancer is prostate cancer and the regional lymph node category identifies hotspots representing lesions located in one or more pelvic lymph nodes (e.g., internal iliac, external iliac, obturator, presacral and (iii) one or more (e.g., distant) metastatic tumor categories (e.g., one or more "M" or "miM" categories) that identify potential cancer metastases (e.g., lesions that have spread beyond the original (e.g., primary) tumor site) and / or subtypes thereof and are represented by corresponding hotspot subsets [e.g., where the cancer is prostate cancer and the one or more metastatic tumor categories identify hotspots representing potential metastatic lesions located outside the patient's pelvic region (e.g., as defined by the pelvic brim, e.g., according to the American Joint Committee on Cancer staging manual)].

[0071] In certain embodiments, the one or more metastatic tumor categories include one or more of the following: a distant lymph node cancer metastasis category (e.g., a "Ma" or "miMa" category) that identifies potential lesions that have metastasized to distant lymph nodes and is represented by a corresponding subset of hotspots [e.g., where the cancer is prostate cancer and the distant lymph node region category identifies hotspots representing lymph nodes located outside the pelvis (e.g., outside the pelvic region) (e.g., common iliac iliac), retroperitoneal, supradiaphragmatic, inguinal, and other extrapelvic lymph nodes]; a distal bone metastasis category (e.g., an "Mb" or "miMb" category), which identifies potential lesions located in one or more bones (e.g., distal bones) of the patient and is represented by a corresponding subset of hotspots; and a visceral (also called distal soft tissue) metastasis category (e.g., an "Mc" or "miMc" category), which identifies potential lesions located in one or more organs or other non-lymphatic soft tissue regions outside of the local tumor-associated tissue region and is represented by a corresponding subset of hotspots (e.g., where the cancer is prostate cancer and the visceral metastasis category identifies hotspots representing potential lesions located in organs outside of the pelvis of the patient, such as the brain, lungs, liver, spleen, and kidneys).

[0072] In some embodiments, step (c) comprises determining, for each particular lesion category, a value for one or more of the following patient indicators: a lesion count that quantifies the number of (e.g., different) lesions represented by a hotspot subset corresponding to the particular lesion category (e.g., calculated as the number of hotspots within the corresponding subset); a maximum absorption value that quantifies the maximum absorption within the corresponding hotspot set (e.g., calculated as the maximum individual voxel intensity of all voxels within the hotspot volume of the corresponding subset; e.g., according to equation (13a)); an average absorption value that quantifies the overall average absorption within the corresponding hotspot subset (e.g., calculated as the overall average intensity of all voxels within the (total combined) hotspot volume of the corresponding subset; e.g., according to equation (13b)); a total volume lesion count that quantifies the number of (e.g., different) lesions represented by a hotspot subset corresponding to the particular lesion category (e.g., calculated as the number of hotspots within the corresponding subset); a maximum absorption value that quantifies the maximum absorption within the corresponding hotspot set (e.g., calculated as the maximum individual voxel intensity of all voxels within the hotspot volume of the corresponding subset; e.g., according to equation (13a)); a mean absorption value that quantifies the overall average absorption within the corresponding hotspot subset (e.g., calculated as the overall average intensity of all voxels within the (total combined) hotspot volume of the corresponding subset; e.g., according to equation (13b)); a variable volume that quantifies the total volume of lesions belonging to a particular lesion class (e.g., calculated as the sum of the volumes of all individual lesions (e.g., hotspots) of the corresponding subset; according to equation (13c)); and an intensity-weighted tumor volume (ILTV) score (e.g., an aPSMA score) calculated as the weighted sum of the volumes of all individual lesions weighted (e.g., multiplied) by a measure of their intensity [e.g., where the measure of its intensity is a lesion index that quantifies the intensity of the hotspot on a standardized scale based on comparison to one or more reference intensities indicative of physiological (e.g., normal, non-cancer-related) radiopharmaceutical absorption in one or more corresponding reference tissue regions, such as portions of the aorta and the liver] [e.g., calculated according to equation (13d)].

[0073] In certain embodiments, the method includes determining, for each of the lesion categories, an alpha-numeric code that classifies the overall burden within the particular lesion category (e.g., a miTNM staging code indicating (i) the particular lesion category and (ii) one or more numbers and / or numbers indicating a particular number, size, spatial extent, spatial pattern, and / or sublocation of hotspots of a corresponding subset, and the underlying body lesions they represent), and optionally in step (e), causing a representation of the alpha-numeric code for each particular lesion category to be generated and / or displayed.

[0074] In certain embodiments, the method further includes determining an overall disease stage (e.g., an alphanumeric code) for the patient based on multiple lesion categories and their corresponding hotspot subsets, which is indicative of the patient's overall disease state and / or burden, and presenting, by a processor, a graphical representation of the overall disease stage (e.g., an alphanumeric code) for inclusion in a report.

[0075] In certain embodiments, the method further includes: determining, by the processor, one or more reference intensity values, each indicating physiological (e.g., normal, non-cancer-related) absorption of the radiopharmaceutical within a particular reference tissue region (e.g., a portion of the aorta; e.g., the liver) within the patient and calculated based on the intensity of image voxels within a corresponding reference volume identified within the medical image; and in step (d), presenting, by the processor, a representation (e.g., a graph) of the one or more reference intensity values ​​for inclusion in the report.

[0076] In another aspect, the present invention relates to a method for characterizing and reporting detected individual lesions based on imaging assessment of patients having cancer and / or at risk of cancer, the method comprising: (a) obtaining a medical image of the patient by a processor of a computing device; (b) detecting by the processor a set of one or more (e.g., multiple) hot spots within the medical image, each hot spot of the set within the medical image corresponding to (e.g., being or containing) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein the stereo pixels of the 3D hot spot volume have elevated intensity relative to their environment (e.g., and / or are otherwise indicative of or increased radiopharmaceutical absorption)] and representing a possible underlying physical lesion within the individual; and (c) assigning by the processor one or more lesion class labels to each of the one or more hot spots of the set, each lesion class label class representing a particular (d) calculating, by a processor, for each specific metric in the one or more individual hotspot quantitative metrics, a value of a specific individual hotspot quantitative metric for each hotspot of the set; and (e) displaying, by a processor, a graphical representation comprising, for each specific hotspot of at least a portion of the hotspot set, an identification of the specific hotspot (e.g., a row in a table, and optionally an alphanumeric identification, such as a number identifying the specific hotspot), one or more lesion category labels assigned to the specific hotspot and the values ​​of one or more individual hotspot quantitative metrics calculated for the specific hotspot [e.g., an overview table (e.g., a scrollable overview table) listing each hotspot as a row and listing the assigned lesion category and hotspot quantitative metrics column-wise].

[0077] In certain embodiments, the lesion category label includes a label representing one or more of the following: (i) a local tumor category (e.g., a "T" or "miT" category) that identifies potential lesions and / or portions thereof located within one or more local tumor-related tissue regions associated with and / or adjacent to a local (e.g., primary) tumor site within a patient and represented by a corresponding subset of hotspots [e.g., where the cancer is prostate cancer and the one or more local tumor-related tissue regions include the prostate and optionally one or more adjacent structures (e.g., seminal vesicles, external sphincter, rectum, bladder, levator muscles, and / or pelvic wall); e.g., where the cancer is breast cancer and the one or more local tumor-related tissue regions include the breast; e.g., where the cancer is colorectal cancer and the one or more local tumor-related tissue regions include the colon; e.g., where the cancer is lung cancer and the one or more local tumor-related tissue regions include the lung]; (ii) a regional nodule category (e.g., "N" or "N" category) that identifies potential lesions and / or portions thereof located within one or more local tumor-related tissue regions associated with and / or adjacent to a local (e.g., primary) tumor site within a patient and represented by a corresponding subset of hotspots [e.g., where the cancer is prostate cancer and the one or more local tumor-related tissue regions include the prostate and optionally one or more adjacent structures (e.g., seminal vesicles, external sphincter, rectum, bladder, levator muscles, and / or pelvic wall); e.g., where the cancer is breast cancer and the one or more local tumor-related tissue regions include the breast; e.g., where the cancer is colorectal cancer and the one or more local tumor-related tissue regions include the colon; e.g., where the cancer is lung cancer and the one or more local tumor-related tissue regions include the lung]; (e.g., "mM" or "miM" categories) that identify potential lesions located in local lymph nodes adjacent to and / or proximal to the original (e.g., primary) tumor site and are represented by a corresponding subset of hotspots [e.g., where the cancer is prostate cancer and the regional lymph node category identifies hotspots representing lesions located in one or more pelvic lymph nodes (e.g., internal iliac, external iliac, obturator, presacral tubercle, or other pelvic lymph nodes)]; and (iii) one or more (e.g., distant) metastatic tumor categories (e.g., one or more "M" or "miM" categories) that identify potential cancer metastases (e.g., lesions that have spread beyond the original (e.g., primary) tumor site) and / or subtypes thereof and are represented by a corresponding subset of hotspots [e.g., where the cancer is prostate cancer and the one or more metastatic tumor categories identify hotspots representing potential metastatic lesions located outside the patient's pelvic region (e.g., as defined by the pelvic brim, e.g., according to the American Joint Committee on Cancer Staging Manual)].

[0078] In certain embodiments, the one or more metastatic tumor categories include one or more of the following: a distal lymph node cancer metastasis category (e.g., a "Ma" or "miMa" category) that identifies potential lesions that have metastasized to the distal lymph nodes and are represented by a corresponding subset of hotspots [e.g., where the cancer is prostate cancer and the distal lymph node region category identifies hotspots representing lesions located in lymph nodes outside the pelvis (e.g., outside the pelvic region) (e.g., common iliac, retroperitoneal, supra-diaphragmatic, inguinal, and other extra-pelvic lymph nodes)]; a distal bone cancer metastasis category (e.g., "Mb" or "miMb ... ” category), which identifies potential lesions located in one or more bones (e.g., distal bones) of the patient and are represented by a corresponding subset of hotspots; and a visceral (also known as distal soft tissue) cancer metastasis category (e.g., the “Mc” or “miMc” category), which identifies potential lesions located in one or more organs or other non-lymphoid soft tissue areas outside the local tumor-associated tissue area and are represented by a corresponding subset of hotspots (e.g., where the cancer is prostate cancer and the visceral cancer metastasis category identifies hotspots representing potential lesions located in the patient's extra-pelvic organs, such as the brain, lungs, liver, spleen, and kidneys).

[0079] In some embodiments, the lesion category label includes one or more tissue markers that identify a particular organ or bone, wherein the lesion (represented by a hotspot) is determined (e.g., based on comparison of the hotspot to an anatomical segmentation map) to be located in the particular organ or bone (e.g., one or more of the organ or bone regions listed in Table 1).

[0080] In some embodiments, one or more individual hotspot quantification metrics include one or more of the following: maximum intensity (e.g., SUV maximum) (e.g., determined according to any of equations (1a), (1b) or (1c)), peak intensity (e.g., SUV peak) (e.g., determined according to any of equations (3a), (3b) or (3c)), average intensity (e.g., SUV average) (e.g., determined according to any of equations (2a), (2b), (2c)), lesion volume (e.g., determined according to any of equations (5a) or (5b)), and lesion index (e.g., measuring the intensity of the hotspot on a standardized scale) (e.g., determined according to equation (4)).

[0081] In another aspect, the present invention relates to a method for quantifying and reporting disease (e.g., tumor) progression and / or risk over time for a patient having cancer and / or at risk of cancer, the method comprising: (a) obtaining, by a processor of a computing device, a plurality of medical images of the patient, each medical image representing a scan of the patient obtained at a particular time (e.g., a longitudinal dataset); (b) for each particular image of the plurality of medical images, detecting, by the processor, a corresponding set of one or more (e.g., a plurality of) hot spots within the particular medical image, each hot spot within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein the stereo pixels of the 3D hot spot volume have elevated intensities relative to their environment (e.g., and / or are otherwise indicative of or increased radiopharmaceutical absorption )] and represents possible underlying physical lesions within an individual; (c) for each specific indicator of one or more (e.g., overall) patient indicators that measure (e.g., quantify) the overall disease (e.g., tumor) load within a patient at a specific time, determining, by a processor, a value of the specific (e.g., overall) patient indicator for each specific medical image of a plurality of medical images based on a corresponding set of hot spots detected for a specific medical image, thereby determining a value set for each specific indicator of the one or more patient indicators, the value set tracking changes in disease load over time by measuring the specific patient indicator value; and (d) displaying, by a processor, a graphical representation of the value set of at least a portion (e.g., a specific one, a specific subset) of the one or more patient indicator values, thereby conveying to the patient a measurement of disease progression over time.

[0082] In some embodiments, the one or more patient indicators include: a lesion count that quantifies the number of (e.g., different) lesions represented by a set of hotspots corresponding to a particular medical image and detected within the particular medical image (e.g., at a particular time point) (e.g., calculated as the number of hotspots within the corresponding hotspot set); a maximum absorption value that quantifies the maximum absorption within the corresponding hotspot set of the particular medical image (e.g., calculated as the maximum individual voxel intensity within all voxels within the hotspot volume of the corresponding hotspot set of the particular medical image; e.g., according to equation (7a) or (7b)); an average absorption value that quantifies the overall average absorption within the corresponding hotspot set (e.g., calculated as the overall average intensity of all voxels within the (total combined) hotspot volume of the corresponding set, e.g., according to equation (10a) or (10b)). (10b)); a total volume lesion volume, which quantifies the total volume of lesions detected within an individual at a particular time point (e.g., calculated as the sum of all individual hotspot volumes of a corresponding set of hotspots detected within a particular medical image); and an intensity-weighted tumor volume (ILTV) score (e.g., an aPSMA score), calculated as a weighted sum of all individual lesion volumes, each lesion volume weighted (e.g., multiplied) by a measure of its intensity [e.g., where the measure of hotspot intensity is a lesion index that quantifies the hotspot intensity on a standardized scale based on comparison to one or more reference intensities indicative of physiological (e.g., normal, non-cancer-related) radiopharmaceutical absorption within one or more corresponding reference tissue regions, such as portions of the aorta and the liver] [e.g., calculated according to equation (12)].

[0083] In certain embodiments, the method further includes determining, for each particular medical image of the plurality of medical images, an overall disease stage (e.g., an alphanumeric code) based on a corresponding set of hotspots and indicating the overall disease state and / or burden of the patient at a particular point in time, and presenting, by a processor, a graphical representation of the overall disease stage (e.g., an alphanumeric code) at each point in time.

[0084] In certain embodiments, the method further includes: determining, by the processor, for each of a plurality of medical images, a set of one or more reference intensity values, each of which indicates physiological (e.g., normal, non-cancer-related) absorption of the radiopharmaceutical within a particular reference tissue region (e.g., a portion of the aorta; e.g., the liver) within the patient and calculated based on intensities of image voxels within a corresponding reference volume identified within the medical image; and presenting, by the processor, a representation (e.g., a table; e.g., traces in a diagram) of the one or more reference intensity values.

[0085] In another aspect, the present invention relates to a method for automatically processing a 3D image of an individual to determine the value of one or more patient indicators measuring the individual's (e.g., overall) disease burden and / or risk, the method comprising: (a) receiving, by a processor of a computing device, a 3D functional image of the individual obtained using a functional imaging modality; (b) segmenting, by the processor, a plurality of 3D hotspot volumes within the 3D functional image, each 3D hotspot volume corresponding to a local region having an elevated intensity relative to its surroundings and representing a potential cancerous lesion within the individual, thereby obtaining a set of 3D hotspot volumes; and (c) calculating, by the processor, for each specific metric in one or more individual hotspot quantification metrics, a specific individual hotspot quantity for each 3D hotspot volume of the set. (d) determining, by a processor, values ​​of one or more patient indicators, wherein each of at least a portion of the patient indicators is associated with one or more specific individual hotspot quantification metrics and is calculated using (e.g., the same) specific function of the intensity and / or number of stereo pixels within a combined hotspot volume, wherein the combined hotspot volume includes at least a portion (e.g., substantially all; e.g., a specific subset) of a set of 3D hotspot volumes (e.g., forming a union thereof).

[0086] In certain embodiments, the patient-specific metric is an ensemble average voxel intensity and is calculated as the ensemble average of the voxel intensities within the combined hotspot volume.

[0087] In another aspect, the present invention relates to a method for automatically determining the prognosis of an individual with prostate cancer from one or more medical images of the individual [e.g., one or more PSMA PET images (PET images obtained after administration of a PSMA targeting compound to the individual) and / or one or more anatomical (e.g., CT) images], the method comprising: (a) receiving and / or accessing, by a processor of a computing device, one or more images of the individual; (b) automatically determining, by the processor, from the one or more images, a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) molecular imaging TNM (miTNM) lesion type classifications (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) of local (T), pelvic node (N), and / or extrapelvic (M) disease; (ii) indicating lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) standard physiological uptake values ​​(e.g., standard physiological uptake values) of the individual [e.g., standard physiological uptake values ​​of the individual]

[00136] The present invention relates to a method for automatically determining a prognosis for an individual based on the quantitative assessment in (b), wherein the prognosis comprises one or more of the following for the individual: (I) expected survival (e.g., number of months), (II) expected time to disease progression, (III) expected time to radiographic progression, (IV) risk of simultaneous (synchronous) cancer metastasis, and (V) risk of future (metachronous) cancer metastasis.

[0088] In certain embodiments, the quantitative assessment of the one or more prostate cancer lesions determined in step (b) comprises one or more of: (A) total tumor volume, (B) change in tumor volume, (C) total SUV, and (D) PSMA score, and wherein the prognosis of the individual determined in step (c) comprises one or more of: (E) expected survival (e.g., months), (F) time to progression, and (G) time to radiographic progression.

[0089] In certain embodiments, the quantitative assessment of the one or more prostate cancer lesions determined in step (b) comprises one or more characteristics of PSMA expression in the prostate, and wherein the prognosis of the individual determined in step (c) comprises the risk of simultaneous (synchronous) cancer metastasis and / or the risk of future (metachronous) cancer metastasis.

[0090] In another aspect, the present invention relates to a method for automatically determining a response to treatment in an individual with prostate cancer from a plurality of medical images of the individual [e.g., one or more PSMA PET images (PET images obtained after administration of a targeted PSMA compound to the individual) and / or one or more anatomical (e.g., CT) images], the method comprising: (a) receiving and / or accessing, by a processor of a computing device, a plurality of images of the individual, wherein at least a first image of the plurality of images was obtained before administration of the treatment and at least a second image of the plurality of images was obtained after administration of the treatment (e.g., after a period of time); (b) automatically determining, by the processor, from the images a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) molecular imaging TNM (miTNM) lesion type classification (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) for local (T), pelvic node (N) and / or extrapelvic (M) disease; (ii) indication of lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) standard physiological absorption value (SUV) (e.g., SUV maximum, SUV peak, SUV average); (iv) total lesion volume; (v) change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein) (e.g., where the quantitative assessment includes response evaluation criteria in PSMA imaging (Response Evaluation Criteria in PSMA-imaging, RECIP) criteria and / or PSMA PET progression (PPP) criteria); and (c) automatically determining whether the individual responds to treatment (e.g., responder / non-responder) and / or the extent to which the individual responds to treatment (e.g., numerical or classification) from the quantitative assessment in (b).

[0091] In another aspect, the present invention relates to a method for automatically identifying whether an individual with prostate cancer (e.g., metastatic prostate cancer) is likely to benefit from a particular treatment for prostate cancer using a plurality of medical images of the individual [e.g., one or more PSMA PET images (PET images obtained after administration of a targeted PSMA compound to the individual) and / or one or more anatomical (e.g., CT) images], the method comprising: (a) receiving and / or accessing, by a processor of a computing device, a plurality of images of the individual; (b) automatically determining, by the processor, from the images a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) molecular imaging TNM (miTNM) lesion type classification (e.g., miT, miN, miMa (lymphatic)) of localized (T), pelvic node (N), and / or extrapelvic (M) disease] , miMb (bone), miMc (other)); (ii) an indication of lesion location (e.g., prostate, ilium, pelvic bone, rib cage, etc.); (iii) a standard physiological absorption value (SUV) (e.g., SUV maximum, SUV peak, SUV mean); (iv) total lesion volume; (v) change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) a calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein) (e.g., where the quantitative assessment comprises the Response Evaluation Criteria PSMA Imaging (RECIP) criteria and / or the PSMA PET Progression (PPP) criteria); and (c) automatically determining from the quantitative assessment in (b) whether an individual is likely to benefit from a specific treatment for prostate cancer [e.g., determining one or more specific treatments and / or a class of treatments for an individual, e.g., a specific radioligand therapy, e.g., leuproreceptor detrachetan eligibility score].

[0092] In another aspect, the present invention relates to a system for automatically processing a 3D image of an individual to determine the value of one or more patient indicators that measure the individual's (e.g., overall) disease burden and / or risk, the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) receive a 3D functional image of the individual obtained using a functional imaging modality; (b) segment a plurality of 3D hotspot volumes within the 3D functional image, each 3D hotspot volume corresponding to a local region having elevated intensity relative to its surroundings and representing a potential cancerous lesion within the individual, thereby obtaining a set of 3D hotspot volumes; (c) for each specific metric in one or more individual hotspot quantification metrics, calculate the value of a specific individual hotspot quantification metric for each 3D hotspot volume of the set; and (d) determine the value of one or more patient indicators, wherein each of at least a portion of the patient indicators is associated with one or more specific individual hotspot quantification metrics and is a function of at least a portion (e.g., substantially all; e.g., a specific subset) of the values ​​of the one or more specific individual hotspot quantification metrics calculated for the set of 3D hotspot volumes.

[0093] In some embodiments, the system has one or more features and / or instructions that cause the processor to perform one or more steps articulated herein (eg, in the above paragraphs, eg, in paragraphs

[0012] -

[0039] ).

[0094] In another aspect, the present invention relates to a system for automatically analyzing a time series of medical images of an individual [e.g., three-dimensional images, e.g., nuclear medicine images (e.g., bone scan (scintigraphy), PET and / or SPECT), e.g., anatomical images (e.g., CT, X-ray, MRI), e.g., combined nuclear medicine and anatomical images (e.g., overlap)], the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) receive and / or access a time series of medical images of the individual; and (b) identify a plurality of hot spots within each of the medical images, and determine, by the processor, one, two or all three of the following (i), (ii) and (iii): (i) a change in the number of identified lesions, (ii) a change in the number of identified lesions, ) a change in the overall volume of the identified lesions (e.g., a change in the sum of the volumes of the individual identified lesions), and (iii) a change in the total volume weighted by PSMA (e.g., a lesion index) (e.g., the sum of the product of the lesion index and the lesion volume for all lesions in the region of interest) [e.g., wherein the change identified in step (b) is used to identify (1) disease status [e.g., progression, regression, or no change], (2) make treatment management decisions [e.g., activity monitoring, prostatectomy, anti-androgen therapy, prednisone, radiation, radiotherapy, radioPSMA therapy, or chemotherapy], or (3) treatment efficacy (e.g., wherein the individual has started treatment or has continued treatment with a drug or other therapy according to an initial set of images in the time series of medical images)] [e.g., wherein step (b) comprises use of a machine learning module / model].

[0095] In another aspect, the present invention relates to a system for analyzing multiple medical images of an individual (e.g., to assess the individual's disease status and / or progression), the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) receive and / or access multiple medical images of the individual and obtain, by the processor, multiple 3D hotspot maps, each of which corresponds to a specific medical image (of the multiple) and identifies one or more hotspots within the specific medical image (e.g., representing possible underlying physical pathology within the individual); b) for each specific image (medical image) of the plurality of medical images, using a machine learning module [e.g., a deep learning network (e.g., a convolutional neural network (CNN))] to determine a corresponding 3D anatomical segmentation map, which identifies a set of organ regions within the specific medical image [e.g., representing soft tissue and / or bone structures within an individual (e.g., one or more cervical vertebrae; thoracic vertebrae; lumbar vertebrae; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and mandible)], thereby generating a plurality of 3D anatomical segmentation maps; ( c) using (i) a plurality of 3D heat maps and (ii) a plurality of 3D anatomical segmentation maps to determine one or more identifications of lesion correspondences, each (lesion correspondence) identifying two or more corresponding hot spots within different medical images and determined (e.g., by a processor) to represent the same underlying body lesion within an individual; and (d) determining one or more metrics {e.g., one or more hot spot quantification metrics and / or changes therein [e.g., quantifying individual hot spots and / or underlying body lesions represented by them (e.g., over time / across multiple medical images) based on the plurality of 3D heat maps and the identifications of the one or more lesion correspondences]; such as a characteristic of a patient (e.g., between images, such as changes in volume, radiopharmaceutical absorption, shape, etc.); such as a patient indicator (e.g., measuring an individual's overall disease burden and / or condition and / or risk) and / or changes therein; such as a value for classifying a patient (e.g., belonging to and / or suffering from a particular disease condition, progression, etc. category), such as a prognostic metric [e.g., indicating and / or quantifying the likelihood (e.g., overall survival) of one or more clinical outcomes (e.g., disease condition, progression, likely survival, therapeutic efficacy, etc.); such as a predictive metric (e.g., indicating a predicted response to therapy and / or other clinical outcomes)}.

[0096] In some embodiments, the system has one or more features and / or instructions that cause the processor to perform one or more steps expressed herein (eg, in the above paragraphs, eg, in paragraphs

[0042] -

[0056] ).

[0097] In another aspect, the present invention relates to a system for analyzing multiple medical images of an individual, the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) obtain (e.g., receive and / or access, and / or generate) a first 3D heat map of the individual; (b) obtain (e.g., receive and / or access, and / or generate) a first 3D anatomical segmentation map associated with the first 3D heat map; (c) obtain (e.g., receive and / or access, and / or generate) a second 3D heat map of the individual; (d) obtain (e.g., receiving and / or accessing, and / or generating) a second 3D anatomical segmented map associated with a second 3D heat map; (e) determining a registration field (e.g., 3D registration field and / or point-by-point registration) using / based on the first 3D anatomical segmented map and the second 3D anatomical segmented map; (f) using the registration field to register the first 3D heat map with the second 3D heat map, thereby generating a co-registered pair of 3D heat maps; (g) using the co-registered pair of 3D heat maps to determine one or more lesion correspondences; and (h) storing and / or providing one or more lesion correspondences for display and / or further processing.

[0098] In another aspect, the present invention relates to a system for analyzing multiple medical images of an individual (e.g., to assess the individual's disease condition and / or progression), the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) receive and / or access multiple medical images of the individual; (b) for each particular image (medical image) of the multiple medical images, determine, using a machine learning module [e.g., a deep learning network (e.g., a convolutional neural network (CNN))], a corresponding 3D anatomical segmentation map that identifies a set of organ regions within the particular medical image [ For example, representing soft tissue and / or bone structures within an individual (e.g., one or more cervical vertebrae; thoracic vertebrae; lumbar vertebrae; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and mandible)], thereby generating a plurality of 3D anatomical segmented maps; (c) using the plurality of 3D anatomical segmented maps to determine one or more registration fields (e.g., full 3D registration fields; e.g., point-by-point registration) and applying the one or more registration fields to register the plurality of medical images, thereby generating a plurality of registered medical images; (d) for each particular image in the plurality of registered medical images, determining a medical map that identifies the particular registration (e) determining one or more lesion correspondences using the plurality of 3D registered heat maps, each (lesion correspondence) identifying two or more corresponding hot spots in different medical images and determined (e.g., by a processor) to represent the same potential body lesion in the individual; and (e) determining one or more metrics {e.g., one or more hot spot quantification metrics and / or changes therein [e.g., quantification of individual hot spots and / or potential body lesions represented by them] based on the plurality of 3D heat maps and the identification of the one or more lesion correspondences. characteristics of body lesions (e.g., changes in volume, radiopharmaceutical uptake, shape, etc.)]; such as patient indicators (e.g., measuring an individual's overall disease burden and / or condition and / or risk) and / or changes therein; such as values ​​for classifying a patient (e.g., belonging to and / or suffering from a particular disease condition, progression, etc.), such as prognostic metrics [e.g., indicating and / or quantifying the likelihood (e.g., overall survival) of one or more clinical outcomes (e.g., disease condition, progression, likely survival, efficacy of treatment, etc.); such as values ​​for predictive metrics (e.g., indicating predicted response to therapy and / or other clinical outcomes)}.

[0099] In another aspect, the present invention relates to a system for analyzing multiple medical images of an individual, the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) obtain (e.g., receive and / or access, and / or generate) a first 3D anatomical image (e.g., CT, X-ray, MRI, etc.) and a first 3D functional image [e.g., a nuclear medicine image (e.g., PET, SPECT, etc.)] of the individual; (b) obtain (e.g., receive and / or access, and / or generate) a second 3D anatomical image and a second 3D functional image of the individual; (c) obtain (e.g., receive and / or access, and / or generate) a first 3D anatomical segmentation map based on (e.g., using) the first 3D anatomical image; and (d) obtain (e.g., receive and / or access, and / or generate) a first 3D anatomical segmentation map based on (e.g., using) the second 3D anatomical image. For example, receiving and / or accessing, and / or generating) a second 3D anatomical segmented map; (e) determining a registration field (e.g., 3D registration field and / or point-by-point registration) using / based on the first 3D anatomical segmented map and the second 3D anatomical segmented map; (f) using the registration field to register (align) the second 3D functional image with the first 3D functional image, thereby generating a registered version of the second 3D functional image; (g) obtaining a first 3D hotspot map associated with the first functional image; (h) using the registered version of the second 3D functional image to determine a second 3D hotspot map, the second 3D hotspot map thereby being registered with the first 3D hotspot map; (i) determining one or more lesion correspondences using the first 3D hotspot map and the second 3D hotspot map registered therewith; and (j) storing and / or providing one or more lesion correspondences for display and / or further processing.

[0100] In another aspect, the present invention relates to a system for automatically or semi-automatically whole-body assessing an individual with metastatic prostate cancer [e.g., metastatic castration-resistant prostate cancer (mCRPC) or metastatic hormone-sensitive prostate cancer (mHSPC)] to assess disease progression and / or treatment efficacy, the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receive a first positron emission tomography (PET) image of the individual that targets prostate-specific membrane antigen (PSMA) (a first PSMA-PET image) and a first 3D anatomical image of the individual [e.g., a computed tomography (CT) image; e.g., a magnetic resonance image (MRI)], wherein the first 3D anatomical image of the individual is acquired simultaneously with, immediately after, or immediately before (e.g., on the same date as) the first PSMA PET image, such that the first 3D anatomical image and the first PSMA PET image correspond to a first date, and wherein the images depict a sufficiently large area of ​​the individual's body to encompass areas of the body to which metastatic prostate cancer has spread (e.g., a full torso image or a full-body image encompassing multiple organs) {e.g., wherein the PSMA-PET image is acquired using F-18piflufolastat PSMA (i.e., 2-(3-{1-carboxy-5-[(6-[18F]fluoro-pyridine-3-carbonyl)amino]-pentyl}ureido)-glutaric acid, also known as [18F]F-DCFPyL), or Ga-68 (a) receiving a second PSMA-PET image of the individual and a second 3D anatomical image of the individual, both of which are obtained on a second date after the first date; (b) automatically determining a registration field (e.g., a full 3D registration field; e.g., point-by-point registration) using automatically identified landmarks within the first and second 3D anatomical images (e.g., the identified regions represent one or more of the cervical spine; thoracic spine; lumbar spine; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and mandible), and aligning the first and second PSMA-PET images using the determined registration field [e.g., before or after segmenting the CT and / or PSMA-PET images to identify boundaries of organs and / or bones, and after automatically performing a registration on the PSMA-PET images] and (d) using the first and second PSMA-PET images thus aligned to automatically detect (e.g., stage and / or quantify) a change in disease from the first date to the second date (e.g., progression or remission) [e.g., automatically identifying and / or identifying as such (e.g., marking, labeling) one or both of the following (i) and (ii): (i) a change in the number of lesions {e.g., one or more new lesions (e.g., organ-specific lesions), or elimination of one or more lesions (e.g., organ-specific)}, and (ii) a change in tumor size {e.g., an increase in tumor size (PSMA-VOL increase / decrease), e.g., total tumor size, or a decrease in tumor size (PSMA-VOL decrease)} {e.g., a change in the volume of each of one or more specific lesions, or a change in the overall volume of a specific type of lesion (e.g., organ-specific tumor), or a change in the total volume of identified lesions}.

[0101] In some embodiments, the system has one or more features and / or instructions that cause the processor to perform one or more steps expressed herein (eg, in the above paragraphs, eg, in paragraphs

[0065] -

[0068] ).

[0102] In another aspect, the present invention relates to a system for quantifying and reporting disease (e.g., tumor) burden in patients suffering from cancer and / or at risk of cancer, the method comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) obtaining a medical image of the patient; (b) detecting one or more (e.g., multiple) hot spots within the medical image, each hot spot within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein the stereo pixels of the 3D hot spot volume have elevated intensity relative to their environment (e.g., and / or are otherwise indicative of or increased radiopharmaceutical absorption)] and representing possible underlying body lesions within the individual; (c) for a particular tissue region and / or For each specific lesion category of multiple lesion categories of lesion subtypes: identifying a corresponding subset of one or more hot spots belonging to the specific lesion category (e.g., based on a determination by a processor that the hot spots represent potential body lesions located within a specific tissue region and / or belonging to a specific lesion subtype represented by the specific lesion category); and determining values ​​of one or more patient indicators that quantify the disease (e.g., tumor) burden within and / or associated with the specific lesion category based on the corresponding subset of hot spots; and (d) presenting a graphical representation of the patient indicator values ​​calculated for each of the multiple lesion categories (e.g., a summary table listing each lesion category and the patient indicator values ​​calculated for each lesion category), thereby providing a user with a graphical report summarizing the tumor burden within the specific tissue region and / or associated with the specific lesion subtype.

[0103] In some embodiments, the system has one or more features and / or instructions that cause the processor to perform one or more steps expressed herein (eg, in the above paragraphs, eg, in paragraphs

[0070] -

[0075] ).

[0104] In certain embodiments, the present invention relates to a system for characterizing and reporting detected individual lesions based on imaging assessments of patients having cancer and / or at risk of cancer, the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) obtain a medical image of the patient; (b) detect a set of one or more (e.g., multiple) hot spots within the medical image, each hot spot of the set within the medical image corresponding to (e.g., being or containing) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein the stereo pixels of the 3D hot spot volume have elevated intensity relative to their environment (e.g., and / or are otherwise indicative of or increased radiopharmaceutical absorption)] and represent possible underlying physical lesions within the individual; and (c) assign one or more lesion category labels to the set. (d) calculating, for each of the one or more hotspots in the set, a value of a specific individual hotspot quantitative metric for each hotspot in the set for each specific metric in the one or more individual hotspot quantitative metrics; and (e) displaying a graphical representation that includes, for each specific hotspot in at least a portion of the hotspots in the set, an identification of the specific hotspot (e.g., a row in a table, and optionally an alphanumeric identification, such as a number identifying the specific hotspot), as well as one or more lesion category labels assigned to the specific hotspot and the values ​​of one or more individual hotspot quantitative metrics calculated for the specific hotspot [e.g., an overview table (e.g., a scrollable overview table) listing each hotspot as a row and the assigned lesion category and hotspot quantitative metric listed as columns].

[0105] In some embodiments, the system has one or more features and / or instructions that cause the processor to perform one or more steps expressed herein (eg, in the above paragraphs, eg, in paragraphs

[0077] -

[0080] ).

[0106] In another aspect, the present invention relates to a system for quantifying and reporting disease (e.g., tumor) progression and / or risk over time for patients having cancer and / or at risk of cancer, the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) obtain a plurality of medical images of the patient, each medical image representing a scan of the patient obtained at a particular time (e.g., a longitudinal dataset); (b) for each particular image in the plurality of medical images, detect a corresponding set of one or more (e.g., a plurality of) hot spots within the particular medical image, each hot spot within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein the stereo pixels of the 3D hot spot volume have an elevated intensity relative to its environment (e.g., and and / or otherwise indicating or increasing radiopharmaceutical absorption)] and representing possible underlying physical pathology within the individual; (c) for each specific indicator of one or more (e.g., overall) patient indicators that measure (e.g., quantify) the overall disease (e.g., tumor) burden within the patient at a specific time, determining a value of the specific (e.g., overall) patient indicator for each specific medical image of a plurality of medical images based on a corresponding set of hot spots detected for a specific medical image, thereby determining a value set for each specific indicator of the one or more patient indicators, the value set tracking changes in disease burden over time by measuring the specific patient indicator value; and (d) displaying a graphical representation of the value set of at least a portion (e.g., a specific one, a specific subset) of the one or more patient indicator values, thereby conveying to the patient a measurement of disease progression over time.

[0107] In some embodiments, the system has one or more features and / or instructions that cause the processor to perform one or more steps expressed herein (eg, in the above paragraphs, eg, in paragraphs

[0082] -

[0084] ).

[0108] In another aspect, the present invention relates to a system for automatically processing a 3D image of an individual to determine the value of one or more patient indicators measuring the individual's (e.g., overall) disease burden and / or risk, the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) receiving a 3D functional image of the individual obtained using a functional imaging modality; (b) segmenting a plurality of 3D hotspot volumes within the 3D functional image, each 3D hotspot volume corresponding to a local region having an elevated intensity relative to its surroundings and representing a potential cancerous lesion within the individual, thereby obtaining a set of 3D hotspot volumes; and (c) for each specific metric in one or more individual hotspot quantification metrics, calculating the set (d) determining the value of one or more patient indicators, wherein each of at least a portion of the patient indicators is associated with the one or more specific individual hotspot quantification metrics and is calculated using (e.g., the same) specific function of the intensity and / or number of stereo pixels within the combined hotspot volume, wherein the combined hotspot volume includes at least a portion (e.g., substantially all; e.g., a specific subset) of the set of 3D hotspot volumes (e.g., formed as a union thereof).

[0109] In certain embodiments, the patient-specific metric is an overall average voxel intensity and is calculated as an overall average of the voxel intensities within the combined hotspot volume.

[0110] In another aspect, the present invention relates to a system for automatically determining the prognosis of an individual with prostate cancer from one or more medical images of the individual [e.g., one or more PSMA PET images (PET images obtained after administration of a targeted PSMA compound to the individual) and / or one or more anatomical (e.g., CT) images], the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) receive and / or access one or more images of the individual; (b) automatically determine a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) from the one or more images [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) molecular imaging TNM (miTNM) lesion type classifications (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) of localized (T), pelvic node (N), and / or extrapelvic (M) disease; (ii) indicate lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) standard absorption value (standard uptake value (SUV) (e.g., maximum SUV, peak SUV, average SUV); (iv) total lesion volume; (v) change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein); and (c) automatically determining the prognosis of the individual from the quantitative assessment in (b), wherein the prognosis comprises one or more of the following for the individual: (I) expected survival (e.g., number of months), (II) expected time to disease progression, (III) expected time to radiographic progression, (IV) risk of simultaneous (synchronous) cancer metastasis, and (V) risk of future (metachronous) cancer metastasis.

[0111] In certain embodiments, the quantitative assessment of the one or more prostate cancer lesions determined in step (b) comprises one or more of: (A) total tumor volume, (B) change in tumor volume, (C) total SUV, and (D) PSMA score, and wherein the prognosis of the individual determined in step (c) comprises one or more of: (E) expected survival (e.g., months), (F) time to progression, and (G) time to radiographic progression.

[0112] In certain embodiments, the quantitative assessment of the one or more prostate cancer lesions determined in step (b) comprises one or more characteristics of PSMA expression in the prostate, and wherein the prognosis of the individual determined in step (c) comprises the risk of simultaneous (synchronous) cancer metastasis and / or the risk of future (metachronous) cancer metastasis.

[0113] In another aspect, the present invention relates to a system for automatically determining a response to treatment in an individual with prostate cancer from a plurality of medical images of the individual [e.g., one or more PSMA PET images (PET images obtained after administration of a targeted PSMA compound to the individual) and / or one or more anatomical (e.g., CT) images], the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to perform the following operations: (a) receiving and / or accessing, by the processor of the computing device, a plurality of images of the individual, wherein at least a first image of the plurality of images is obtained before administration of the treatment and at least a second image of the plurality of images is obtained after administration of the treatment (e.g., after a period of time); (b) automatically determining a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) from the images [e.g., wherein the quantitative The quantitative assessment comprises one or more members selected from the group consisting of: (i) a molecular imaging TNM (miTNM) lesion type classification (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) for local (T), pelvic node (N) and / or extrapelvic (M) disease; (ii) an indication of lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) a standard physiological uptake value (SUV) (e.g., SUV maximum, SUV peak, SUV mean); (iv) total lesion volume; (v) a change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) a calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein) (e.g., wherein the quantitative assessment comprises a PSMA imaging of response evaluation criteria (RECIP) criteria and / or a PSMA PET progression (PPP) criteria); and (c) automatically determining from the quantitative assessment in (b) whether the individual is responsive to treatment (e.g., responsive / non-responsive) and / or the extent to which the individual is responsive to treatment (e.g., numerical or categorical).

[0114] In another aspect, the present invention relates to a system for automatically identifying whether an individual with prostate cancer (e.g., metastatic prostate cancer) is likely to benefit from a particular treatment for prostate cancer using a plurality of medical images of the individual [e.g., one or more PSMA PET images (PET images obtained after administration of a targeted PSMA compound to the individual) and / or one or more anatomical (e.g., CT) images], the system comprising: a processor of a computing device; and a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receive and / or access a plurality of images of the individual; (b) automatically determine from the images a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i ) molecular imaging TNM (miTNM) lesion type classification (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) for local (T), pelvic node (N), and / or extrapelvic (M) disease; (ii) indication of lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) standard physiological uptake value (SUV) (e.g., SUV maximum, SUV peak, SUV mean); (iv) total lesion volume; (v) change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein) (e.g., where the quantitative assessment comprises the response evaluation criteria PSMA imaging (RECIP) criteria and / or the PSMA PET progression (PPP) criteria); and (c) automatically determining from the quantitative assessment in (b) whether an individual is likely to benefit from a specific treatment for prostate cancer [e.g., determining one or more specific treatments and / or a class of treatments for the individual, such as a specific radioligand therapy, such as leuprorelin eligibility score].

[0115] In another aspect, the invention relates to a therapeutic agent for treating (e.g., with multiple cycles of the therapeutic agent) an individual having a particular disease (e.g., prostate cancer (e.g., metastatic castration-resistant prostate cancer)) and / or at risk for a particular disease, wherein the individual has (i) been administered a first cycle of the therapeutic agent and imaged (e.g., before and / or during and / or after the first cycle of the therapeutic agent), and (ii) identified as a responder to the therapeutic agent using the methods described herein, e.g., in the above paragraphs, e.g., paragraphs

[0011] -

[0060] ) (e.g., based on the values ​​of one or more risk indicators determined using the methods described herein, e.g., in the above paragraphs, e.g., paragraphs

[0011] -

[0060] , the individual has been identified / classified as a responder).

[0116] In another aspect, the invention relates to a second (e.g., second line) therapeutic agent for treating an individual having a particular disease (e.g., prostate cancer (e.g., metastatic castration-resistant prostate cancer)) and / or at risk for a particular disease, wherein the individual has (i) been administered an initial, cycle of a first therapeutic agent and imaged (e.g., before and / or during and / or after the cycle of the first therapeutic agent), and (ii) identified as a non-responder to the first therapeutic agent using the methods described herein, e.g., in the above paragraphs, e.g., paragraphs

[0011] -

[0060] ) (e.g., the individual has been identified / classified as a non-responder based on the values ​​of one or more risk indicators determined using the methods described herein, e.g., in the above paragraphs, e.g., paragraphs

[0011] -

[0060] ) (e.g., thereby allowing the individual to receive a potentially more effective therapy).

[0117] Features of the embodiments described with respect to one aspect of the invention may be applied with respect to another aspect of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0118] The foregoing and other objects, aspects, features and advantages of the present disclosure will become more apparent and better understood by referring to the following description in conjunction with the accompanying drawings, in which:

[0119] Figure 1A is a CT image obtained from a 3D PET / CT scan, a PET image, and a set of corresponding slices of a PET / CT fusion in accordance with an illustrative embodiment.

[0120] Figure 1B is a set of two slices of a PET / CT composite image in which the PET image is overlaid on a CT scan in accordance with an illustrative embodiment.

[0121] Figure 2 is a diagram depicting an example procedure for segmenting anatomical images and identifying anatomical boundaries in co-aligned functional images in accordance with an illustrative embodiment.

[0122] Figure 3 is a diagram illustrating an example process for segmenting and classifying hotspots according to an illustrative embodiment.

[0123] Figure 4A is a graphical user interface (GUI) that displays a computer-generated patient report obtained by the image analysis and decision support tool of the present disclosure according to an illustrative embodiment.

[0124] Figure 4B is another screen shot of a computer-generated report presenting longitudinal data tracking disease burden and evolution over time in accordance with an illustrative embodiment.

[0125] Figure 4C is a schematic diagram showing a method for calculating a lesion index value according to an illustrative embodiment.

[0126] Figure 5 is a block diagram illustrating an example procedure for tracking lesions and determining changes in hot spot quantification and / or patient index values.

[0127] Fig. 6A is a schematic diagram illustrating the evolution of hot spots identified in an initial baseline scan and then in a second subsequent scan in accordance with an illustrative embodiment.

[0128] Figure 6B is a schematic diagram illustrating the evolution of hot spots identified during an initial baseline scan and subsequently during a second follow-up scan in accordance with an illustrative embodiment.

[0129] Figure 6C is a schematic diagram illustrating the evolution of hot spots identified during an initial baseline scan and subsequently during a second follow-up scan in accordance with an illustrative embodiment.

[0130] Figure 7 is a block diagram of an example procedure for determining and using lesion correspondence to determine patient metrics and / or classifications in accordance with an illustrative embodiment.

[0131] Figure 8 is a block diagram showing an example procedure for determining lesion correspondence in accordance with an illustrative embodiment.

[0132] Fig. 9A are images showing example registration using an anatomical segmentation map in accordance with an illustrative embodiment.

[0133] Fig. 9B is another image showing an example registration using an anatomical segmentation map in accordance with an illustrative embodiment.

[0134] Fig. 9C is another image showing an example registration using an anatomical segmentation map in accordance with an illustrative embodiment.

[0135] Fig.10 is a set of three composite images (shown twice as a "first scan" for purposes of illustration) showing the registration of a composite image obtained by a second scan with a composite image obtained by a first scan in accordance with an illustrative embodiment.

[0136] Fig.11A is a schematic diagram illustrating registration between a second image obtained through a second scan and a first image obtained through a first scan according to an illustrative embodiment.

[0137] Fig. 11Bis a schematic diagram illustrating registration between a second image obtained through a second scan and a first image obtained through a first scan according to an illustrative embodiment.

[0138] Fig.12 is a set of three schematic diagrams showing three lesion correspondence metrics according to an illustrative embodiment.

[0139] Fig.13 is a block diagram of an exemplary cloud computing environment used in certain embodiments.

[0140] Fig.14 is a block diagram of an example computing device and an example mobile computing device for use in certain embodiments.

[0141] The features and advantages of the present disclosure will become more apparent from the embodiments described below in conjunction with the drawings, in which the same reference characters always identify corresponding elements. In the drawings, the same reference numerals generally indicate the same, functionally similar and / or structurally similar elements. Some definitions

[0142] In order to make the present disclosure more easily understood, certain terms are first defined below. The following terms and additional definitions of other terms are set forth throughout this specification.

[0143] A, an: The articles "a" and "an" are used herein to refer to one or to more than one (i.e., to at least one) of the grammatical object of the article. By way of example, "an" means one element or more than one element. Thus, in this specification and the appended claims, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. Thus, for example, reference to a pharmaceutical composition comprising "an agent" includes reference to two or more agents.

[0144] About, approximately: As used in the present application, the term "about" is used equivalently to "approximately". Any numerical value used in the present application in the presence or absence of about / roughly is intended to cover any normal fluctuations understood by a person of ordinary skill in the relevant art. In certain embodiments, unless otherwise stated or otherwise apparent from the context, the term "about" or "approximately" refers to a range of values ​​(unless such numerical value will exceed 100% of a possible value) in either direction of the stated reference value (greater than or less than) of 25%, 20%, 19%, 18%, 17%, 16%, 15%, 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1% or less.

[0145] First, second, etc.: It should be understood that any reference to elements herein using names such as "first," "second," etc. does not limit the number or order of those elements unless such limitation is explicitly stated. In fact, these names may be used herein as a conventional method of distinguishing two or more elements or instances of elements. Thus, reference to a first element and a second element does not mean that only two elements may be used or that the first element must precede the second element in some way. Additionally, unless otherwise stated, a collection of elements may include one or more elements.

[0146] Image: As used herein, "image" - e.g., a 3D image of an individual, includes any visual representation such as a photograph, a video frame, a streaming video, and any electronic, digital, or mathematical simulation of a photograph (e.g., a digital image), a video frame, or a streaming video, displayed or stored in a memory (e.g., a digital image may be displayed but need not be displayed for visual inspection). In certain embodiments, any device described herein includes a display for displaying an image or any other result produced by a processor. In certain embodiments, any method described herein includes a step of displaying an image or any other result produced by the method. In certain embodiments, an image is a 3D image that conveys information that varies with position within a 3D volume. Such an image may be represented, for example, digitally as a 3D matrix (e.g., an N×M×L matrix), where each stereo pixel of the 3D image is represented by an element of the 3D matrix. Other representations are also contemplated and included, for example, a 3D matrix may be reshaped into a vector (e.g., a 1×K dimensional vector, where K is the total number of stereo pixels) by splicing rows or columns end to end. Examples of images include, for example, medical images, such as bone scan images (also known as scintigraphy images), computed tomography (CT) images, magnetic resonance images (MRI), optical images (e.g., bright field microscopy images, fluorescence images, reflection or transmission images, etc.), positron emission tomography (PET) images, single photon emission tomography (SPECT) images, ultrasound images, x-ray images, etc. In some embodiments, the medical image is or includes a nuclear medicine image, which is generated by radiation emitted from within the individual being imaged. In some embodiments, the medical image is or includes an anatomical image (e.g., a 3D anatomical image) that conveys information about the location and extent of anatomical structures such as viscera, bones, soft tissues, and blood vessels within the individual. Examples of anatomical images include, but are not limited to, x-ray images, CT images, MRI, and ultrasound images. In some embodiments, the medical image is or includes a functional image (e.g., a 3D functional image) that conveys information related to physiological activities within a specific organ and / or tissue, such as metabolism, blood flow, regional chemical composition, absorption, etc. Examples of functional images include, but are not limited to, nuclear medicine images, such as PET images, SPECT images, and other functional imaging modalities, such as functional MRI (fMRI), which measures small changes in blood flow for assessing brain activity.

[0147] Map: As used herein, the term "map" is understood to mean a visual display or any data representation that can be interpreted for a visual display, which contains spatially related information. For example, a three-dimensional map of a given volume may include a data set of values ​​of a given number that varies across the three spatial dimensions of the volume. A three-dimensional map may be displayed in two dimensions, such as on a two-dimensional screen or on a two-dimensional printed output.

[0148] Segmentation map: As used herein, the term "segmentation map" refers to a computer representation that identifies one or more 2D or 3D regions determined by segmenting an image. In certain embodiments, a segmentation map distinguishably identifies multiple different (e.g., segmented) regions, thereby allowing the regions to be individually and distinguishably accessed and manipulated, e.g., one or more images and / or used to operate on, e.g., one or more images.

[0149] 3D, three-dimensional: As used herein, "3D" or "three-dimensional" with reference to an "image" means conveying information about three dimensions. A 3D image may be presented as a three-dimensional data set and / or may be displayed as a collection of two-dimensional representations or as a three-dimensional representation. In certain embodiments, a 3D image is represented as voxel (e.g., volumetric pixel) data.

[0150] Full body: As used herein, in the context of segmenting and otherwise identifying regions within an image of an individual, the terms "full body" and "whole body" are used (interchangeably) to refer to methods of evaluating a large portion (e.g., greater than 50%) of a graphical representation of an individual's body in a 3D anatomical image to identify a target tissue region of interest. In certain embodiments, full body and whole body segmentation refers to identification of a target tissue region within at least the entire torso of an individual. In certain embodiments, portions of limbs as well as the head of an individual are also included.

[0151] Radionuclides: As used herein, "radionuclide" refers to a moiety that contains a radioactive isotope of at least one element. Exemplary suitable radionuclides include, but are not limited to, those described herein. In some embodiments, the radionuclide is one used in positron emission tomography (PET). In some embodiments, the radionuclide is one used in single photon emission computed tomography (SPECT). In some embodiments, a non-limiting list of radionuclides includes 99m Tc, 111 In, 64 Cu, 67 Ga, 68 Ga, 186 Re, 188 Re, 153 Sm, 177 Lu, 67 Cu, 123 I. 124 I. 125 I. 126 I. 131 I. 11 C.13 N. 15 O. 18 F. 153 Sm, 166 Ho, 177 Lu, 149 Pm, 90 Y. 213 Bi, 103 Pd, 109 Pd, 159 Gd, 140 La, 198 Au, 199 Au, 169 Yb, 175 Yb, 165 Dy, 166 Dy, 105 Rh, 111 Ag, 89 Zr, 225 Ac, 82 Rb, 75 Br, 76 Br, 77 Br, 80 Br, 80m Br, 82 Br, 83 Br, 211 At and 192 Ir.

[0152] Radiopharmaceuticals: As used herein, the term "radiopharmaceutical" refers to a compound that contains a radionuclide. In certain embodiments, radiopharmaceuticals are used for diagnostic and / or therapeutic purposes. In certain embodiments, radiopharmaceuticals include small molecules labeled with one or more radionuclides, antibodies labeled with one or more radionuclides, and antigen-binding portions of antibodies labeled with one or more radionuclides.

[0153] Machine learning module: Certain embodiments described herein are aided by (e.g., include) software instructions that include one or more machine learning modules, also referred to herein as artificial intelligence software. As used herein, the term "machine learning module" refers to a computer-implemented program (e.g., function) that implements one or more specific machine learning algorithms in order to determine one or more output values ​​for a given input (e.g., an image (e.g., a 2D image; e.g., a 3D image), a data set, etc.). For example, a machine learning module may receive a 3D image (e.g., a CT image; e.g., an MRI) of an individual as input, and for each stereo pixel of the image, determine a value representing the likelihood that the stereo pixel is within a region of the 3D image corresponding to a representation of a specific organ or tissue of the individual. In certain embodiments, two or more machine learning modules may be combined and implemented in a single module and / or a single software application. In certain embodiments, two or more machine learning modules may also be implemented separately, such as in separate software applications. A machine learning module may be software and / or hardware. For example, a machine learning module may be implemented entirely in software, or certain functions of a CNN module may be performed by dedicated hardware (e.g., by an application specific integrated circuit (ASIC)).

[0154] Subject: As used herein, "subject" means a human or other mammal (eg, rodent (mice, rats, hamsters), pig, cat, dog, horse, primate, rabbit, etc.).

[0155] Administration: As used herein, "administering" an agent means introducing a substance (e.g., an imaging agent) into a subject. Generally, any route of administration can be utilized, including, for example, parenteral (e.g., intravenous), oral, topical, subcutaneous, peritoneal, intraarterial, inhalation, vaginal, rectal, nasal, introduction into cerebrospinal fluid, or instillation into a body compartment.

[0156] Tissue: As used herein, the term "tissue" refers to bone (osseous tissue) as well as soft tissue. DETAILED DESCRIPTION

[0157] The claimed inventive systems, architectures, devices, methods, and programs are considered to cover changes and adaptations made using the information of the embodiments described herein. Adaptations and / or modifications of the systems, architectures, devices, methods, and programs described herein may be made as contemplated by this specification.

[0158] Throughout this specification, where articles, devices, systems, and architectures are described as having, including, or comprising particular components, or where processes and methods are described as having, including, or comprising particular steps, it is contemplated that there are additional articles, devices, systems, and architectures of the invention that consist essentially of, or consist of, the enumerated components, and there are processes and methods according to the invention that consist essentially of, or consist of, the enumerated processing steps.

[0159] It should be understood that the order of steps or the order in which a certain action is performed is not important, as long as the present invention remains operable. In addition, two or more steps or actions may be performed simultaneously.

[0160] Any publication mentioned herein (e.g., in the prior art section) is not an admission that the publication serves as prior art with respect to any of the claims presented herein. The prior art section is present for the purpose of clarity and is not intended to be a description of prior art with respect to any claim.

[0161] As noted, the documents are incorporated herein by reference. In the event of any discrepancy in the meaning of a particular term, the meaning provided in the above definitions section shall prevail.

[0162] The headings are provided for the convenience of the reader - the presence and / or placement of the headings is not intended to limit the scope of the subject matter described herein. A. Nuclear Medicine Images

[0163] Nuclear medicine images may be obtained using nuclear medicine imaging modalities such as bone scan imaging (also known as scintigraphy), positron emission tomography (PET) imaging, and single photon emission tomography (SPECT) imaging.

[0164] In certain embodiments, nuclear medicine images are obtained using an imaging agent that includes a radiopharmaceutical. Nuclear medicine images can be obtained after administration of a radiopharmaceutical to a patient (eg, a human subject) and provide information about the distribution of the radiopharmaceutical within the patient.

[0165] Nuclear medicine imaging techniques detect radiation emitted by radionuclides of radiopharmaceuticals to form images. The distribution of a particular radiopharmaceutical in a patient's body may be affected and / or regulated by biological mechanisms such as blood flow or perfusion and by specific enzymatic or receptor binding interactions. Different radiopharmaceuticals may be designed to utilize different biological mechanisms and / or specific enzymatic or receptor binding interactions, and therefore selectively concentrate in specific tissue types and / or regions within the patient when administered to the patient. Larger amounts of radiation are emitted from regions within the patient that have higher radiopharmaceutical concentrations than other regions, making these regions appear brighter in nuclear medicine images. Therefore, intensity changes within nuclear medicine images can be used to map the distribution of radiopharmaceuticals within the patient. This mapped distribution of radiopharmaceuticals within a patient can be used, for example, to infer the presence of cancerous tissue in different regions of the patient's body. In certain embodiments, the intensity of a stereoscopic pixel of a nuclear medicine image such as a PET image represents a standard absorption value (SUV) (e.g., calibrated for the injected radiopharmaceutical dose and / or patient weight).

[0166] For example, after administration to a patient, technetium-99m methylene diphosphonate ( 99m Tc MDP) selectively accumulates in bone regions of the patient, particularly at sites of abnormal bone formation associated with malignant bone lesions. The selective concentration of the radiopharmaceutical at these sites produces identifiable hot spots, i.e., localized areas of high intensity, in nuclear medicine images. Thus, the presence of malignant bone lesions associated with metastatic prostate cancer can be inferred by identifying such hot spots within a whole-body scan of the patient. In certain embodiments, the analysis of the presence of malignant bone lesions associated with metastatic prostate cancer is performed, for example, by detecting and evaluating the characteristics of the hot spots. 99m The intensity changes in whole body scans obtained after Tc MDP can be used to calculate risk indicators related to overall patient survival and other prognostic metrics indicative of disease status, progression, treatment efficacy, etc. In certain embodiments, other radiopharmaceuticals can also be used with 99m Tc MDP is used in a similar manner.

[0167] In certain embodiments, the specific radiopharmaceutical used depends on the specific nuclear medicine imaging modality used. For example, 18F sodium fluoride (NaF) also accumulates in bone lesions (similar to 99m Tc MDP), but can be used for PET imaging. In certain embodiments, PET imaging can also utilize a radioactive form of the vitamin choline, which is readily taken up by prostate cancer cells.

[0168] In certain embodiments, radiopharmaceuticals may be used that selectively bind to specific proteins or receptors of interest, particularly those whose expression is increased in cancer tissue. Such proteins or receptors of interest include, but are not limited to, tumor antigens such as CEA, which is expressed in colorectal cancer; Her2 / neu, which is expressed in a variety of cancers; BRCA 1 and BRCA 2, which are expressed in breast and ovarian cancers; and TRP-1 and TRP-2, which are expressed in melanoma.

[0169] For example, human prostate-specific membrane antigen (PSMA) is upregulated in prostate cancer, including metastatic disease. Almost all prostate cancers express PSMA and its expression is further increased in poorly differentiated metastatic and hormone-refractory carcinomas. Therefore, a radiopharmaceutical containing a PSMA binder (e.g., a compound with high affinity for PSMA) labeled with one or more radionuclides can be used to obtain a nuclear medicine image of a patient, from which the presence and / or state of prostate cancer in multiple regions of the patient (e.g., including but not limited to bone regions) can be assessed. In certain embodiments, when the disease is localized, nuclear medicine images obtained using a PSMA binder are used to identify the presence of cancerous tissue within the prostate. In certain embodiments, when the disease is metastatic, nuclear medicine images obtained using a radiopharmaceutical containing a PSMA binder are used to identify the presence of cancerous tissue in multiple regions, including not only the prostate, but also other organs and tissue regions associated with the patient, such as lungs, lymph nodes, and bones.

[0170] Specifically, after administration to a patient, the radionuclide-labeled PSMA-binding agent selectively accumulates in cancer tissue based on its affinity for PSMA. 99m In a manner similar to that described for the Tc MDP, the selective concentration of a radionuclide-labeled PSMA binder at a specific site within a patient creates a detectable hot spot in a nuclear medicine image. When the PSMA binder is concentrated in multiple cancerous tissues and regions of the body that express PSMA, localized cancer within the patient's prostate and / or metastatic cancer in different regions of the patient's body can be detected and assessed. Various metrics that indicate and / or quantify the severity of individual lesions (e.g., possible malignancies), the patient's overall disease burden and risk, etc. can be calculated based on automated analysis of intensity changes in nuclear medicine images obtained after administration of the PSMA binder radiopharmaceutical to the patient. These disease burden and / or risk metrics can be used to stage the disease and make assessments regarding the patient's overall survival and other prognostic metrics that indicate disease status, progression, and efficacy of treatment.

[0171] A variety of radionuclide-labeled PSMA binding agents can be used as radiopharmaceutical imaging agents for nuclear medicine imaging to detect and evaluate prostate cancer. In certain embodiments, the specific radionuclide-labeled PSMA binding agent used depends on factors such as the specific imaging modality (e.g., PET; e.g., SPECT) and the specific area (e.g., organ) of the patient to be imaged. For example, certain radionuclide-labeled PSMA binding agents are suitable for PET imaging, while others are suitable for SPECT imaging. For example, certain radionuclide-labeled PSMA binding agents help image the patient's prostate and are used primarily when the disease is localized, while others help image organs and areas throughout the patient's body and are suitable for evaluating metastatic prostate cancer.

[0172] Several exemplary PSMA-binding agents and versions of their radionuclide labeling are described in further detail in Section H herein, as well as in U.S. Pat. Nos. 8,778,305, 8,211,401, and 8,962,799, and U.S. Patent Publication No. US2021 / 0032206 A1, the contents of each of which are incorporated herein by reference in their entirety. B. Image Segmentation in Nuclear Medicine Imaging

[0173] Nuclear medicine images are functional images. Functional images convey information related to physiological activities within specific organs and / or tissues, such as metabolism, blood flow, regional chemical composition and / or absorption. In certain embodiments, nuclear medicine images are acquired and / or analyzed in combination with anatomical images, such as computed tomography (CT) images. Anatomical images provide information about the location and extent of anatomical structures such as internal organs, bones, soft tissues, and blood vessels within an individual. Examples of anatomical images include, but are not limited to, x-ray images, CT images, magnetic resonance images, and ultrasound images.

[0174] Thus, in certain embodiments, anatomical images as well as nuclear medicine images may be analyzed to provide anatomical context for the functional information they convey. For example, while nuclear medicine images such as PET and SPECT convey the three-dimensional distribution of a radiopharmaceutical within an individual, adding anatomical context from anatomical imaging modalities such as CT imaging allows for the determination of specific organs, soft tissue regions, bones, etc., in which the radiopharmaceutical has accumulated.

[0175] For example, the functional image can be aligned with the anatomical image so that locations within each image that correspond to the same body location and therefore correspond to each other can be identified. For example, coordinates and / or pixels / voxels within the functional and anatomical images can be defined relative to a common coordinate system, or a mapping (i.e., a functional relationship) between voxels within the anatomical image and voxels within the functional image can be established. In this way, one or more voxels within the anatomical image and one or more voxels within the functional image that represent the same body location or volume can be identified as corresponding to each other.

[0176] For example, FIG1 shows axial slices of a 3D CT image 102 and a 3D PET image 104, and a fused image 106, wherein the slices of the 3D CT image are displayed in grayscale and wherein the PET image is displayed as a semi-transparent overlay. With the help of the registration between the CT and PET images, the location of hot spots within the PET image indicating accumulated radiopharmaceuticals and corresponding potential lesions can be identified in the corresponding CT image and viewed in an anatomical situation, such as within a specific location in the pelvic region (e.g., within the prostate). Figure 1B Another PET / CT fusion is shown showing transverse and sagittal slices.

[0177] In some embodiments, the registration pair is a composite image, such as PET / CT or SPECT / CT. In some embodiments, anatomical images (e.g., 3D anatomical images, such as CT images) and functional images (e.g., 3D functional images, such as PET or SPECT images) are acquired using separate anatomical and functional imaging modalities, respectively. In some embodiments, anatomical images (e.g., 3D anatomical images, such as CT images) and functional images (e.g., 3D functional images, such as PET or SPECT images) are acquired using a single multimodal imaging system. The functional images and anatomical images can be acquired using a single multimodal imaging system, for example, by two scans, such as a CT scan first and then a PET scan a second time, during which the individual remains in a substantially fixed position.

[0178] In some embodiments, the 3D boundaries of specific tissue regions of interest can be accurately identified by analyzing the 3D anatomical images. For example, automatic segmentation of the 3D anatomical images can be performed so that the 3D boundaries of regions such as specific organs, organ sub-regions and soft tissue regions, and bones are segmented. In some embodiments, organs such as the prostate, bladder, liver, aorta (e.g., portions of the aorta, such as the thoracic aorta), parotid glands, etc. are segmented. In some embodiments, one or more specific bones are segmented. In some embodiments, the overall framework is segmented.

[0179] In certain embodiments, automatic segmentation of 3D anatomical images can be performed using one or more machine learning modules that are trained to receive a 3D anatomical image and / or a portion thereof as input and segment one or more specific regions of interest, thereby generating a 3D segmentation map as output. For example, as described in PCT Publication WO / 2020 / 144134, entitled “Systems and Methods for Platform Agnostic Whole Body Segmentation” and published on July 16, 2020, the contents of which are incorporated herein by reference in their entirety, multiple machine learning modules implementing convolutional neural networks (CNNs) can be used to segment a 3D anatomical image, such as a CT image, of the entire body of an individual and thereby generate a 3D segmentation map that identifies multiple target tissue regions in the individual's body.

[0180] In certain embodiments, for example, to segment certain organs (where it is believed that functional images provide additional useful information that facilitates segmentation), the machine learning module may receive both anatomical images and functional images as inputs, such as two different channels as inputs (e.g., similar to color, RGB, multiple color channels in an image) and use these two inputs to determine anatomical segmentation. This multi-channel approach is further described in detail in U.S. Patent Publication No. 2021 / 0334974A1, entitled "Systems and Methods for Deep-Learning-Based Segmentation of Composite Images" and published on October 28, 2021, the contents of which are incorporated herein by reference in their entirety.

[0181] In certain embodiments, such as Figure 2As shown, an anatomical image 204 (e.g., a 3D anatomical image, such as a CT image) and a functional image 206 (e.g., a 3D functional image, such as a PET or SPECT image) can be aligned (e.g., co-registered) with each other, such as in a composite image 202, such as a PET / CT image. The anatomical image 204 can be segmented 208 to generate a segmentation map 210 (e.g., a 3D segmentation map) that can distinguishably identify one or more tissue regions and / or sub-regions of interest, such as one or more specific organs and / or bones. The segmentation map 210 that has been generated from the anatomical image 204 is aligned with the anatomical image 204, which is in turn aligned with the functional image 206. Therefore, the boundaries of specific regions (e.g., segmentation masks), such as specific organs and / or bones, identified by the segmentation map 210 can be transferred to the functional image 206 and / or overlapped 212 with it to identify volumes within the functional image 206 for the purpose of hotspot classification, and determine useful indicators that serve as measurements and / or predictions of cancer status, progression, and response to treatment. Segment maps and masks may also be displayed as graphical representations, for example, overlaid on medical images to guide physicians and other medical practitioners. C. Lesion Detection and Characterization

[0182] In certain embodiments, the methods described herein include techniques for detecting and characterizing lesions within an individual by (e.g., automatically) analyzing medical images, such as nuclear medicine images. As described herein, in certain embodiments, a hotspot is a localized (e.g., continuous) area of ​​high intensity relative to its surroundings within an image, such as a 3D functional image, and may indicate the presence of a potential cancerous lesion within an individual.

[0183] A variety of methods can be used to detect, segment and classify hot spots. In some embodiments, hot spots are detected and segmented using analytical methods, such as filtering techniques, including but not limited to Gaussian difference (Difference of Gaussians, DoG) filters and Gaussian Laplacian (Laplacian of Gaussians, LoG) filters. In some embodiments, hot spots are segmented using a machine learning module, which receives a 3D functional image such as a PET image as input and generates a hot spot segmentation map ("hot spot map") that separates the boundaries of the identified hot spots from the background as output. In some embodiments, each segmented hot spot in the hot spot map can be identified separately (e.g., marked separately). In some embodiments, in addition to the 3D functional image, the machine learning module for segmenting hot spots can also treat one or two of the 3D anatomical image (e.g., CT image) and the 3D anatomical segmentation map as input. The 3D anatomical segmentation map can be generated by automatic segmentation of the 3D anatomical image (e.g., described herein).

[0184] In some embodiments, the segmented hotspots may be classified according to the anatomical region in which the segmented hotspots are located. For example, in some embodiments, the locations of individual segmented hotspots within a heat map (representing and identifying segmented hotspots) may be compared with 3D boundaries of segmented tissue regions such as various organs and bones within a 3D anatomical segmented map and labeled according to their locations (e.g., based on proximity to and / or overlap with a specific organ). In some embodiments, a machine learning module may be used to classify hotspots. For example, in some embodiments, a machine learning module may generate a heat map as an output in which segmented hotspots are not only individually labeled and identifiable (e.g., distinguishable from each other), but are also labeled as corresponding to one of bone, lymphatic, or prostate lesions, for example. In some embodiments, one or more machine learning modules may be combined with each other, as well as with analytical segmentation (e.g., thresholding) techniques, to perform various tasks in parallel and sequentially to produce a final labeled heat map.

[0185] Various methods for performing detailed segmentation of 3D anatomical images and identifying hotspots representing lesions in 3D functional images that can be used in conjunction with the various methods described herein are described in: PCT Publication WO / 2020 / 144134, entitled “Systems and Methods for Platform-Independent Whole-Body Segmentation” and published on July 16, 2020; U.S. Patent Publication No. US2021 / 0334974 A1, entitled “Systems and Methods for Deep Learning-Based Composite Image Segmentation” and published on October 28, 2021; and PCT Publication WO / 2022 / 008374, entitled “Systems and Methods for Artificial Intelligence-Based Image Analysis for Detection and Characterization of Lesions” and published on January 13, 2022, the contents of each of which are incorporated herein in their entirety.

[0186] Figure 3 An example procedure 300 is shown for segmenting and classifying hotspots based on an example method further described in detail in PCT Publication WO 2022 / 008374, which is entitled "Systems and methods for artificial intelligence-based image analysis for detecting and characterizing lesions" and was published on January 13, 2022. Figure 3The method shown in uses two machine learning modules, each of which receives a 3D functional image 306, a 3D anatomical image 304, and a 3D anatomical segmentation map 310 as input. The machine learning module 312a is a binary classifier that generates a single-class heat map 320a by marking stereo pixels as hot spots or backgrounds (not hot spots). The machine learning module 312b performs multi-class segmentation and generates a multi-class heat map 320b, wherein the hot spots are segmented and marked as one of three categories, prostate, lymph, or bone. In addition, in this way, the need to segment a specific area is avoided by classifying the hot spots by the machine learning module 312b (e.g., relative to directly comparing the hot spot location with the segmentation boundary from the segmentation map 310). For example, in some embodiments, the machine learning module 312b can classify the hot spots as belonging to the prostate, lymph, or bone, without the prostate region that has been identified and segmented from the 3D anatomical image 304 (e.g., in some embodiments, the 3D anatomical segmentation map 310 does not include the prostate region). In some embodiments, the heat maps 320a and 320b are merged, for example, by transferring labels from the multi-class heat map 320b to the hotspot segments identified in the single-class heat map 320a (e.g., based on overlap). Without wishing to be bound by any particular theory, it is believed that this method combines the improved segmentation and detection of hotspots from the single-class machine learning module 312a with the classification results from the multi-class machine learning module 312b. In some embodiments, the hotspot areas identified by this final merged heat map are further improved using analysis techniques such as the adaptive limit technique described in PCT Publication WO / 2022 / 008374, which is entitled "Systems and Methods for Artificial Intelligence-Based Image Analysis for Detecting and Characterizing Lesions" and published on January 13, 2022.

[0187] In some embodiments, once detected and segmented, the hot spot can be identified and assigned a mark according to the specific anatomical (e.g., tissue) region where it is located and / or the specific lesion subtype that it is likely to represent. For example, in some embodiments, the hot spot can be assigned with an anatomical position, which is identified as representing a position in a set of tissue regions such as listed in Table 1 below. In some embodiments, the list of tissue regions can include those tissue regions in Table 1 and gluteus maximus (e.g., left and right) and gallbladder. In some embodiments, based on machine learning classification and / or by comparing the position of the 3D hot spot volume of the hot spot and / or with the overlap of various tissue volumes identified by the mask in the anatomical segmentation diagram, the hot spot is assigned to a specific tissue region and / or marked as belonging to a specific tissue region. In some embodiments, the prostate is not segmented. For example, as described above, in some embodiments, the machine learning module 312b can classify the hot spot as belonging to the prostate, lymph or bone, without the prostate region identified and segmented from the 3D anatomical image 304. Table 1: Certain tissue regions (*in certain embodiments, the prostate may optionally be segmented if present, may not be present if the patient has, for example, undergone a radical prostatectomy, or may not be segmented in any case)

[0188] In some embodiments, in addition or alternatively, the hotspot may be classified as belonging to one or more lesion subtypes. In some embodiments, the lesion subtype classification may be performed by comparing the hotspot location with the category of the anatomical region. For example, in some embodiments, the miTNM classification scheme may be used, wherein the hotspot is labeled as belonging to one of the three categories of miT, miN, or miM based on whether the hotspot represents a lesion located in the prostate (miT), pelvic lymph nodes (miN), or distant cancer metastasis (miM). In some embodiments, a five-type miTNM scheme may be used, wherein distant cancer metastasis is further divided into three subclasses: miMb for bone cancer metastasis, miMa for lymphatic cancer metastasis, and miMc for other soft tissue cancer metastasis.

[0189] For example, in some embodiments, hot spots located within the prostate are marked as belonging to, for example, a "T" class or a "miT" class representing a local tumor. In some embodiments, hot spots located outside the prostate but within the pelvic region are marked as "N" class or "miN" class. In some embodiments, such as the system and method for Automated Identification and Classification of Lesions in Local Lymph and Distant Metastases (Systems and Methods for Automated Identification and Classification of Lesions in Local Lymph and Distant Metastases) filed on October 4, 2022, US2023 / 0115732A1, U.S. Application No. 17 / 959,357 published on April 13, 2023 (the contents of which are incorporated herein by reference in their entirety), for the purpose of identifying pelvic lymph node lesions, a pelvic atlas can be registered to identify the boundaries of the pelvic region and / or different sub-regions therein. The pelvic atlas may, for example, include boundaries and / or plane references of the pelvic region (e.g., a plane passing through the aortic bifurcation), which may be compared to hotspot locations (e.g., such that hotspots located outside the pelvic region and / or above the plane reference passing through the aortic bifurcation are labeled "M" or "miM," e.g., distal metastases). In certain embodiments, based on comparison of the hotspot locations to the anatomical segmentation map, distal metastases may be classified as lymphatic (miMa), bone (miMb), or visceral (miMc). For example, hotspots located within one or more bones (e.g., and outside the pelvic region) may be labeled as distal metastases, hotspots located within one or more segmented organs or a subset of organs (e.g., brain, lung, liver, spleen, kidney) may be labeled as visceral (miMc) distal metastases, and the remaining hotspots located outside the pelvic region are labeled as distant lymphatic metastases (miMa).

[0190] Additionally or alternatively, in some embodiments, a hotspot may be assigned a miTNM class based on a determination that it is located within a specific anatomical region, such as based on a table such as Table 2, wherein each column corresponds to a specific miTNM label (the first row indicates a specific miTNM class) and the second row and below include specific anatomical regions associated with each miTNM class. In some embodiments, a hotspot may be assigned to be located within a specific tissue region listed in Table 2 based on a comparison of the hotspot location with an anatomical segmentation map, thereby allowing for automatic miTNM class assignment. Table 2. List of examples of tissue regions corresponding to the five categories of lesion anatomy labeling methods

[0191] In certain embodiments, hot spots can be further classified in terms of their anatomical location and / or lesion subtype. For example, in certain embodiments, hot spots identified as being located in pelvic lymph nodes (miN) can be identified as belonging to a specific pelvic lymph node subregion, such as the left / right internal iliac, left or right external iliac, left or right common iliac, left or right obturator, presacral region, or other pelvic regions. In certain embodiments, distal lymph node cancer metastasis (miMa) can be classified as retroperitoneal (RP), supradiaphragmatic (SD), or other pelvic outside (OE). Methods for classifying regional (miN) and distant (miMa) lymphoma metastases may include registration of pelvic atlas images and / or identification of various systemic landmarks, which are further described in detail in U.S. application Ser. No. 17 / 959,357, filed on October 4, 2022, entitled "Systems and methods for automatic identification and classification of local lymphatic lesions and distant metastases," published on April 13, 2023 as US2023 / 0115732A1, the contents of which are incorporated herein by reference in their entirety. D. Individual hotspot quantification metrics

[0192] In certain embodiments, detected (e.g., identified and segmented) hotspots may be characterized by various individual hotspot quantification metrics. Specifically, for a particular individual hotspot, the individual hotspot quantification metrics may be used to quantify a measurement of the size (e.g., 3D volume) and / or intensity of the particular hotspot in a manner that indicates the size and / or content of radiopharmaceutical absorption within the (e.g., possible) underlying body lesion represented by the particular hotspot. Thus, the individual hotspot quantification metrics may, for example, convey to a physician or radiologist the likelihood that a hotspot appearing in an image represents a true underlying body lesion and / or convey its likelihood or content of malignancy (e.g., allowing differentiation between benign and malignant lesions).

[0193] In certain embodiments, image segmentation, lesion detection and characterization techniques as described herein are used to determine a corresponding set of hot spots for each of one or more medical images. As described herein, image segmentation techniques can be used to determine a specific 3D volume (3D hot spot volume) for each hot spot detected in a particular image, which represents and / or indicates the volume (e.g., 3D location and extent) of a possible underlying physical lesion within an individual. Each 3D hot spot volume, in turn, includes a set of image voxels, each of which has a specific intensity value.

[0194] Once determined, the set of 3D hotspot volumes can be used to calculate one or more hotspot quantification metrics for each hotspot. Individual hotspot quantification metrics can be calculated according to various methods and formulas described herein, such as those described below. In the following description, the variable L is used to refer to the set of hotspots detected by a particular image, where L = {1, 2, ..., l, ..., N L} represents the number of N detected in the imageL (i.e., N L is the number of hotspots) and the variable l indexes the lth hotspot. As described herein, each hotspot corresponds to a specific 3D hotspot volume within the image, where R l Represents the 3D hotspot volume of the lth hotspot.

[0195] The hotspot quantification metrics may be presented to a user via a graphical user interface (GUI) and / or (e.g., automatically or semi-automatically) generated reports. As described in further detail herein, separate hotspot quantification metrics may include a hotspot intensity metric and a hotspot volume metric (e.g., lesion volume), which quantify the intensity and size of a particular hotspot and / or the underlying lesion it represents, respectively. The hotspot intensity and size may in turn indicate the amount of radiopharmaceutical absorption within an individual and the size of the underlying body lesion, respectively. Hotspot intensity metrics

[0196] In some embodiments, the hotspot quantification metric is or includes an individual hotspot intensity metric that quantifies the intensity of an individual 3D hotspot volume. The hotspot intensity metric may be calculated based on the individual stereo pixel intensities within the identified hotspot volume. For example, for a particular hotspot, the value of the hotspot intensity metric may be calculated based on at least a portion (e.g., a particular subset, such as all) of the stereo pixel intensities of the hotspot. The hotspot intensity metric may include, but is not limited to, metrics such as maximum hotspot intensity, average hotspot intensity, and peak hotspot intensity. Like stereo pixel intensities in nuclear medicine images, in some embodiments, the hotspot intensity metric may represent (e.g., in units) an SUV value.

[0197] In some embodiments, the value of a particular hotspot intensity metric is calculated for an individual hotspot, for example based only on (eg, dependent on) the voxel intensity of the individual hotspot, and not based on the intensities of other image voxels outside the 3D volume of the individual hotspot.

[0198] For example, the hotspot intensity metric may be a maximum hotspot intensity (e.g., SUV) or "SUV max", calculated as the maximum voxel intensity (e.g., SUV or absorption) within the 3D hotspot volume. In some embodiments, the maximum hotspot intensity may be calculated according to the following equations (1a), (1b), or (1c): (1a) (1b) (1c)SUV = maximum (absorbed voxels ∈ lesion volume) Where, in equations (1a) and (1b), l represents a specific (e.g., the lth) hotspot, as described above, and q i is the intensity of voxel i and i∈R l is the specific 3D hotspot volume Rl In equation (1b), SUV i A specific unit for indicating voxel intensity, the Standard Absorption Value (SUV), is described herein.

[0199] In some embodiments, the hotspot intensity metric may be an average hotspot intensity (e.g., SUV) or "SUV average", and may be calculated as the average of all voxel intensities (e.g., SUV or absorption) within the 3D hotspot volume. In some embodiments, the average hotspot intensity may be calculated according to equations (2a), (2b), or (2c) below. (2a) (2b) (2c) where n l is the number of individual voxels within a particular 3D hotspot volume.

[0200] In some embodiments, the hotspot intensity metric can be a peak hotspot intensity (e.g., SUV) or "SUV peak" and can be calculated as the average of the intensities (e.g., SUV or absorption) of stereo pixels where the points are within a (e.g., predetermined) particular distance (e.g., within 5 mm) of the midpoint of the hotspot stereo pixels, where the maximum intensity (e.g., SUV maximum) can be located within the hotspot and can therefore be calculated according to the following equations (3a)-(3c). (3a) (3b) (3c) Absorbed Voxels i Where i: distance (i 最大值 ,i)≤d is the same as the stereo pixel i 最大值 The set of midpoint (hotspot) voxels within distance d, where the voxel is the maximum intensity voxel (e.g., Q 最大值 (l) = q i-最大值 . Lesion Index Metrics

[0201] In certain embodiments, the hotspot intensity metric is a separate lesion index value that maps the intensity of a stereo pixel within a particular 3D hotspot volume to a value on a standardized scale. Such lesion index values ​​are further described in detail in PCT / EP2020 / 050132 filed on January 6, 2020 and PCT / EP2021 / 068337 filed on July 2, 2021, the contents of each of which are incorporated herein by reference in their entirety. The calculation of the lesion index value may include the calculation of a reference intensity value within a specific reference tissue region of, for example, a portion of the aorta (also referred to as the blood pool) and / or the liver.

[0202] For example, in one particular implementation, a first blood pool reference intensity value is determined based on a measurement of intensity (e.g., mean SUV) within an aortic region, and a second liver reference intensity value is determined based on a measurement of intensity (e.g., mean SUV) within a liver region. As further described in detail, for example, in PCT / EP2021 / 068337, filed on July 2, 2021 (the contents of which are incorporated herein by reference in their entirety), calculation of the reference intensity may include, for example, methods of: identifying a reference volume (e.g., the aorta or a portion thereof; e.g., a liver volume) within a functional image such as a PET or SPECT image; washing out and / or dilating certain reference volumes, for example, to avoid including stereo pixels at the edge of the reference volume; and selecting a subset of reference stereo pixel intensities based on a modeling approach, for example, to account for abnormal tissue features within the liver, such as cysts and lesions. In some embodiments, a third reference intensity value may be determined in the form of a multiple (e.g., twice) of the liver reference intensity value or based on the intensity of another reference tissue region, such as the parotid gland.

[0203] In some embodiments, the hot spot intensity can be compared to one or more reference intensity values ​​to determine the lesion index as a value on a standardized scale, which facilitates comparison among different images. Figure 4C A method for assigning lesion index values ​​to hot spots ranging from 0 to 3 is shown. Figure 4C In the method shown in , the blood pool (aorta) intensity value is assigned a lesion index of 1, the liver intensity value is assigned a lesion index of 2, and twice the liver intensity value is assigned a lesion index of 3. The lesion index of a particular hotspot can be determined by first calculating the value of an initial hotspot intensity metric for the particular hotspot, such as the average hotspot intensity (e.g., Q 平均值 (l) or SUV 平均值 ), and compare the value of the initial hotspot intensity metric with the reference intensity value. For example, the value of the initial hotspot intensity metric can be in four ranges, [0, SUV 血液 ]、(SUV 血液 , SUV 肝脏 ]、(SUV 肝脏 , 2×SUV肝脏 ] and greater than 2×SUV 肝脏 (For example (2×SUV 肝脏 ,∞)). The lesion index value of a particular hotspot can then be calculated based on: (i) the value of the initial hotspot intensity metric and (ii) a linear interpolation of the specific range into which the value of the initial hotspot intensity metric falls, such as Figure 4C , where the filled and open points on the horizontal (SUV) and vertical (LI) axes show example values ​​of the initial hotspot intensity metric and the resulting lesion index value, respectively. In certain embodiments, if the SUV reference for the liver or aorta cannot be calculated, or if the aorta value is higher than the liver value, then the lesion index will not be calculated and will be displayed as '-'.

[0204] According to the above description and Figure 4C The mapping scheme shown in can be used to calculate the lesion index value as shown in the following equation (4). (4) where f1, f2 and f3 are linear interpolations between the respective spans in equation (4). Hotspot / lesion volume

[0205] In some embodiments, the hotspot quantification metric may be a volume metric, such as the lesion volume Q 体积 , which provides a measure of the size (eg, volume) of the underlying body lesion represented by the hotspot. In certain embodiments, the lesion volume may be calculated as shown in equations (5a) and (5b) below. (5a) (5b)Q 体积 (l) = v × n l In equation (5a), v i is the volume of the ith voxel, and equation (5b) assumes a uniform voxel volume v, and as before n l is a number of voxels in a particular hotspot volume l. In some embodiments, the voxel volume is calculated as v=δx×δy×δz, where δx, δy, and δz are grid spacings in x, y, and z (e.g., in millimeters, mm). In some embodiments, the lesion volume has units of milliliters (ml). E. Aggregate Hot Spot Metrics

[0206] In certain embodiments, the systems and methods described herein calculate patient index values ​​that quantify disease load and / or risk for a particular individual. Values ​​of various patient indexes can be calculated using (e.g., based on) individual hotspot quantification measurements. In particular, in certain embodiments, a specific patient index value aggregates the values ​​of multiple individual hotspot quantification metrics calculated for an entire set of hotspots detected for a patient and / or for a specific subset of hotspots associated with, for example, a specific tissue region and / or lesion subtype. In certain embodiments, a specific patient index is associated with one or more specific individual hotspot quantification measurements and is calculated using the values ​​of (e.g., multiple) specific individual hotspot quantification metrics calculated for each of at least a portion of the individual 3D hotspot volumes in the set. Overall patient indicators

[0207] For example, in some embodiments, a specific patient metric can be an overall patient metric that aggregates one or more specific individual hotspot quantitative measurements calculated across substantially the entire set of 3D hotspot volumes detected for a patient at a specific point in time to, for example, provide an overall measure of the individual's total disease burden at the specific point in time.

[0208] In some embodiments, a specific patient index may be associated with a single specific individual hotspot quantitative measurement and may be calculated based on substantially all values ​​of the specific individual hotspot quantitative measurement for a collection of 3D hotspot volumes. Such a patient index may be considered to have a functional form, (6) P p,m =f (p) (Q (m),L ) Where Q (m) represents a specific individual hot spot quantitative metric, such as the Q described above 最大值 , Q 平均值 , Q 峰值 , Q 体积 , Q LI , and Q (m),L is the set of values ​​of a specific individual hotspot quantization metric computed for each hotspot l in the set of hotspots L. That is, Q (m),L is a set {Q (m) (l=1),Q (m) (l=2),…,Q (m) (lN L )}.

[0209] Function f (p) can be a variety of functions that appropriately aggregate (combine) the overall value set Q of the values ​​of the specific individual hotspot quantitative metrics (m) For example, the function f (p) It can be the sum, mean, median, mode, maximum, etc. Different specific functions can be used for f(p) , which depends on the aggregated specific hotspot quantization metric Q (m) Thus, the various individual hotspot quantification measurements (e.g., mean intensity, median intensity, mode intensity, peak intensity, individual lesion indices, volumes) may be combined in a variety of ways, such as by taking an overall sum, mean, median, mode, etc., among substantially all values ​​calculated for a 3D hotspot volume set.

[0210] For example, in some embodiments, the overall patient index may be an overall intensity maximum value, which is calculated as the maximum of all individual hotspot maximum intensity values, as shown in equation (7a) or (7b) below: (7a) (7b) Where Q max (1) may be calculated according to equation (1a) above, generally or according to equation (1b) or (1c), where the image intensity represents the SUV value as reflected in equation (7b), for example.

[0211] In certain embodiments, a particular patient index value may be calculated as a combination of substantially all individual hotspot average intensity values, such as as a sum of the average intensity values, such as shown in equations (8a) and (8b) below. (8a) (8b)

[0212] In certain embodiments, the overall patient indicator is a total lesion volume calculated, for example, as the sum of all individual hotspot volumes, thereby providing a measure of the total lesion volume. The total lesion volume may be calculated, for example, as shown in equations (9a) and / or (9b) below, (9a) (9b) where (9b) assumes uniform voxel size, i.e., each voxel has the same volume, v i =v.

[0213] In some embodiments, an overall patient metric may be calculated (e.g., directly) as a function of the intensity, volume, and / or number of voxels within the entire set of hotspots (e.g., as a function of all hotspot voxels within the union of all 3D hotspot volumes; e.g., not necessarily as a function of individual hotspot quantization metrics). For example, in some embodiments, the patient metric may be an overall average, and may be calculated, e.g., as shown in equations (10a) and (10b) below (i.e., by summing the intensities of all individual hotspot voxels for the entire set of hotspots L, and dividing by the total number of hotspot voxels (for the entire set L)): (10a) (10b)

[0214] In certain embodiments, a specific patient index may be calculated using two or more specific individual hot spot quantitative measurements, e.g. (11) P p,m =f (p) (Q (m1),L ,Q (m2),L …)

[0215] For example, both the measure of hotspot intensity and the measure of hotspot volume can be used to calculate an intensity-weighted measure of volume. For example, the total intensity-weighted volume can be calculated at the patient level by calculating, for each hotspot, the product of the lesion index calculated for the individual hotspot and the volume of the hotspot. The sum of substantially all of the intensity-weighted volumes can then be calculated to determine a total score according to, for example, the following equation, where Q LI (l) and Q 体积 (l) are the values ​​of individual lesion index and volume of the i-th 3D hotspot volume, respectively. (12)

[0216] For example, as described above, other measures of intensity may be used to weight the hotspot volume or calculate other metrics. In some embodiments, additionally or alternatively, a patient index may be determined by multiplying the total lesion volume (e.g., as calculated in equation (9a) or (9b)) by the total SUV average (e.g., as calculated in equation (10a) or (10b)) to provide an assessment that also combines intensity with volume.

[0217] In some embodiments, the patient index is or includes a total lesion count, which is a total count of substantially all hot spots detected (eg, N L ). Patient indicators for regional and lesion subtype stratification

[0218] In certain embodiments, additionally or alternatively, multiple values ​​of a specific patient indicator may be calculated, each value being associated with and calculated for a specific subset of the 3D hotspot volume (eg, relative to a set L of substantially all hotspots).

[0219] Specifically, in certain embodiments, the 3D hotspot volumes within the set may be assigned to one or more subsets based on, for example, the specific tissue region in which they are located or a subtype based on a classification scheme such as the miTNM classification. Methods for grouping hotspots based on tissue region and / or based on an anatomical classification such as miTNM are further described in detail in PCT / EP2020 / 050132 filed on January 6, 2020 and PCT / EP2021 / 068337 filed on July 2, 2021, the contents of each of which are hereby incorporated by reference in their entirety.

[0220] In this way, values ​​of patient indicators as described herein may be calculated for one or more specific tissue regions, such as bone regions, prostate regions, or lymphatic regions. In certain embodiments, lymphatic regions may be further graded in a fine-grained manner, for example using methods as described in PCT / EP22 / 77505, filed October 4, 2022 (published as WO2023 / 057411 on April 13, 2023), the contents of which are incorporated herein by reference in their entirety. Additionally or alternatively, in certain embodiments, each 3D hotspot volume may be assigned a specific miTNM subtype and grouped into subsets according to a miTNM classification, and values ​​of various patient indicators may be calculated for each miTNM classification.

[0221] For example, where the miTNM staging system, miTNM class-specific pattern of the overall patient index described above, assigns hotspots to specific lesion subtypes. For example, in some embodiments, the hotspots can be identified (e.g., automatically based on their location) as local tumors (T), nodes in the pelvis (N), or distant metastases (M), and labels (e.g., miT, miN, and miM) are assigned to identify the three subsets. In some embodiments, distant metastases can be further subdivided based on whether the lesion appears in a distant lymph node region (a), bone (b), or other site (e.g., another organ (c)) (e.g., determined by the hotspot location). Therefore, the hotspot can be assigned to one of five lesion (e.g., miTNM) categories (e.g., miT, miN, miMa, miMb, miMc). Therefore, each hotspot can be assigned to a specific subset S, so that, for example, the value of the patient index P(S) can be calculated for each subset S of hotspots in the image. For example, the patient index value for a specific subset of hotspots can be calculated using the following equations (13a-d). (13a) (13b) (13c) (13d) Where S represents a specific subset of hotspots, such as local tumors (e.g., miT), nodes in the pelvis (e.g., labeled miN), distant metastases (e.g., labeled miM), or specific types of distant metastases, such as distant lymph nodes (e.g., labeled miMa), bones (e.g., labeled miMb), or other sites (e.g., labeled miMc). In each of equations (13a)-(13d), l∈S represents a hotspot within the subset S. Equation (13a) is similar to equation (7a), where Q 最大值,S represents the maximum hotspot intensity of the hotspots in the subset S, and Q 最大值 (1) can be calculated generally according to equation (1a) above or according to equation (1b) or (1c), where the image intensity represents the SUV value. Equation (13b) is similar to equation (10a), where q i The combined hotspot volume representing the intensity of the ith voxel (which can be measured in SUV units) and taking the average value is the union of all hotspot volumes within the subset S. Equation (13c) is similar to equation (9b) and yields the overall lesion volume for the particular subset S. Equation (13d) is similar to equation (12) and provides the overall intensity-weighted lesion volume for the particular subset S.

[0222] In some embodiments, the lesion counts may be calculated for a specific subset S (e.g., N S ) is calculated by basically calculating the number of all detected hotspots within. Proportionally adjusted patient index values

[0223] In certain embodiments, various patient indicator values ​​may be proportionally adjusted, for example, based on physical characteristics of the individual (e.g., weight, height, BMI, etc.) and / or the volume of a tissue region (e.g., total bone region volume, prostate volume, total lymphatic volume, etc.) determined by analyzing an image (e.g., a 3D anatomical image) of the individual. Reporting patient index values

[0224] Steering Figure 4A , patient indicator values ​​calculated as described herein may be displayed (e.g., in the form of a chart, graph, table, etc.) in a report (e.g., an automatically generated report), such as an electronic file or part of a graphical user interface, such as for user review and verification / checkout.

[0225] In addition, if Figure 4AAs shown in , a report 400 generated as described herein may include an overview 402 of patient index values ​​that quantifies the disease burden in a patient, such as grouping hotspot subsets according to lesion type (e.g., miTNM classification) and displaying one or more calculated patient index values ​​for each subtype for each lesion type. For example, based on the miTNM staging system, the overview portion 402 of the report 400 displays patient index values ​​for five hotspot subsets, namely, labeled miT, miN, miMa (lymph), miMb (bone), and miMc (other). For each lesion subtype, the overview table 402 displays a number of detected hotspots belonging to the subtype (e.g., within a particular subset), the maximum SUV (SUV 最大值 ), mean SUV (SUV mean), total volume, and a quantity called "aPSMA score". For each lesion subtype S, the values ​​of SUV maximum, SUV mean, total volume, and aPSMA score can be calculated as described above, for example, according to equations (13a), (13b), (13c), and (13d), respectively. Figure 4A In the present invention, the term "aPSMA score" is used to reflect the use of PSMA-binding agents such as [18F]DCFPyL for imaging.

[0226] For each lesion type, Figure 4AThe overview table 402 also includes alphanumeric codes characterizing the severity, number, and location of lesions in different areas (e.g., miTx, miN1a, miM0a, miM1b, miM0c shown from top to bottom), according to the systemic miTNM staging system described in: Siefert et al., "Second Version of the Prostate Cancer Molecular Imaging Standardized Evaluation Framework Including Response Evaluation for Clinical Trials (PROMISE V2)", Eur Urol. May 2023;83(5):405-412.doi:10.1016 / j.eururo.2023.02.002. The notation miTx for miT (localized tumor) subtypes uses "x" as a placeholder for various alphanumeric codes used in the miTNM system to indicate, for example, whether a localized tumor is unifocal or multifocal, whether it is organ-confined or has invasion of structures (e.g., seminal vesicles) or other adjacent structures (e.g., external sphincter, rectum, bladder, levator muscles, pelvic wall), and whether it represents a local recurrence following radical prostatectomy. In certain embodiments, such fine-grained information may not be calculated, e.g., due to specific imaging parameters and / or specific anatomical structures of the segmentation. In certain embodiments, additional fine-grained numeric codes (e.g., miT2, miT3, miT4) and alphanumeric codes (e.g., miT2u, miT2m, miT3a, miT3b, miT4, miTr) may be calculated (e.g., automatically based on automated anatomical segmentation) and displayed. In some embodiments, such codes are computable, but are not displayed in a report such as 400 (e.g., intentionally) for simplicity / readability of the report (e.g., to avoid overburdening a physician or radiologist). In situations where the level of detail in the information displayed in a high-level report, such as detailed miTNM (or other staging system) code information, may be limited (e.g., intentionally), the systems and methods described herein may include features for providing additional detail. For example, when a report such as report 400 is provided through a graphical user interface, a user may be provided with the option to view additional code information, such as by clicking (or touching, such as in a touch screen device) on a portion of report 400 or hovering a mouse over the portion.For example, a single click or touch interaction may be used to expand the overview table 402, allowing a larger view in which additional code information may be presented, or clicking on a specific code such as "miTx" may be used to generate (e.g., via a pop-up) additional information.

[0227] The generated report, such as report 400, may also include information such as reference values ​​(e.g., SUV uptake) 404 determined for various reference organs, such as the blood pool (e.g., calculated from the aortic region or portion thereof) and the liver, which quantifies physiological uptake within the patient; a disease stage code 406, such as an alphanumeric code based on the miTNM scheme or other scheme. In some embodiments, the disease stage indication 406 includes an indication of the specific staging criteria used. For example, if Figure 4A As shown in , disease stage representation 406 includes the text "miTNM" to indicate use of the miTNM staging criteria, as well as a specific code determined by analysis of the specific scan on which report 400 is based.

[0228] Additionally or alternatively, the report may include a hotspot table 410 that provides a list of the individual hotspots identified, along with information for each hotspot, such as lesion subtype, lesion location (e.g., the specific tissue volume in which the lesion is located), and values ​​for various individual hotspot quantification metrics as described herein.

[0229] Therefore, if Figure 4A The reports shown in can be generated from a single imaging session (eg, functional and anatomical images, such as PET / CT or SPECT / CT images), and are used to provide a snapshot of the patient's disease at a specific time.

[0230] In certain embodiments, as described in further detail herein, multiple images acquired over time can be used to track the evolution of a disease over time. Such information can also be included in a report or portion thereof, e.g. Figure 4B as shown in . F. Lesion Tracking in Medical Images

[0231] In certain embodiments, the image analysis and decision support tools of the present disclosure provide, among other things, systems and methods for tracking lesions and assessing disease progression and / or treatment response of patients through analysis of nuclear medicine images. Specifically, in certain embodiments, the methods described herein can be used to analyze longitudinal image data, i.e., a series of medical images (e.g., two or more images) collected over time.

[0232] The lesion tracking techniques described herein can be used in conjunction with a variety of medical image types and / or imaging modalities. For example, a medical image can be or include an anatomical image. An anatomical image conveys anatomical information about structures / morphologies within an individual's body and is obtained using anatomical imaging modalities such as CT, MRI, ultrasound, etc.

[0233] Although described herein particularly with respect to tracking lesions in a time series of medical images, the lesion tracking methods of the present disclosure may additionally or alternatively be used to identify lesion correspondence between medical images (e.g., of the same individual) acquired using different contrast imaging agents (e.g., different radiopharmaceuticals), their doses, image reconstruction techniques, acquisition devices (e.g., different cameras), combinations thereof, and the like.

[0234] Steering Figure 5 In certain embodiments, the methods herein may be used when a patient undergoes an initial, baseline scan and subsequently (eg, at a later time) undergoes a follow-up scan, for example, to assess response to treatment and / or to track the patient's disease.

[0235] In some embodiments, the medical images analyzed by the methods described herein are or include nuclear medicine images, such as three-dimensional (3D) images, such as bone scan (scintigraphy) images, PET images, and / or SPECT images. In some embodiments, the nuclear medicine images are supplemented (e.g., overlaid) with anatomical images, such as computed tomography (CT) images, X-rays, or MRIs.

[0236] Following an initial baseline scan of the patient, a medical image 502, such as a PET / CT image resulting from the scan, is obtained and analyzed to detect hot spots and segment them 504 to identify image regions that indicate potential cancerous lesions in the individual, such as those described herein (e.g., in portions B and C).

[0237] The identified hot spots may be analyzed, for example, to calculate various individual hot spot quantification metrics and / or patient index metrics 506 as described herein. As described herein, hot spot quantification metrics may, for example, include intensity measurements (e.g., peak, mean, median, etc., intensity within a particular hot spot), size measurements (e.g., hot spot volume), and lesion index values ​​that combine both size and intensity, for example, to derive an overall severity of a particular underlying lesion. In certain embodiments, the intensity of one or more reference organs, such as the liver, aorta, parotid gland, may be used to scale the hot spot intensity, allowing the lesion index value to be calculated on a standardized scale.

[0238] Individual hotspot quantification metrics may be combined / aggregated to provide an overall risk / disease severity profile for the patient as a whole and / or for specific anatomical regions (e.g., prostate, bone burden, lymphatic) and / or tumor classifications (e.g., various categories of lesions according to the miTNM classification or other schemes). For example, the volumes of hotspots may be summed and / or otherwise aggregated across the entire patient (e.g., or a selected region) to calculate a total lesion volume for a particular patient.

[0239] For example, the values ​​of the hotspot quantification metric and / or the patient level risk metric (patient index) may be used to provide an initial assessment of the patient, and / or may be stored and / or provided for further processing.

[0240] Turn again Figure 5 , after a period of time (e.g., after a treatment period), one or more subsequent images (Time 2 images) 522 are obtained, hot spots are identified 524, and quantification / risk metrics are calculated 526 as discussed above. Changes in one or more metrics between the initial image and the Time 2 image are calculated. For example, (i) changes in the number of identified lesions may be identified (automatically and / or semi-automatically) and / or (ii) changes in the overall volume of identified lesions (e.g., changes in the sum of the volumes of each identified lesion) may be calculated (automatically and / or semi-automatically), and / or (iii) changes in the total volume weighted by the PSMA (e.g., lesion index) (e.g., the sum of the product of the lesion index and the lesion volume for all lesions in the region of interest) may also or alternatively be automatically determined to indicate changes. Similarly, other subsequent images may be obtained at later time points (e.g., Time 3, Time 4, etc.) and analyzed accordingly. This longitudinal data set of lesion tracking may be used by a medical provider, for example, to determine treatment effectiveness.

[0241] For example, in certain embodiments, a heat map is retained with the patient record, and each subsequent map is compared to a baseline map (or a previous subsequent map) to identify corresponding (same) lesions, e.g., to identify which lesions are new and / or to generate per-lesion longitudinal data, thereby allowing tracking of volume, intensity, lesion index score or other parameters for each lesion. Thus, the methods described herein provide for semi-automatic and / or automatic analysis of medical image data acquired over time to generate a longitudinal data set that provides for the evolution of a patient's risk and / or disease over time during monitoring and / or treatment response.

[0242] In certain embodiments, the methods described herein provide for the calculation of metrics that can be used to classify patient disease for treatment / decision-making purposes and / or to stratify cohorts for clinical trial data collection and analysis. For example, in certain embodiments, changes in one or more metrics can be used to classify a patient as belonging to one of the following three categories: (i) response / partial response, characterized by a decrease in PSMA-volume greater than or equal to 30% and a decrease in the number of lesions, such as Fig. 6A Stable disease, characterized by a greater than 30% decrease in PSMA-volume, but with the appearance of new lesions ( Figure 6B ); and (ii) progressive disease, characterized by a 20% or greater increase in PSMA-volume and the appearance of one or more new lesions, such as those classified according to RECIP ( Figure 6C ). Registering multiple medical images

[0243] Steering Figure 7 In some embodiments, two or more different medical images may be obtained 702, for example, from the same individual at different points in time (e.g., a time series). Each particular medical image may have a particular heat map associated therewith that identifies one or more hot spots within the particular medical image. In some embodiments, the medical images and associated heat maps may be analyzed to identify corresponding hot spots in two or more medical images that are determined to represent the same potential lesion. In this way, the presence (e.g., appearance and / or disappearance) and / or characteristics of a lesion, such as size / volume, radiopharmaceutical absorption, etc., may be compared between multiple different medical images.

[0244] In some embodiments, the plurality of medical images may be or include a time series of medical images obtained for the same particular individual, each medical image, for example, having been obtained at a different time. Additionally or alternatively, the plurality of medical images may include medical images obtained using different imaging agents (e.g., different radiopharmaceuticals), their doses, image reconstruction techniques, acquisition devices (e.g., different cameras), combinations thereof, and the like.

[0245] In certain embodiments, a plurality of heat maps 704 may be obtained. Each heat map is associated with a particular medical image and identifies one or more hot spots therein. A hot spot is a region of interest (ROI) identified within a particular medical image and / or sub-images thereof (e.g., in the case of a composite image) as representing a potential physical lesion within an individual. A heat map may identify a hot spot volume (e.g., a 3D volume) that has been determined, for example, by segmentation of a 3D image.

[0246] In certain embodiments, hot spots are identified and / or segmented within the 3D functional image, for example, as localized areas of higher intensity.

[0247] In some embodiments, the heat map can be generated by manual and / or automatic detection and / or segmentation or a combination thereof. The manual and / or semi-automatic methods can include, for example, receiving user input through an image analysis graphical user interface (GUI). A user can review the presentation of one or more medical images and / or sub-images thereof, with or without various computer-generated annotations, such as organ segmentation in combination with the display, and perform operations, such as selecting areas to include and / or exclude heat maps. In some embodiments, automatic hot spot identification and segmentation are performed prior to user review to generate a preliminary heat map, which is then reviewed by the user, for example, to generate a final heat map.

[0248] In certain embodiments, hotspots are classified (e.g., assigned a label) as belonging to specific anatomical regions (e.g., bone, lymphatic, pelvic, prostate, visceral (e.g., soft tissue organs (other than prostate, lymph), such as liver, kidney, spleen, lung, and brain)) and / or lesion types, such as those of the miTNM classification scheme.

[0249] In certain embodiments, each medical image is segmented to identify a set of organ regions therein and generate a corresponding anatomical segmentation map 706. Within a particular medical image, the anatomical segmentation map identifies a set of organ regions, each member of the set corresponding to a particular organ, including various soft tissue and / or bone regions. As described herein, anatomical segmentation can be performed using a machine learning module. The machine learning module can receive an anatomical image as input and analyze the anatomical image to generate an anatomical segmentation map.

[0250] In some embodiments, the anatomical segmentation maps determined from each medical image can be used to perform image registration. Specifically, at least a portion of the identified set of organ regions (e.g., including regions corresponding to one or more of the cervical spine; thoracic spine; lumbar spine; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and lower bones) can be used to determine one or more registration fields for co-registering two or more anatomical segmentation maps. Once determined, the one or more registration fields can be used to co-register the medical images from which the anatomical segmentation maps are determined and / or their corresponding heat maps 708.

[0251] For example, turning Figure 8 , the method can be used to co-register the first medical image and the second medical image and / or their corresponding heat maps. In process 800, the first medical image and the second medical image are composite images, each containing an anatomical and a functional image pair (802a / 802b and 804a / 804b).

[0252] The first heat map 814 identifies a first set of hot spots within the first medical image and may be generated by and / or by detecting and / or segmenting the hot spots 812 within the first functional image 802b. The second heat map 824 identifies a second set of hot spots within the second medical image and may be generated by and / or by detecting and / or segmenting the hot spots 822 within the second functional image 804b.

[0253] The first anatomical image 802a may be segmented, for example, using a machine learning module (anatomical segmentation module) to determine a first anatomical segmentation map 834 (832) that identifies a set of one or more organ regions within the first medical image (i.e., within the first anatomical image and / or the first functional image). The second anatomical image 804a may be segmented, for example, using the anatomical segmentation module to determine a second anatomical segmentation map 844 (842) that identifies a set of one or more organ regions within the second medical image (i.e., within the second anatomical image and / or the second functional image). Full-field image registration

[0254] In some embodiments, the first anatomical segmented map 834 and the second anatomical segmented map 844 are used to determine one or more registration fields. The registration fields can be calculated based on (e.g., performing) an affine transformation. For example, in some embodiments, one or more specific subsets of the identified set of organ regions are used as markers for registering the first anatomical segmented map and the second anatomical segmented map. Specifically, each specific subset of the identified organ regions can be used to determine a corresponding registration field that aligns a specific subset within the first anatomical segmented map with the same specific subset within the second anatomical segmented map. This procedure can be performed for multiple subsets of the identified organ regions to determine multiple registration fields 850, which can then be incorporated to produce a final overall registration field for performing final image registration.

[0255] For example, each subset may include organ regions corresponding to locations within a particular anatomical region or part of an individual's body. Fig. 9A and 9B As shown in FIG. 1 , a first left pelvic region, a registration field can be determined using a subset of organ regions corresponding to the pelvic bones on the left side of the individual ( Fig. 9A ); and a second right pelvic region, determining the registration field using a subset of organ regions corresponding to the pelvic bones on the right side of the individual ( Fig. 9B ).like Fig. 9CAs shown in , these two (left and right pelvic region) registration fields can be combined, for example, by a distance-weighted voxel-by-voxel average, whereby each stereo pixel of the final registration field is calculated as a weighted average of the stereo pixel values ​​in the left and right pelvic region registration fields. For each stereo pixel, the weights of the left and right stereo pixel values ​​for averaging can be determined based on the identified distances of the stereo pixel to the left and right pelvic bones, respectively. An example of this registration method is further described in detail in PCT / EP22 / 77505 filed on October 4, 2022 (published as WO2023 / 057411 on April 13, 2023) with respect to the portion of the image located around the pelvic region. This method can be extended to multiple organ region subsets throughout an individual's body (e.g., organ subsets associated with specific parts of the body, such as the head, neck, chest, abdomen, pelvic region, left side, right side, front, back, etc. and combinations thereof (e.g., left pelvic region, right pelvic region, right chest, right chest, etc.) to determine multiple local registration fields, each using a specific organ region subset as a landmark, which are then merged (e.g., by distance weighted averaging) to produce a final overall registration field.

[0256] like Fig.10 As shown in , this method can be used to perform accurate whole-body image registration. For example, Fig.10 A first PET / CT composite image obtained by a first scan and a second PET / CT composite image as originally obtained by a second scan are shown (top row). Each CT scan shows the identified organ regions (colored portions) of the overlaid anatomical segmentation maps. Fig.10 The bottom row of shows again the first PET / CT image and a transformed version of the second PET / CT image, which is now registered with the first image by a weighted piecewise affine registration method as described herein.

[0257] Fig.11A A schematic diagram showing a second image registered with a first image, illustrating the change in voxels. Fig. 11B A schematic diagram of a registration field is shown, which includes vectors of a subset of voxels. Fig. 11B As shown in FIG. 1 , in some embodiments, the registration field includes a reference to a location (e.g., a stereo pixel) in the first image relative to a corresponding point (e.g., a stereo pixel) in the second image (the target stereo pixel in the second image is at Fig. 11BIn some embodiments, an inverted registration field may be determined. The inverted registration field contains a reference of locations (e.g., stereo pixels) in the second image relative to locations (e.g., stereo pixels) in the first image. In some embodiments, an inverted reference field is first generated for each of the affine registrations. The inverted fields may then be weighted together in the same manner as the affine registrations to generate a full body inverted registration field.

[0258] In certain embodiments, without wishing to be bound by any particular theory, the first scan resides in one space (e.g., in world coordinates) and the second scan resides in another space. A registration field from the first image space to the second image space is generated by finding a registration that best aligns the organ segments from the second scan with the organ segments in the first scan (e.g., by finding a local optimum in an optimization problem). The registration field can then be applied to any image (e.g., PET, CT, organ segmentation, heat map) that resides in the same space as the second scan to register it with the space of the first scan. Point-by-point registration

[0259] Additionally or alternatively, in some embodiments, the methods described herein can be used to generate a point-by-point registration 850. In some embodiments, the point-by-point registration can be used to triangulate, for example, between two PET / CT image stacks acquired at two different time points. In some embodiments, as described herein, the point-by-point registration method uses an "anchor point," which is a single point correspondence, for example relative to a corresponding mask that identifies corresponding 3D tissue regions (e.g., skeletal bones) as described above.

[0260] In some embodiments, a point-by-point registration method utilizes an anatomical segmentation map determined for two different images, such as PET / CT images of the same patient acquired at two different time points, to identify a set of anchor points. For example, the set of anchor points may be or include the following points: the center of mass of all left ribs, the center of mass of all right ribs, the center of mass of the left hip bone, the center of mass of the right hip bone, and the center of mass of the thoracic spine. For a particular medical image, for example, an anatomical segmentation map acquired at a particular time point may be used to determine the coordinates of each anchor point in a particular set of anchor points. The anchor point coordinates may be determined accordingly for each of a plurality of medical images, such as in a time series of medical images.

[0261] In some embodiments, the point-by-point registration method determines a transformation operation, such as a translation, that matches corresponding anchor points between two images. For example, in some embodiments, the set of anchor points may include N anchor points. Coordinate values ​​(e.g., (x, y, z) coordinates in three dimensions) may be calculated for each of the N anchor points in the first image and the second image to be registered with each other. For each anchor point i, in the set, a separate anchor point translation may be determined that matches its position in the first image with its position in the second image. The individual anchor point translations can then be used to determine a weighted translation for a particular point in the first image. The weighted translation aligns it with or identifies a corresponding point in the second image (eg, representing the same potential body location).

[0262] For example, for a particular selected point and a set of N anchor points, the weighted translation It can be determined based on a weighted sum of the (inverse) distances of the individual anchor point translations, where each anchor point translation is weighted (e.g., multiplied) by the inverse of its distance from a particular selected point. This particular point-by-point registration method can be expressed, for example, according to the following equation (14): (14) Where D i is the distance from a specific selected point to the i-th anchor point, is the translation that matches the coordinate values ​​of the i-th anchor point in the two images. Therefore, is a weighted translation computed for a specific (selected) point based on all distances from the anchor point.

[0263] Turn again Figure 7 and 8 The registration field and / or point-by-point registration 850 determined as described herein may be used to transform the second heat map and / or the first heat map, 824 and / or 814, respectively, so that they are registered 708, 852 with each other. In this way, a set of hotspots identified within different (e.g., first and second) medical images may be aligned, thereby allowing accurate identification of corresponding hotspots 710, 854 that represent the same body lesion.

[0264] In certain embodiments, additionally or alternatively, a registration field and / or a point-by-point registration may be determined as described herein and used to register the second medical image with the first medical image (e.g., collected at an earlier time), e.g., prior to generating the second heat map. The registered version of the second medical image may be used to generate a second heat map, which is generated from the registered version of the second medical image to be registered with the first heat map generated from the first medical image. Identify the corresponding hot spots

[0265] Steering Fig.12In one embodiment, corresponding hot spots may be identified by computing one or more lesion correspondence metrics, which, for example, quantify the proximity and / or similarity between two or more hot spots identified in different medical images. Example metrics include, but are not limited to, the following:

[0266] Hotspot Overlap : In some embodiments, hot spots in the first and second images that overlap (and subsequently register) can be identified as corresponding hot spots for inclusion in lesion correspondence. In some embodiments, the relative fraction (percentage) of volumetric overlap can be calculated and compared to one or more overlap thresholds. Hot spots with an overlap fraction above a particular threshold (e.g., 20 percent or more, 30 percent or more, 40 percent or more, 50 percent or more, 70 percent or more) can be identified as lesion correspondence, e.g. Fig.12 As shown in Group A.

[0267] Hotspot distance : In some embodiments, for example Fig.12 As shown in Group B of , the hotspot distance can be calculated as, for example, the distance between two points, such as the center of mass (COM) of each hotspot. Hotspot pairs separated by a hotspot distance less than a specific distance threshold (e.g., 10 mm or less, 20 mm or less, 30 mm or less, 40 mm or less, 50 mm or less, etc.) can be identified as belonging to lesion correspondence. In some embodiments, multiple distance thresholds are used, such as for different regions. For example, in some embodiments, a larger threshold (e.g., 50 mm) is used for the rib / chest region to account for respiratory motion and a smaller distance threshold (e.g., 10 mm, 20 mm, etc.) is used elsewhere.

[0268] Type / position matching : In some embodiments, each hotspot may be assigned a lesion classification (e.g., miTNM classification) and / or location (e.g., pelvic, bone, lymphatic). In some embodiments, a hotspot may be required to have a matching lesion classification and / or assigned location to be identified as a corresponding hotspot in a lesion correspondence.

[0269] In this way, hot spots appearing in different images can be matched 854 to each other and identified as representing the same underlying body lesion. The correspondence between such matching hot spots can be determined by identifying lesion correspondence codes of corresponding hot spots in two or more different medical images (e.g., the first and second images). Lesion correspondence can be bidirectional. Lesion tracking metrics

[0270] In certain embodiments, the systems and methods described herein provide for the calculation of metrics 712 that can be used to classify patient disease for treatment / decision making purposes and / or to stratify cohorts for clinical trial data collection and analysis 714. As described herein, such metrics can include total lesion volume, e.g., calculated as the sum of hotspot volume and / or change thereof for the entire individual, and a plurality of newly identified lesions and / or their absence (or reduction in the total number of lesions), as well as other metrics, such as the various hotspot quantifications and / or patient indicators / metrics described herein, e.g., in Sections D and E. In certain embodiments, these metrics can be presented in a report, e.g., in a tabular format or as a series of graphs or as tracks in a graph, e.g. Figure 4B In certain embodiments, values ​​for normal (non-cancerous) physiological absorption may also be displayed, such as Figure 4B as shown in .

[0271] In certain embodiments, the methods described herein for identifying corresponding hot spots can be used to match other target regions (e.g., corresponding to other physical features of an individual) identified within different images (e.g., collected at different times, from different individuals, with different tracers, etc.). These methods can be used to align and identify corresponding target regions identified within different images to assess the presence, progression, status, response to treatment, etc. of a variety of conditions that are not necessarily limited to cancer (e.g., muscle, ligament, tendon injuries; aneurysm diagnosis; assessing cognitive activity (e.g., via fMRI), etc.). G. Provide information for clinical decision making and treatment evaluation

[0272] In certain embodiments, metrics calculated based on analysis of images as described herein may also be used to determine values ​​for and / or stratify individuals according to various metrics indicating, among other things, disease status, progression, prognosis, an individual's response to therapy, and / or a prediction of an individual's likely response to one or more specific therapies.

[0273] In certain embodiments, these metrics can be alone and / or associated with endpoints, such as clinical endpoints (e.g., which measure how well a patient functions, feels, or survives) and can be used to assess treatment efficacy, such as in the context of population analyses in clinical trials, and can be used alone and / or in combination with other markers, such as prostate-specific antigen (PSA).

[0274] In certain embodiments, endpoints that may be determined and / or associated with patient metrics and / or classifications described herein include, but are not limited to, overall survival (OS), radiographic progression-free survival (rPFS), various symptomatic endpoints (e.g., patient reported outcomes), disease-free survival (DFS), event-free survival (EFS), objective response rate (ORR), complete response (CR) / partial response (PR) / stable disease (SD) / progressive disease (PD), progression-free survival (PFS), time to progression (TTP), and time to radiographic progression.

[0275] In certain embodiments, the various metrics described herein and / or endpoint values ​​determined therefrom can be used to guide treatment decisions. For example, the methods described herein can be used to identify whether an individual is a responder to a particular treatment, providing an opportunity to discontinue an ineffective treatment, adjust the dosage, or switch to a new therapy in advance.

[0276] Thus, the image analysis and decision support tools described herein may be used, among other things, to determine prognostic information, measure response to therapy, stratify patients for radioligand therapy, and / or provide predictive information for other therapies.

[0277] For example, in certain embodiments, metrics computed from images as described herein, such as miTNM classification of individual lesions and / or overall disease stage (e.g., Figure 4A ), expression scores, PRIMARY scores, measurements of tumor volume (e.g., total tumor volume of the patient and / or stratified by lesion category), the presence and / or count of new lesions can be used to calculate a specific response classification. For example, the lesion tracking tool described herein can be used to identify new lesions and quantify the increase in tumor size, and changes in aPSMA scores (e.g., lesion index scores and / or intensity-weighted total volume, as described herein) can also be used to assess prostate cancer progression criteria, such as PSMAPET progression (PPP) scores (see, e.g., Fanti et al., "Proposal of Systemic Therapy Response Assessment Criteria in time of PSMAPET / CT imaging: PSMA PET Progression (PPP)", Journal of Nuclear Medicine (J. Nucl. Med.), 2019 https: / / doi.org / 10.2967 / jnumed.119.233817), RECIP criterion scores, etc.

[0278] In certain embodiments, the quantified values ​​of patient indicators at single and / or multiple time points can be used as inputs to a prognostic model to determine a prognostic metric that indicates and / or quantifies the likelihood of a particular clinical event, disease recurrence, or progression in a patient (e.g., having prostate cancer or at risk for prostate cancer). The prognostic metric can include overall survival (OS), radiographic progression-free survival (rPFS), various symptomatic endpoints (e.g., patient-reported outcomes), disease-free survival (DFS), event-free survival (EFS), objective response rate (ORR), complete response (CR) / partial response (PR) / stable disease (SD) / progressive disease (PD), progression-free survival (PFS), time to progression (TTP), radiographic time to progression.

[0279] The prognostic model can be a statistical model, such as regression, and can include additional clinical variables, i.e., inputs, such as patient physical characteristics, such as race / ethnicity; prostate-specific antigen (PSA) content and / or flow rate; hemoglobin content; lactate dehydrogenase content; albumin content; clinical T stage; biopsy Gleason score (Gleason score); and percentage positive core score (percentage positive core score). In certain embodiments, the prognostic model compares a calculated value (e.g., a patient index) with one or more thresholds to classify the patient and / or place the patient in a "bucket", such as one of a set of OS value ranges, etc. In certain embodiments, the prognostic model can be a machine learning model, for example, various individual hot spot quantitative measures and / or aggregated patient-level indicators can be considered as features input to the machine learning model, which produces a predicted value of one or more prognostic endpoints described herein as an output. Such machine learning models can be, for example, artificial neural networks (ANNs). The machine learning model can also include clinical variables as inputs (i.e., features).

[0280] For example, in certain embodiments, quantitative measures of disease burden from a single time point can be used to calculate values ​​for patient-level metrics such as total tumor volume, overall intensity measures such as total SUV mean / max / peak, aPSMA score (e.g., intensity-weighted total volume). These metrics can be used as inputs to a prognostic model to produce one or more of the following as outputs: expected survival (e.g., in months), time to progression (TTP), and time to radiographic progression.

[0281] In certain embodiments, quantitative data at multiple time points, such as changes in total lesion volume, SUV, aPSMA score, measures of changes in lesions over time (e.g., number of new lesions, number of disappeared lesions, number of tracked lesions) can be used as input to a prognostic model to produce one or more of the following as outputs: expected survival (e.g., in months), time to progression, time to radiographic progression.

[0282] In certain embodiments, additionally or alternatively, features such as PSMA expression in the prostate (and / or other tissue regions, such as may be identified by the anatomical segmentation techniques described herein) may be used as inputs to a prognostic model. For example, spatial intensity patterns (e.g., intensity from functional images such as PET or SPECT images), particularly tissue regions, may be used alone and / or in conjunction with the quantitative metrics and clinical variables described herein as inputs to a machine learning module to generate predictions such as risk of simultaneous (synchronous) cancer metastasis, risk of future (metachronous) cancer metastasis. For example, data from lesion tracking techniques described herein may be used as inputs to improve prediction techniques such as those described in U.S. Pat. No. 11,564,621, the contents of which are hereby incorporated by reference in their entirety. In certain embodiments, the intensity pattern can be used to determine, for example, a score for each image of an individual at a particular time point, such as or similar to the PRIMARY score as described in Seifert et al., "A standardized assessment framework for molecular imaging of prostate cancer including response assessment for use in clinical trials (Provisions V2)", European Urology 2023 May;83(5):405-412.doi:10.1016 / j.eururo.2023.02.002. Such automatically calculated intensity scores can be included in patient reports, such as Figure 4A Those ratings shown in .

[0283] In certain embodiments, the methods described herein can be used to generate a model for classifying patients who respond to therapy. For example, the lesion tracking technology that can be described herein can be used to determine inputs, such as changes in tumor volume, intensity, lesion appearance / disappearance. These inputs can be used by one or more response models to determine whether the patient responds to treatment (e.g., responds / does not respond) and / or the degree (e.g., numerical value) to which the patient responds to treatment. As described herein, this class of methods can make full use of existing response criteria, such as RECIP and PPP, which currently rely on variable and time-consuming manual radiologist assessments, and therefore can be improved by the present invention technology to improve the accuracy of patient staging, stability (e.g., uniformity in different operators, imaging sites, etc.) and speed and the response to therapy assessment.

[0284] In certain embodiments, the methods described herein can be used to assess which patients are likely to experience favorable benefits and / or adverse effects of a particular treatment, which may, for example, be expensive and / or associated with adverse side effects. For example, the software can be used to provide an indication of whether a patient is likely to benefit from a particular radioligand therapy. In this way, the methods described herein can satisfy the needs of radioligand therapies (e.g., Pluvicto TM ) and help physicians choose between the large and growing number of therapies, especially in advanced disease. For example, for a set of possible treatments (e.g., abiraterone, enzalutamide, apalutamide, darutamide, sipraluzumab, Ra223, docetaxel, cabazitaxel, pembrolizumab, olaparib, rucaparib, 177Lu-PSMA617, etc.), a predictive model can accept the various imaging metrics described herein as input and generate as output a score for each treatment (or treatment class, e.g., a specific treatment class, e.g., androgen biosynthesis inhibitors (e.g., abiraterone), androgen receptor inhibitors (e.g., enzalutamide, apalutamide, darutamide), cellular immunotherapy (e.g., sipraluzumab), internal radiation therapy treatment (e.g., Ra223), anti-neoplastic drugs (e.g., docetaxel, cabazitaxel), immune checkpoint inhibitors (pembrolizumab), PARP inhibitors (e.g., olaparib, rucaparib), PSMA binders (e.g., with radioligand therapy, e.g., Lu177) that indicates the likelihood that the patient will respond positively to the treatment. H. Imaging Agents

[0285] As described herein, various radionuclide-labeled PSMA binding agents can be used as radiopharmaceutical imaging agents for nuclear medicine imaging to detect and assess prostate cancer. In certain embodiments, certain radionuclide-labeled PSMA binding agents are suitable for PET imaging, while others are suitable for SPECT imaging. i. PET Imaging of Radionuclide-Labeled PSMA Binding Agents

[0286] In certain embodiments, the radionuclide-labeled PSMA binding agent is a radionuclide-labeled PSMA binding agent suitable for PET imaging.

[0287] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises [18F]DCFPyL (also known as PyL TM ; also known as DCFPyL-18F): or a pharmaceutically acceptable salt thereof.

[0288] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises [18F]DCFBC: or a pharmaceutically acceptable salt thereof.

[0289] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 68 Ga-PSMA-HBED-CC (also known as 68 Ga-PSMA-11): or a pharmaceutically acceptable salt thereof.

[0290] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises PSMA-617: or a pharmaceutically acceptable salt thereof. In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 68 Ga-PSMA-617 (which is used 68 In certain embodiments, the PSMA-binding agent labeled with a radionuclide comprises 177 Lu-PSMA-617 (which is used 177 Lu-labeled PSMA-617) or a pharmaceutically acceptable salt thereof.

[0291] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises PSMA-I&T: or a pharmaceutically acceptable salt thereof. In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 68 Ga-PSMA-I&T (which is 68 Ga-labeled PSMA-I&T) or a pharmaceutically acceptable salt thereof.

[0292] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises PSMA-1007: or a pharmaceutically acceptable salt thereof. In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 18 F-PSMA-1007 (which is used 18 F-labeled PSMA-1007) or a pharmaceutically acceptable salt thereof.

[0293] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 18F-JK-PSMA-7: or a pharmaceutically acceptable salt thereof. ii. SPECT imaging of radionuclide-labeled PSMA binders

[0294] In certain embodiments, the radionuclide-labeled PSMA binding agent is a radionuclide-labeled PSMA binding agent suitable for SPECT imaging.

[0295] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 1404 (also referred to as MIP-1404): or a pharmaceutically acceptable salt thereof.

[0296] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 1405 (also referred to as MIP-1405): or a pharmaceutically acceptable salt thereof.

[0297] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 1427 (also known as MIP-1427): or a pharmaceutically acceptable salt thereof.

[0298] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 1428 (also known as MIP-1428): or a pharmaceutically acceptable salt thereof.

[0299] In certain embodiments, PSMA-binding agents utilize radionuclides by chelating them to a radioisotope of a metal [e.g., a radioisotope of technetium (Tc) (e.g., technetium-99m ( 99m Tc)); for example, radioactive isotopes of rhenium (Re) (e.g., rhenium-188 ( 188 Re); for example, rhenium-186 ( 186 Re)); for example, radioactive isotopes of yttrium (Y) (e.g., 90 Y); for example, radioactive isotopes of lutetium (Lu) (e.g., 177 Lu); for example, radioactive isotopes of gallium (Ga) (such as 68 Ga; e.g. 67 Ga); for example, radioactive isotopes of indium (e.g., 111In); for example, radioactive isotopes of copper (Cu) (e.g., 67 Cu)] to mark.

[0300] In certain embodiments, 1404 is labeled with a radionuclide (e.g., a radioisotope chelated to a metal). In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 99m Tc-MIP-1404, which is labeled with (e.g., chelated to) 99m TC 1404: or a pharmaceutically acceptable salt thereof. In certain embodiments, 1404 can be chelated to other metal radioisotopes [e.g., radioisotopes of rhenium (Re) (e.g., rhenium 188 ( 188 Re); for example, rhenium 186 ( 186 Re)); for example, radioactive isotopes of yttrium (Y) (e.g., 90 Y); for example, radioactive isotopes of lutetium (Lu) (e.g., 177 Lu); for example, radioactive isotopes of gallium (Ga) (such as 68 Ga; e.g. 67 Ga); for example, radioactive isotopes of indium (e.g., 111 In); for example, radioactive isotopes of copper (Cu) (e.g., 67 Cu)] to form a compound having the same 99m A similar structure is shown by Tc-MIP-1404, in which another metal radioisotope is 99m Tc substitution.

[0301] In certain embodiments, 1405 is labeled with a radionuclide (e.g., a radioisotope chelated to a metal). In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 99m Tc-MIP-1405, which is labeled with (e.g., chelated to) 99m TC 1405: or a pharmaceutically acceptable salt thereof. In certain embodiments, 1405 can be chelated to other metal radioisotopes [e.g., radioisotopes of rhenium (Re) (e.g., rhenium 188 ( 188 Re); for example, rhenium 186 ( 186 Re)); for example, radioactive isotopes of yttrium (Y) (e.g., 90 Y); for example, radioactive isotopes of lutetium (Lu) (e.g., 177 Lu); for example, radioactive isotopes of gallium (Ga) (such as 68 Ga; e.g.67 Ga); for example, radioactive isotopes of indium (e.g., 111 In); for example, radioactive isotopes of copper (Cu) (e.g., 67 Cu)] to form a compound having the same 99m The structure of Tc-MIP-1405 is similar to that of the 99m Tc.

[0302] In certain embodiments, 1427 is labeled with (eg, chelated to) a radioactive isotope of a metal to form a compound according to the formula: or a pharmaceutically acceptable salt thereof, wherein M is a radioisotope of a metal labeled with 1427 [e.g., a radioisotope of technetium (Tc) (e.g., technetium 99m ( 99m Tc)); for example, radioactive isotopes of rhenium (Re) (e.g., rhenium 188 ( 188 Re); for example, rhenium 186 ( 186 Re)); for example, radioactive isotopes of yttrium (Y) (e.g., 90 Y); for example, radioactive isotopes of lutetium (Lu) (e.g., 177 Lu); for example, radioactive isotopes of gallium (Ga) (e.g., 68 Ga; e.g. 67 Ga); for example, radioactive isotopes of indium (e.g., 111 In); for example, radioactive isotopes of copper (Cu) (e.g., 67 Cu)].

[0303] In certain embodiments, 1428 is labeled with (eg, chelated to) a radioactive isotope of a metal to form a compound according to the formula: or a pharmaceutically acceptable salt thereof, wherein M is a radioisotope of a metal labeled with 1428 [e.g., a radioisotope of technetium (Tc) (e.g., technetium 99m ( 99m Tc)); for example, radioactive isotopes of rhenium (Re) (e.g., rhenium 188 ( 188 Re); for example, rhenium 186 ( 186 Re)); for example, radioactive isotopes of yttrium (Y) (e.g., 90 Y); for example, radioactive isotopes of lutetium (Lu) (e.g., 177 Lu); for example, radioactive isotopes of gallium (Ga) (e.g., 68 Ga; e.g. 67 Ga); for example, radioactive isotopes of indium (e.g., 111In); for example, radioactive isotopes of copper (Cu) (e.g., 67 Cu)].

[0304] In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises PSMA I&S: or a pharmaceutically acceptable salt thereof. In certain embodiments, the radionuclide-labeled PSMA-binding agent comprises 99m Tc-PSMA I&S (which is used 99m Tc-labeled PSMA I&S) or a pharmaceutically acceptable salt thereof. I. Computer system and network environment

[0305] Certain embodiments described herein utilize computer algorithms in the form of software instructions executed by a computer processor. In certain embodiments, the software instructions include a machine learning module, also referred to herein as artificial intelligence software. As used herein, a machine learning module refers to a computer-implemented process (e.g., software function) that implements one or more specific machine learning techniques, such as an artificial neural network (ANN), such as a convolutional neural network (CNN), such as a recursive neural network, such as a recursive neural network, such as a long short-term memory (LSTM) or a bilateral long short-term memory (Bi-LSTM), a random forest, a decision tree, a support vector machine, etc., to determine one or more output values ​​for a given input.

[0306] In some embodiments, a machine learning module implementing machine learning techniques is trained, for example, using a data set including data categories described herein (e.g., CT images, MRI images, PET images, SPECT images). Such training can be used to determine various parameters of a machine learning algorithm implemented by a machine learning module, such as weights associated with a layer in a neural network. In some embodiments, once a machine learning module is trained, for example, to achieve a specific task (e.g., segmented anatomical regions, segmented and / or classified hot spots, or determining a prognostic value, a therapeutic response value, and / or a predictive metric value), the determined parameter values ​​are fixed and (e.g., unchanged, static) machine learning modules are used to process new data (e.g., different from training data) and to achieve its training tasks without further updating its parameters (e.g., the machine learning module does not receive feedback and / or updates). In some embodiments, a machine learning module may receive feedback, such as based on a user's review of accuracy, and such feedback may be used as additional training data to dynamically update the machine learning module. In some embodiments, two or more machine learning modules may be combined and implemented with a single module and / or a single software application. In some embodiments, two or more machine learning modules may also be implemented separately, such as with a separate software application. The machine learning module can be software and / or hardware. For example, the machine learning module can be implemented entirely in software, or certain functions of the ANN module can be performed by dedicated hardware (e.g., by an application-specific integrated circuit (ASIC)).

[0307] like Fig.13 As shown in FIG. 1 , an implementation of a network environment 1300 for providing the systems, methods, and architectures described herein is shown and described. Fig.13 , a block diagram of an exemplary cloud computing environment 1300 is shown and described. The cloud computing environment 1300 may include one or more resource providers 1302a, 1302b, 1302c (collectively referred to as 1302). Each resource provider 1302 may include computing resources. In some implementations, computing resources may include any hardware and / or software for processing data. For example, computing resources may include hardware and / or software capable of executing algorithms, computer programs, and / or computer applications. In some implementations, exemplary computing resources may include application servers and / or databases with storage and retrieval capabilities. Each resource provider 1302 may be connected to any other resource provider 1302 in the cloud computing environment 1300. In some implementations, the resource provider 1302 may be connected via a computer network 1308. Each resource provider 1302 may be connected to one or more computing devices 1304a, 1304b, 1304c (collectively referred to as 1304) via a computer network 1308.

[0308] The cloud computing environment 1300 may include a resource manager 1306. The resource manager 1306 may be connected to the resource providers 1302 and the computing devices 1304 via a computer network 1308. In some implementations, the resource manager 1306 may facilitate providing computing resources to one or more computing devices 1304 via one or more resource providers 1302. The resource manager 1306 may receive a request for computing resources from a particular computing device 1304. The resource manager 1306 may identify one or more resource providers 1302 that are capable of providing the computing resources requested by the computing device 1304. The resource manager 1306 may select a resource provider 1302 to provide the computing resources. The resource manager 1306 may facilitate a connection between a resource provider 1302 and a particular computing device 1304. In some embodiments, the resource manager 1306 may establish a connection between a particular resource provider 1302 and a particular computing device 1304. In some embodiments, the resource manager 1306 may redirect the specific computing device 1304 to the specific resource provider 1302 having the requested computing resources.

[0309] Fig.14 Examples of computing devices 1400 and mobile computing devices 1450 that can be used to implement the techniques described in the present disclosure are shown. Computing device 1400 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Mobile computing device 1450 is intended to represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, and other similar computing devices. The components, their connections and relationships, and their functions shown here are intended to be examples only and are not intended to be limiting.

[0310] The computing device 1400 includes a processor 1402, a memory 1404, a storage device 1406, a high-speed interface 1408 connected to the memory 1404 and a plurality of high-speed expansion ports 1410, and a low-speed interface 1412 connected to a low-speed expansion port 1414 and the storage device 1406. Each of the processor 1402, the memory 1404, the storage device 1406, the high-speed interface 1408, the high-speed expansion port 1410, and the low-speed interface 1412 is interconnected using various buses and can be mounted on a common motherboard or otherwise installed as appropriate. The processor 1402 can process instructions for execution within the computing device 1400, including instructions stored in the memory 1404 or on the storage device 1406 to display graphical information of a GUI on an external input / output device, such as a display 1416 coupled to the high-speed interface 1408. In other implementations, multiple processors and / or multiple buses can be used as needed along with multiple memories and memory types. Furthermore, multiple computing devices may be connected with each device providing a necessary portion of the operations (e.g., as a server bank, a group of blade servers, or a multi-processor system). Thus, when the term is used herein, where multiple functions are described as being performed by a "processor," this encompasses embodiments where the multiple functions are performed by any number of processors of any number of computing devices. Furthermore, when a function is described as being performed by a "processor," this encompasses embodiments where the function is performed by any number of processors of any number of computing devices (e.g., in a distributed computing system).

[0311] The memory 1404 stores information within the computing device 1400. In some implementations, the memory 1404 is one or more volatile memory units. In some implementations, the memory 1404 is one or more non-volatile memory units. The memory 1404 may also be another form of computer-readable medium, such as a magnetic disk or optical disk.

[0312] The storage device 1406 can provide mass storage for the computing device 1400. In some implementations, the storage device 1406 can be or contain a computer-readable medium, such as a floppy disk device, a hard disk device, an optical disk device, or a tape device, a flash memory or other similar solid-state memory device, or an array of devices, including devices in the form of a storage area network or other configuration. Instructions can be stored in an information carrier. The instructions, when executed by one or more processing devices (such as the processor 1402), perform one or more methods, such as those described above. Instructions can also be stored by one or more storage devices, such as a computer-readable or machine-readable medium (such as the memory 1404, the storage device 1406, or a memory on the processor 1402).

[0313] The high-speed interface 1408 manages bandwidth-intensive operations of the computing device 1400, while the low-speed interface 1412 manages less bandwidth-intensive operations. Such configuration of functions is merely an example. In some implementations, the high-speed interface 1408 is coupled to the memory device 1404, the display 1416 (e.g., via a graphics processor or accelerator), and the high-speed expansion port 1410, which can accommodate various expansion cards (not shown). In an implementation, the low-speed interface 1412 is coupled to the storage device 1406 and the low-speed expansion port 1414. Various communication ports (e.g., USB, The low-speed expansion port 1414 of Ethernet, wireless Ethernet) can be coupled to one or more input / output devices, such as a keyboard, a pointing device, a scanner, or a network connection device, such as a switch or a router, for example, through a network adapter.

[0314] The computing device 1400 can be implemented in a variety of different forms, as shown in the figure. For example, it can be implemented in a standard server 1420, or multiple times in a group of such servers. In addition, it can be implemented in the form of a personal computer, such as a laptop computer 1422. It can also be implemented in the form of a part of a framework server system 1424. Alternatively, components from the computing device 1400 can be combined with other components (not shown) in a mobile device such as a mobile computing device 1450. Each of these devices can contain one or more of the computing device 1400 and the mobile computing device 1450, and the entire system can be composed of multiple computing devices that communicate with each other.

[0315] The mobile computing device 1450 includes a processor 1452, a memory 1464, an input / output device (e.g., a display 1454), a communication interface 1466, and a transceiver 1468, among other components. The mobile computing device 1450 may also be provided with a storage device (e.g., a microdrive or other device) to provide additional storage. Each of the processor 1452, the memory 1464, the display 1454, the communication interface 1466, and the transceiver 1468 are interconnected using various buses, and several of the components may be mounted on a common motherboard or otherwise appropriately mounted.

[0316] Processor 1452 can execute instructions within mobile computing device 1450, including instructions stored in memory 1464. Processor 1452 can be implemented in the form of a chipset of chips that can include independent or multiple analog and digital processors. Processor 1452 can provide, for example, other components for cooperating with mobile computing device 1450, such as controls for a user interface, applications run by mobile computing device 1450, and wireless communications of mobile computing device 1450.

[0317] The processor 1452 can communicate with the user through the control interface 1458 and the display interface 1456 coupled to the display 1454. The display 1454 can be, for example, a thin film transistor (Thin-Film-Transistor, TFT) liquid crystal display or an organic light emitting diode (Organic Light Emitting Diode, OLED) display, or other suitable display technology. The display interface 1456 can include appropriate circuits for driving the display 1454 to present graphics and other information to the user. The control interface 1458 can receive commands from the user and transform them for submission to the processor 1452. In addition, the external interface 1462 can provide communication with the processor 1452 to enable nearby area communication between the mobile computing device 1450 and other devices. The external interface 1462 can provide, for example, wired communication in some implementations, or wireless communication in other implementations, and multiple interfaces can also be used.

[0318] Memory 1464 stores information within the mobile computing device 1450. Memory 1464 may be implemented in one or more of the following forms: one or more computer-readable media, one or more volatile memory units, or one or more non-volatile memory units. An expansion memory 1474 may also be provided and connected to the mobile computing device 1450 via an expansion interface 1472, which may include, for example, a Single In Line Memory Module (SIMM) card interface. The expansion memory 1474 may be additional storage space provided by the mobile computing device 1450, or may be a mobile computing device 1450 storing application programs or other information. Specifically, the expansion memory 1474 may include instructions for performing or supplementing the procedures described above, and may also include security information. Thus, for example, the expansion memory 1474 may be provided as a security module for the mobile computing device 1450, and may be programmed with instructions that permit secure use of the mobile computing device 1450. Additionally, secure applications and additional information may be provided through the SIMM card, such as placing authentication information on the SIMM card in a non-intrusive manner.

[0319] The memory may include, for example, flash memory and / or NVRAM memory (non-volatile random access memory), as discussed below. In some implementations, the instructions are stored in an information carrier. The instructions, when executed by one or more processing devices (e.g., processor 1452), perform one or more methods, such as those described above. The instructions may also be stored by one or more storage devices, such as one or more computer-readable or machine-readable media (e.g., memory 1464, expansion memory 1474, or memory on processor 1452). In some implementations, the instructions may be received in the form of a propagated signal, such as via transceiver 1468 or external interface 1462.

[0320] The mobile computing device 1450 can communicate wirelessly through the communication interface 1466, which may include digital signal processing circuits if necessary. The communication interface 1466 can provide communications under various modes or protocols, such as GSM voice calls (Global System for Mobile Communications), Short Message Service (SMS), Enhanced Messaging Service (EMS) or MMS messaging (Multimedia Messaging Service), Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Personal Digital Cellular (PDC), Wideband Code Division Multiple Access (WCDMA), CDMA2000 or General Packet Radio Service (GPRS). Such communications can be performed, for example, using radio frequencies through the transceiver 1468. In addition, short-range communications can be performed, for example, using Wi-Fi TM Or other such transceiver (not shown). In addition, the Global Positioning System (GPS) receiver module 1470 can provide additional navigation-related and location-related wireless data to the mobile computing device 1450, which can be used appropriately by applications running on the mobile computing device 1450.

[0321] The mobile computing device 1450 may also communicate audibly using an audio codec 1460, which may receive voice information from a user and convert it into usable digital information. The audio codec 1460 may likewise produce audible sounds to the user, such as through a speaker, such as in an earpiece of the mobile computing device 1450. Such sounds may include sounds from voice telephone calls, may include recorded sounds (e.g., voice messages, music files, etc.), and may also include sounds produced by applications operating on the mobile computing device 1450.

[0322] The mobile computing device 1450 may be implemented in a variety of different forms, as shown in the figure. For example, it may be implemented in the form of a cellular phone 1480. It may also be implemented in the form of a portion of a smart phone 1482, a personal digital assistant, or other similar mobile device.

[0323] Various implementations of the systems and techniques described herein can be realized in the form of digital electronic circuitry, integrated circuitry, specially designed application specific integrated circuits (ASICs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs executable and / or interpretable on a programmable system including at least one programmable processor, which can be special purpose or general purpose, coupled to receive data and instructions from and transmit data and instructions to: a storage system, at least one input device, and at least one output device.

[0324] These computer programs (also referred to as programs, software, software applications or code) include machine instructions for a programmable processor and may be implemented in a high-level procedural and / or object-oriented programming language and / or in a combination / machine language. As used herein, the terms machine-readable medium and computer-readable medium refer to any computer program product, apparatus and / or device (e.g., a disk, an optical disk, a memory, a programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, the programmable processor including a machine-readable medium that receives machine instructions as a machine-readable signal. The term machine-readable signal refers to any signal for providing machine instructions and / or data to a programmable processor.

[0325] To provide interaction with a user, the systems and techniques described herein may be implemented on a computer having a display device (e.g., a cathode ray tube (CRT) or liquid crystal display (LCD) monitor) for displaying information to the user, and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and the input from the user may be received in any form, including acoustic, voice, or tactile input.

[0326] The systems and techniques described herein may be implemented in a computing system that includes a back-end component (e.g., as a data server), or includes a middleware component (e.g., an application server), or includes a front-end component (e.g., a client computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or any combination of such back-end, middleware, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0327] A computing system may include clients and servers. The client and server are substantially remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

[0328] In some implementations, the various modules described herein may be separated, combined, or incorporated into a single or combined module.The modules depicted in the figures are not intended to limit the systems described herein to the software architecture shown therein.

[0329] The elements of the different implementations described herein may be combined to form other implementations not specifically described above. Elements may be removed from the programs, computer programs, databases, etc. described herein without adversely affecting their operation. In addition, the logic flow depicted in the figure does not require the specific order or sequential order shown to achieve the desired results. Various separate elements may be combined into one or more individual elements to perform the functions described herein.

[0330] Throughout the specification, where devices and systems are described as having, including, or comprising specific components, or where programs and methods are described as having, including, or comprising specific steps, it is contemplated that there are also devices and systems of the present invention that consist essentially of, or consist of, the described components, and there are programs and methods of the present invention that consist essentially of, or consist of, the described processing steps.

[0331] It should be understood that the order of steps or the order in which a certain action is performed is not important, as long as the present invention remains operable. In addition, two or more steps or actions may be performed simultaneously.

[0332] While the invention has been particularly shown and described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined in the following claims.

Claims

1. A method for automatically processing a 3D image of an individual to determine one or more patient indicator values ​​measuring the individual's (e.g., overall) disease burden and / or risk, the method comprising: (a) receiving, by a processor of a computing device, a 3D functional image of the individual obtained using a functional imaging modality; (b) segmenting, by the processor, a plurality of 3D hotspot volumes within the 3D functional image, each 3D hotspot volume corresponding to a local region having an increased intensity relative to its surroundings and representing a potential cancerous lesion within the individual, thereby obtaining a set of 3D hotspot volumes; (c) calculating, by the processor, for each specific metric of the one or more individual hotspot quantification metrics, a value of a specific individual hotspot quantification metric for each 3D hotspot volume of the set; and (d) determining, by the processor, the one or more patient indicator values, wherein at least a portion of the patient indicators are each associated with one or more specific individual hotspot quantitative metrics and are at least a portion (e.g., substantially all) of the one or more specific individual hotspot quantitative metric values ​​calculated for the set of 3D hotspot volumes; For example, a function that specifies a specific subset of a class.

2. A method according to claim 1, wherein at least one specific patient indicator of the one or more patient indicator values ​​is associated with a single specific individual hotspot quantitative metric and is calculated as a function of substantially all values ​​(e.g., all; e.g., excluding only statistical outliers) of the specific individual hotspot quantitative metric calculated for the 3D hotspot volume set (e.g., mean, median, mode, sum, etc.).

3. The method of claim 2, wherein the single specific individual hotspot quantification metric is an individual hotspot intensity metric that quantifies the intensity within a 3D hotspot volume (e.g., calculated for an individual 3D hotspot volume as a function of the intensity of voxels of the 3D hotspot volume).

4. The method of claim 3, wherein the individual hotspot intensity metric is an average hotspot intensity (eg, calculated, for an individual 3D hotspot volume, as an average of the intensities of the voxels within the 3D hotspot volume).

5. The method according to any one of claims 3 to 4, wherein the specific patient indicator is calculated as the sum of substantially all values ​​of the individual hotspot intensity metrics calculated for the set of 3D hotspot volumes.

6. The method of claim 6, wherein the single specific individual hotspot quantitative metric is the lesion volume (eg, for a specific 3D hotspot volume, calculated as the sum of the volumes of the individual voxels within the specific 3D hotspot volume).

7. A method according to claim 6, wherein the patient-specific indicator (the value of) is calculated as the sum of substantially all lesion volume values ​​calculated for the set of 3D hotspot volumes (e.g., so that the patient-specific indicator value provides a measure of the total lesion volume within the individual).

8. A method according to any of the preceding claims, wherein a particular one of the one or more overall patient indicators is associated with two or more particular individual hotspot quantification metrics and is calculated as a function of substantially all values ​​of the two or more particular individual hotspot quantification metrics calculated for the 3D hotspot volume set (e.g., a weighted sum, a weighted average, etc.).

9. The method of claim 8, wherein the two or more specific individual hotspot quantification metrics include (i) an individual hotspot intensity metric and (ii) a lesion volume.

10. The method of claim 9, wherein the individual hotspot intensity metric is an individual lesion index that maps values ​​of hotspot intensity to values ​​on a standardized scale.

11. The method according to claim 9 or 10, wherein the patient-specific indicator (the value) is calculated as the sum of intensity-weighted lesion (e.g. hotspot) volumes by: For each of substantially all of the 3D hotspot volumes, weighting the value of the lesion volume by the value of the individual hotspot intensity metric (e.g., calculating the product of the lesion volume value and the value of the individual hotspot intensity metric), thereby calculating a plurality of intensity-weighted lesion volumes; and A sum of substantially all of the intensity-weighted lesion volumes is calculated as a value for the patient-specific indicator.

12. A method according to any of the preceding claims, wherein the one or more individual hotspot quantification metrics include one or more individual hotspot intensity metrics that quantify the intensity within a 3D hotspot volume (e.g., calculated for an individual 3D hotspot volume as a function of the intensity of the stereo pixels of the 3D hotspot volume).

13. The method of claim 12, wherein the one or more individual hot spot quantification metrics include one or more members selected from the group consisting of: Average hotspot intensity (e.g., for a specific 3D hotspot volume, calculated as the average value of the intensities of the three-dimensional pixels within the specific 3D hotspot volume); Maximum hotspot intensity (e.g., for a particular 3D hotspot volume, calculated as the maximum value of the voxel intensities within the particular 3D hotspot volume); and Median hotspot intensity (eg, for a particular 3D hotspot volume, calculated as the median of the voxel intensities within the 3D hotspot volume).

14. The method of claim 12 or 13, wherein the one or more individual hotspot intensity metrics comprise a peak intensity of a 3D hotspot volume [For example, for a specific 3D hotspot volume, the value of the peak intensity is calculated as follows: (i) identifying maximum intensity voxels within the specific 3D hotspot volume; (ii) identifying voxels within a sub-region surrounding the maximum intensity voxel (e.g., voxels within a specific distance threshold including the maximum intensity voxel) and voxels within a specific 3D hotspot; and (iii) calculating the average value of the stereo pixel intensity in the sub-region as the corresponding peak intensity].

15. The method of any one of claims 12 to 14, wherein the one or more individual hotspot intensity metrics include an individual lesion index that maps values ​​of hotspot intensity to values ​​on a standardized scale.

16. The method according to claim 15, comprising: identifying, by the processor, within the 3D functional image, one or more 3D reference volumes each corresponding to a specific reference tissue region; determining, by the processor, one or more reference intensity values, each associated with a particular 3D reference volume of the one or more 3D reference volumes and corresponding to an intensity metric within the particular 3D reference volume; and In step (c), for each 3D hotspot volume in the set: Determining, by the processor, a corresponding value of a particular individual hotspot intensity metric (e.g., average hotspot intensity, median hotspot intensity, maximum hotspot intensity, etc.); and The corresponding value of the individual lesion indicator is determined by the processor based on the corresponding value of the specific individual hotspot intensity metric and the one or more reference intensity values.

17. The method according to claim 16, comprising: mapping each of the one or more reference intensity values ​​to a corresponding reference index value on a scale; and For each 3D hotspot volume, a corresponding value of the individual lesion index is determined using the reference intensity value and a corresponding reference index value to interpolate the corresponding individual lesion index value on a scale based on the corresponding value of the particular individual hotspot intensity metric.

18. The method of any one of claims 16 or 17, wherein the reference tissue region comprises one or more members selected from the group consisting of: liver, aorta, and parotid gland.

19. A method according to any one of claims 16 to 18, wherein: The first reference intensity value (i) is a blood reference intensity value associated with a reference volume corresponding to a portion of the aorta, and (ii) is mapped to a first reference index value; The second reference intensity value (i) is a liver reference intensity value associated with a reference volume corresponding to the liver, and (ii) is mapped to a second reference index value; and The second reference intensity value is greater than the first reference intensity value and the second reference index value is greater than the first reference index value.

20. A method according to any one of claims 16 to 19, wherein the reference intensity value comprises a maximum reference intensity value mapped to a maximum reference index value, and wherein a 3D hotspot volume whose corresponding value of the specific individual hotspot intensity metric is greater than the maximum reference intensity value is assigned an individual lesion index value equal to the maximum reference index value.

21. A method according to any one of the preceding claims, comprising: identifying one or more subsets within the set of 3D hotspot volumes, each associated with a specific tissue region and / or lesion classification; and For each specific subset of the one or more subsets, corresponding values ​​of one or more specific patient indicators are calculated using the values ​​of the individual hotspot quantification metrics calculated for the 3D hotspot volumes within the specific subset.

22. The method of claim 21, wherein the one or more subsets are each associated with a specific region of one or more tissue regions, and the method comprises identifying, for each specific tissue region, a subset of the 3D hotspot volumes located within a volume of interest corresponding to the specific tissue region.

23. The method of claim 22, wherein the one or more tissue regions comprise one or more members selected from the group consisting of: a skeletal region comprising one or more bones of the individual, a lymphatic region, and a prostate region.

24. A method according to any one of claims 21 to 23, wherein the one or more subsets are each associated with a specific subtype of one or more lesion subtypes [e.g., according to a lesion classification scheme (e.g., miTNM classification)], and the method comprises determining a corresponding lesion subtype for each 3D hotspot volume and assigning the 3D hotspot volume to the one or more subsets according to its corresponding lesion subtype.

25. A method according to any of the preceding claims, comprising using at least a portion of the one or more patient indicator values ​​as input into a prognostic model (e.g., a statistical model, such as a regression; such as a classification model, whereby patients are assigned to particular categories based on comparison of the one or more patient indicator values ​​to one or more threshold values; such as a machine learning model, into which the one or more patient indicator value inputs are received) that produces an expected value and / or range (e.g., category) output indicative of likely values ​​for a particular patient outcome (e.g., time, such as in months, representing expected survival, time to progression, time to radiographic progression, etc.).

26. The method of any of the preceding claims, comprising using at least a portion of the one or more patient indicator values ​​as input into a predictive model (e.g., a statistical model, such as a regression; such as a classification model, whereby a patient is assigned to a particular category based on a comparison of the one or more patient indicator values ​​to one or more thresholds; such as a machine learning model, into which the one or more patient indicator values ​​are received as input) that generates a prediction model for one or more treatment options (e.g., Abiraterone, Enzalutamide, Apalutamide, Darolutamide, Sipuleucel-T, Ra223, Docetaxel, Carbazitaxel, Pembrolizumab, Olaparib, Rucaparib, 177 The eligibility score output for each of the following treatment options and / or therapeutic agent classes [e.g., androgen biosynthesis inhibitors (e.g., abiraterone), androgen receptor inhibitors (e.g., enzalutamide, apalutamide, darutamide), cellular immunotherapy (e.g., siplucet-T), internal radiation therapy treatment (Ra223), anti-tumor drugs (e.g., docetaxel, cabazitaxel), immune checkpoint inhibitors (pembrolizumab), PARP inhibitors (e.g., olaparib, rucaparib), PSMA binders], wherein the eligibility score for a particular treatment option and / or therapeutic agent class indicates a prediction of whether the patient will benefit from the particular treatment and / or therapeutic agent class.

27. A method according to any preceding claim, comprising (e.g. automatically) generating a report [e.g. an electronic file, e.g. within a graphical user interface (e.g. for verification / sign-off by a user)] comprising at least a portion of the one or more patient indicator values.

28. The method of any of the preceding claims, wherein step (b) comprises using one or more machine learning modules [e.g., one or more neural networks (e.g., one or more convolutional neural networks)] to perform one or more functions selected from the group consisting of: detecting a plurality of hot spots, wherein at least a portion of the plurality of 3D hot spot volumes each corresponds to a particular detected hot spot and is generated by segmenting the particular detected hot spot; segmenting at least a portion of the plurality of 3D hotspot volumes; and At least a portion of the 3D hotspot volumes are classified (eg, a likelihood that each 3D hotspot volume represents a potential cancerous lesion is determined).

29. The method of any one of the preceding claims, wherein the 3D functional images comprise PET or SPECT images obtained after administration of an agent to the subject.

30. The method of claim 29, wherein the agent comprises a PSMA binding agent.

31. The method of claim 29 or 30, wherein the agent comprises 18 F.

32. The method of claim 30 or 31, wherein the agent comprises [18F]DCFPyL.

33. The method of claim 30, wherein the agent comprises PSMA-11.

34. The method of claim 30, wherein the agent comprises one or more selected from 99m Tc, 68 Ga, 177 Lu, 225 Ac, 111 In, 123 I. 124 I and 131 I is a member of a group composed of.

35. A method for automatically analyzing a time series of medical images [e.g., three-dimensional images, e.g., nuclear medicine images (e.g., bone scan (scintigraphy), PET and / or SPECT), e.g., anatomical images (e.g., CT, X-ray, MRI), e.g., combined nuclear medicine and anatomical images (e.g., overlapped)] of an individual, the method comprising: (a) receiving and / or accessing, by a processor of a computing device, a time series of medical images of the individual; and (b) identifying, by the processor, a plurality of hot spots within each of the medical images and determining, by the processor, one, two or all three of (i), (ii) and (iii): (i) a change in the number of identified lesions, (ii) a change in the overall volume of the identified lesions (e.g., a change in the sum of the volumes of each identified lesion), and (iii) a change in the PSMA (e.g., lesion index) weighted total volume (e.g., the sum of the product of the lesion index and the lesion volume for all lesions in the region of interest) [e.g., wherein the changes identified in step (b) are used to identify (1) disease status [e.g., progression, regression or no change], (2) make treatment management decisions [e.g., active surveillance, prostatectomy, anti-androgen therapy, prednisone, radiation, radiotherapy, radioPSMA therapy or chemotherapy], or (3) treatment efficacy (e.g., wherein the individual has started treatment or has continued treatment with a drug or other therapy according to an initial set of images in the time series of medical images)] [e.g., wherein step (b) comprises using a machine learning module / model].

36. A method for analyzing a plurality of medical images of an individual (e.g., to assess disease status and / or progression in the individual), the method comprising: (a) receiving and / or accessing, by a processor of a computing device, a plurality of medical images of the individual, and obtaining, by the processor, a plurality of 3D heat maps, each corresponding to a particular medical image (of the plurality of medical images) and identifying one or more hot spots within the particular medical image (e.g., representing possible underlying body lesions within the individual); (b) for each specific image (medical image) of the plurality of medical images, determining, by the processor, a corresponding 3D anatomical segmentation map using a machine learning module [e.g., a deep learning network (e.g., a convolutional neural network (CNN))], which identifies a set of organ regions within the specific medical image [e.g., representing soft tissue and / or bone structures within the individual (e.g., one or more cervical vertebrae; thoracic vertebrae; lumbar vertebrae; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and mandible)], thereby generating a plurality of 3D anatomical segmentation maps; (c) determining, by the processor, using (i) the plurality of 3D hotspot maps and (ii) the plurality of 3D anatomical segmentation maps, one or more lesion correspondences, each (lesion correspondence) identifying two or more corresponding hotspots within different medical images and determined (e.g., by the processor) to represent the same underlying physical lesion within the individual; and (d) determining, by the processor, one or more metrics {e.g., one or more hotspot quantification metrics and / or changes therein [e.g., quantifying the characteristics of individual hotspots and / or the potential body lesions they represent (e.g., over time / between multiple medical images), such as changes in volume, radiopharmaceutical absorption, shape, etc.] based on the multiple 3D hotspot maps and the identification of the one or more lesions corresponding thereto; such as patient indicators (e.g., measuring an individual's overall disease burden and / or condition and / or risk) and / or changes therein; such as values ​​for classifying a patient (e.g., belonging to and / or having a specific disease condition, progression, etc. category), such as prognostic metrics [e.g., indicating and / or quantifying the likelihood (e.g., overall survival) of one or more clinical outcomes (e.g., disease condition, progression, possible survival, treatment efficacy, etc.); such as predictive metrics (e.g., indicating predicted response to therapy and / or other clinical outcomes)} based on the multiple 3D hotspot maps and the identification of the one or more lesions corresponding thereto; 37. The method of claim 36, wherein the plurality of medical images comprises one or more anatomical images (eg, CT, X-ray, MRI, ultrasound, etc.).

38. The method of any one of claims 36 to 37, wherein the plurality of medical images comprises one or more nuclear medicine images [e.g., bone scan (scintigraphy) (e.g., obtained after administration of a radiopharmaceutical, such as 99mTc-MDP, to the subject), PET (e.g., obtained after administration of a radiopharmaceutical, such as [18F]DCFPyL, [68Ga]PSMA-11, [18F]PSMA-1007, rhPSMA-7.3(18F), [18F]-JK-PSMA-7, etc., to the subject), or SPECT (e.g., obtained after administration of a radiopharmaceutical, such as a 99mTc-labeled PSMA binder, to the subject)].

39. A method according to any one of claims 36 to 38, wherein the plurality of medical images comprises one or more composite images, each of which comprises an anatomical and a nuclear medicine pair (e.g., overlaid / co-registered with each other); For example, one or more PET / CT images have been acquired by the individual at substantially the same time.

40. The method according to any one of claims 36 to 39, wherein the plurality of medical images is or comprises a time series of medical images, each medical image of the time series relating to and having been acquired at a different specific time.

41. The method of claim 40, wherein the time series of medical images comprises a first medical image acquired before the administration (e.g., one or more cycles) of a specific therapeutic agent [e.g., a PSMA binder (e.g., PSMA-617; e.g., PSMA I&T); e.g., a radiopharmaceutical; e.g., a PSMA binder labeled with a radionuclide (e.g., 177Lu-PSMA-617; e.g., 177Lu-PSMA I&T)] to the individual, and a second medical image acquired after the administration (e.g., the one or more cycles) of the specific therapeutic agent to the individual.

42. The method of claim 41, comprising classifying the individual as a responder and / or non-responder to the particular therapeutic agent based on the values ​​of the one or more metrics determined in step (d).

43. A method according to any one of claims 36 to 42, wherein step (a) comprises generating each hotspot map by (e.g. automatically) segmenting at least a portion of the corresponding medical image (e.g. a sub-image thereof, such as a nuclear medicine image) (e.g. using a second hotspot segmentation, machine learning module [e.g. wherein the hotspot segmentation machine module comprises a deep learning network (e.g. a convolutional neural network (CNN))]).

44. A method according to any one of claims 36 to 43, wherein for each of at least a portion of the identified hotspots, each heat map comprises one or more markers (e.g., miTNM classification markers) identifying one or more assigned anatomical regions and / or lesion subtypes.

45. A method according to any one of claims 36 to 44, wherein: The plurality of heat maps include (i) a first heat map corresponding to a first medical image (e.g., and identifying a first set of one or more hot spots therein), and (ii) a second heat map corresponding to a second medical image (e.g., and identifying a second set of one or more hot spots therein); The plurality of 3D anatomical segmentation maps include (i) a first 3D anatomical segmentation map identifying a set of organ regions within the first medical image, and (ii) a second 3D anatomical segmentation map identifying a set of organ regions within the second medical image; and Step (c) includes registering (i) the first hotspot map with (ii) the second hotspot map using the first 3D anatomical segmented map and the second 3D anatomical segmented map (e.g., using the set of organ regions and / or one or more subsets thereof as landmarks within the first 3D anatomical segmented map and the second 3D anatomical segmented map to determine one or more registration fields (e.g., full 3D registration fields; e.g., point-by-point registration) and using the one or more determined registration fields to co-register the first hotspot map and the second hotspot map).

46. ​​The method of any one of claims 36 to 45, wherein step (c) comprises: for a set of two or more hot spots, each hot spot being a member of a different heat map and identified in a different medical image, determining values ​​of one or more lesion correspondence metrics (e.g., volume overlap; e.g., centroid distance; e.g., lesion type match); and The two or more hot spots of the group are determined to represent the same particular potential body lesion based on the values ​​of the one or more lesion correspondence metrics, thereby including the two or more hot spots of the group in one of the one or more lesion correspondences.

47. The method of any one of claims 36 to 46, wherein step (d) comprises determining one, two, or all three of (i), (ii), and (iii): (i) a change in the number of identified lesions, (ii) a change in the overall volume of the identified lesions (e.g., a change in the sum of the volumes of each identified lesion), and (iii) a change in the PSMA (e.g., lesion index) weighted total volume (e.g., the sum of the product of the lesion index and the lesion volume for all lesions in the region of interest) [e.g., wherein the change identified in step (b) is used to identify (1) disease status [e.g., progression, regression, or no change], (2) make a treatment management decision [e.g., active surveillance, prostatectomy, anti-androgen therapy, prednisone, radiation, radiotherapy, radioPSMA therapy, or chemotherapy], or (3) treatment efficacy (e.g., wherein the individual has started treatment or has continued treatment with a drug or other therapy according to an initial set of images in the time series of medical images)].

48. A method according to any one of claims 36 to 47, comprising determining (e.g. based on the values ​​of the one or more metrics; e.g. in step (d)) the value of one or more prognostic metrics indicative of disease status / progression and / or treatment [e.g. determining the expected overall survival (OS) (e.g. predicted number of months) of the individual].

49. The method of any one of claims 36 to 48, comprising using the values ​​of the one or more metrics (e.g., change in tumor volume, SUV mean, SUV maximum, PSMA score, number of new lesions, number of disappeared lesions, total number of follow-up lesions) to input a prognostic model (e.g., a statistical model, such as a regression; such as a classification model, whereby patients are assigned to a particular category based on comparison of the one or more patient indicator values ​​to one or more thresholds; such as a machine learning model, in which values ​​of the one or more patient indicators are received as input), which produces an expected value and / or range (e.g., category) output indicative of possible values ​​for a particular patient outcome (e.g., time, such as in months, representing expected survival, time to progression, time to radiographic progression, etc.).

50. The method of any one of claims 36 to 49, comprising using the values ​​of the one or more metrics (e.g., change in tumor volume, SUV mean, SUV maximum, PSMA score, number of new lesions, number of disappeared lesions, total number of follow-up lesions) to input a response model (e.g., a statistical model, such as regression; such as a classification model, whereby patients are assigned to a particular category based on comparison of the one or more patient indicator values ​​to one or more thresholds; such as a machine learning model, in which values ​​of the one or more patient indicators are received as input), which produces a classification (e.g., a binary classification) output indicative of the patient's response to treatment.

51. The method of any one of claims 36 to 50, comprising using the values ​​of the one or more metrics (e.g., change in tumor volume, SUV mean, SUV maximum, PSMA score, number of new lesions, number of disappeared lesions, total number of follow-up lesions) to input a predictive model (e.g., a statistical model, such as regression; such as a classification model, whereby a patient is assigned to a particular category based on a comparison of the one or more patient indicator values ​​to one or more thresholds; such as a machine learning model, in which the values ​​of the one or more patient indicators are received as input) that generates a prediction model for one or more treatment options (e.g., abiraterone, enzalutamide, apalutamide, darutamide, sipralex-T, Ra223, docetaxel, cabazitaxel, pembrolizumab, The eligibility score output is for each of a plurality of therapeutic options and / or therapeutic classes [e.g., androgen biosynthesis inhibitors (e.g., abiraterone), androgen receptor inhibitors (e.g., enzalutamide, apalutamide, darutamide), cellular immunotherapy (e.g., siplucezet-T), internal radiation therapy (Ra223), anti-neoplastic agents (e.g., docetaxel, cabazitaxel), immune checkpoint inhibitors (pembrolizumab), PARP inhibitors (e.g., olaparib, rucaparib), PSMA binders], wherein the eligibility score for a particular treatment option and / or therapeutic class indicates a prediction of whether the patient will benefit from the particular treatment and / or therapeutic class.

52. A method for analyzing a plurality of medical images of an individual, the method comprising: (a) obtaining (e.g., receiving and / or accessing, and / or generating) a first 3D heat map of the individual by a processor of a computing device; (b) obtaining (e.g., receiving and / or accessing, and / or generating) by the processor a first 3D anatomical segmentation map associated with the first 3D heat map; (c) obtaining (e.g., receiving and / or accessing, and / or generating) by the processor a second 3D heat map of the individual; (d) obtaining (e.g., receiving and / or accessing, and / or generating) by the processor a second 3D anatomical segmentation map associated with the second 3D heat map; (e) determining, by the processor, a registration field (e.g., a full 3D registration field; e.g., point-by-point registration) using / based on the first 3D anatomical segmented map and the second 3D anatomical segmented map; (f) registering, by the processor, the first 3D heat map with the second 3D heat map using the determined registration field, thereby generating a co-registered pair of 3D heat maps; (g) determining, by the processor, identification of one or more lesions corresponding to the co-registered pairs of the 3D heat maps; and (h) storing and / or providing, by the processor, identifications corresponding to the one or more lesions for display and / or further processing.

53. A method for analyzing a plurality of medical images of an individual (e.g., to assess disease status and / or progression in the individual), the method comprising: (a) receiving and / or accessing, by a processor of a computing device, a plurality of medical images of the individual; (b) for each specific image (medical image) of the plurality of medical images, determining, by the processor, a corresponding 3D anatomical segmentation map using a machine learning module [e.g., a deep learning network (e.g., a convolutional neural network (CNN))], which identifies a set of organ regions within the specific medical image [e.g., representing soft tissue and / or bone structures within the individual (e.g., one or more cervical vertebrae; thoracic vertebrae; lumbar vertebrae; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and mandible)], thereby generating a plurality of 3D anatomical segmentation maps; (c) determining, by the processor, one or more registration fields (e.g., full 3D registration fields; e.g., point-by-point registration) using the plurality of 3D anatomical segmented maps and applying the one or more registration fields to register the plurality of medical images, thereby generating a plurality of registered medical images; (d) for each particular image in the plurality of registered medical images, determining, by the processor, a corresponding registered 3D heat map that identifies one or more hot spots within the particular registered medical image (e.g., representing possible underlying physical lesions within the individual), thereby generating a plurality of registered 3D heat maps; (e) determining, by the processor, using the plurality of 3D registered heat maps, identifications of one or more lesion correspondences, each (lesion correspondence) identifying two or more corresponding hot spots within different medical images and determined (e.g., by the processor) to represent the same underlying body lesion within the individual; and (f) determining, by the processor, one or more metrics {e.g., one or more hotspot quantification metrics and / or changes therein [e.g., quantifying the characteristics of individual hotspots and / or the potential body lesions they represent (e.g., over time / between multiple medical images), such as changes in volume, radiopharmaceutical absorption, shape, etc.] based on the multiple 3D hotspot maps and the identification of the one or more lesions corresponding thereto; such as patient indicators (e.g., measuring an individual's overall disease burden and / or condition and / or risk) and / or changes therein; such as values ​​for classifying a patient (e.g., belonging to and / or suffering from a specific disease condition, progression, etc. category), such as prognostic metrics [e.g., indicating and / or quantifying one or more clinical outcomes (e.g., disease condition, progression, possible survival, treatment efficacy, etc.) (e.g., overall survival); such as predictive metrics (e.g., indicating predicted response to therapy and / or other clinical outcomes)} based on the multiple 3D hotspot maps and the identification of the one or more lesions corresponding thereto; 54. A method for analyzing a plurality of medical images of an individual, the method comprising: (a) obtaining (e.g., receiving and / or accessing, and / or generating) by a processor of a computing device a first 3D anatomical image (e.g., CT, X-ray, MRI, etc.) and a first 3D functional image [e.g., a nuclear medicine image (e.g., PET, SPECT, etc.)] of the individual; (b) obtaining (e.g., receiving and / or accessing, and / or generating) by the processor a second 3D anatomical image and a second 3D functional image of the individual; (c) obtaining (e.g., receiving and / or accessing, and / or generating) by the processor a first 3D anatomical segmentation map based on (e.g., using) the first 3D anatomical image; (d) obtaining (e.g., receiving and / or accessing, and / or generating), by the processor, a second 3D anatomical segmentation map based on (e.g., using) the second 3D anatomical image; (e) determining, by the processor, a registration field (e.g., a full 3D registration field; e.g., point-by-point registration) using / based on the first 3D anatomical segmented map and the second 3D anatomical segmented map; (f) registering (aligning), by the processor, the second 3D functional image with the first 3D functional image using the registration field, thereby producing a registered version of the second 3D functional image; (g) obtaining, by the processor, a first 3D heat map associated with the first functional image; (h) determining, by the processor, a second 3D heat map using the registered version of the second 3D functional image, whereby the second 3D heat map is registered with the first 3D heat map; (i) determining, by the processor, identification corresponding to one or more lesions using the first 3D heat map and the second 3D heat map registered therewith; and (j) storing and / or providing, by the processor, identifications corresponding to the one or more lesions for display and / or further processing.

55. A method for evaluating the efficacy of an intervention, the method comprising: (a) for each specific individual in a test population (e.g. comprising a plurality of individuals, e.g. enrolled in a clinical trial) presenting with a specific disease (e.g. prostate cancer (e.g. metastatic castration-resistant prostate cancer)) and / or at risk for a specific disease, performing a method according to any of the preceding claims to obtain a plurality of medical images of the specific patient, wherein the plurality of medical images of the specific patient comprises a time series of medical images obtained over a time period (e.g. before, during and / or after) a tested intervention, and the one or more risk indicators comprise one or more endpoints indicative of a patient response to the tested intervention, thereby determining a plurality of values ​​for each of the one or more endpoints in the test population; and (b) determining the efficacy of the tested intervention based on the values ​​of the one or more endpoints in the test population.

56. A method for treating an individual having and / or at risk for a particular disease, such as prostate cancer (e.g., metastatic castration-resistant prostate cancer), the method comprising: administering a first cycle of a therapeutic agent to the individual; and A second cycle of the therapeutic agent is administered to the individual based on the individual having been imaged (e.g., before and / or during and / or after the first cycle of the therapeutic agent) and identified as a responder to the therapeutic agent using the method according to any one of claims 1 to 52 (e.g., the individual has been identified / classified as a responder based on the values ​​of the one or more risk indicators determined using the method according to any one of claims 1 to 52).

57. A method for treating an individual having and / or at risk for a particular disease, such as prostate cancer (e.g., metastatic castration-resistant prostate cancer), the method comprising: a period of administering a first therapeutic agent to the individual; and A cycle of administering a second therapeutic agent to the individual based on the individual having been imaged (e.g., before and / or during and / or after a cycle of the first therapeutic agent) and identified as a non-responder to the first therapeutic agent using the method according to any one of claims 1 to 52 (e.g., based on the values ​​of one or more risk indicators determined using the method according to any one of claims 1 to 52, the individual has been identified / classified as a non-responder) (e.g., thereby subjecting the individual to a potentially more effective therapy).

58. A method for treating an individual having and / or at risk for a particular disease, such as prostate cancer (e.g., metastatic castration-resistant prostate cancer), the method comprising: a period for administering a therapeutic agent to the individual; and discontinuing administration of a therapeutic agent to the individual based on the individual having been imaged (e.g. before and / or during and / or after a cycle of the first therapeutic agent) and identified as a non-responder to the therapeutic agent using the method according to any one of claims 1 to 52 (e.g. based on the values ​​of the one or more risk indicators determined using the method according to any one of claims 1 to 52, the individual has been identified / classified as a non-responder) (e.g. thereby subjecting the individual to a potentially more effective therapy).

59. A method for automatically or semi-automatically systemically assessing an individual with metastatic prostate cancer [e.g., metastatic castration-resistant prostate cancer (mCRPC) or metastatic hormone-sensitive prostate cancer (mHSPC)] to assess disease progression and / or treatment efficacy, the method comprising: (a) receiving, by a processor of a computing device, a first positron emission tomography (PET) image of the individual that targets prostate specific membrane antigen (PSMA) (a first PSMA-PET image) and a first 3D anatomical image of the individual [e.g., a computed tomography (CT) image; e.g., a magnetic resonance image (MRI)], wherein the first 3D anatomical image of the individual was obtained simultaneously with, immediately after, or immediately before (e.g., on the same date as) the first PSMA PET image, such that the first 3D anatomical image and the first PSMA PET image correspond to a first date, and wherein the images depict a sufficiently large area of ​​the individual's body to encompass areas of the body to which the metastatic prostate cancer has spread (e.g., a full torso image or a full body image encompassing multiple organs) {e.g., wherein the PSMA-PET image was acquired using F-18piflufolastat PSMA (i.e., 2-(3-{1-carboxy-5-[(6-[18F]fluoro-pyridine-3-carbonyl)amino]-pentyl}ureido)-glutaric acid, also known as [18F]F-DCFPyL), or Ga-68 PSMA-11, or other radiolabeled prostate specific membrane antigen inhibitor imaging agents}; (b) receiving, by the processor, a second PSMA-PET image of the individual and a second 3D anatomical image of the individual, both acquired on a second date after the first date; (c) automatically determining, by the processor, a registration field using automatically identified landmarks within the first 3D anatomical image and the second 3D anatomical image (e.g., the identified regions represent one or more of the cervical spine; thoracic spine; lumbar spine; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and mandible), and aligning, by the processor, the first PSMA-PET image and the second PSMA-PET image using the determined registration field [e.g., before or after segmenting the CT and / or PSMA-PET images to identify boundaries of organs and / or bones, and before or after automatic hotspot (e.g., lesion) detection from the PSMA-PET images]; and (d) using the first PSMA-PET image and the second PSMA-PET image thus aligned, to automatically detect (e.g., stage and / or quantify) a change (e.g., progression or remission) in the disease from the first date to the second date by the processor [e.g., automatically identifying and / or identifying as such (e.g., tagging, labelling) one or both of the following (i) and (ii): (i) a change in the number of lesions {e.g., one or more new lesions (e.g., organ-specific lesions), or elimination of one or more lesions (e.g., organ-specific)}, and (ii) a change in tumor size {e.g., an increase in tumor size (PSMA-VOL increase / decrease), e.g., total tumor size, or a decrease in tumor size (PSMA-VOL decrease)} {e.g., a change in the volume of each of one or more specific lesions, or a change in the overall volume of a specific type of lesion (e.g., organ-specific tumor), or a change in the total volume of identified lesions}.

60. The method of claim 59, wherein the method comprises one or more members selected from the group consisting of: lesion location assignment, tumor staging, nodal staging, distant metastasis staging, assessment of intraprostatic lesions, and determination of PSMA expression score.

61. The method of claim 59 or 60, wherein the subject has been administered a therapy {e.g., hormone therapy, chemotherapy and / or radiation therapy, e.g., androgen ablation therapy, e.g., a 177Lu-containing compound, e.g., a 177Lu-PSMA radioligand therapy, e.g., 177Lu-PSMA-617, e.g., Lu 177 vipivotide tetraxetan (Pluvicto), e.g., cabazitaxel} for treating the metastatic prostate cancer one or more times from the first date to the second date (after obtaining the first image and before obtaining the second image), such that the method is used to assess treatment efficacy.

62. The method of any one of claims 59 to 61, further comprising obtaining one or more additional PSMA PET images and 3D anatomical images of the subject after the second date, aligning the additional PSMA PET images with corresponding 3D anatomical images, and using the aligned additional PSMA PET images to assess the disease progression and / or treatment efficacy.

63. The method of any one of claims 59 to 62, further comprising determining and presenting, by the processor, a predicted PSMA-PET image based at least in part on the change in the disease detected from the first date to the second date, the image depicting predicted progression (or remission) of disease until a future date (e.g., a future date later than the second date or any other subsequent date for which a PSMA-PET image has been obtained).

64. A method of quantifying and reporting disease (e.g., tumor) burden in a patient having cancer and / or at risk for cancer, the method comprising: (a) obtaining, by a processor of a computing device, a medical image of the patient; (b) detecting, by the processor, one or more (e.g., a plurality of) hot spots within the medical image, each hot spot within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein voxels of the 3D hot spot volume have elevated intensity relative to their surroundings (e.g., and / or other indication or increased radiopharmaceutical uptake)] and representing possible underlying physical pathology within the individual; (c) for each specific lesion category of a plurality of lesion categories representing specific tissue regions and / or lesion subtypes: identifying, by the processor, a corresponding subset of the one or more hot spots as belonging to the particular lesion class (e.g., based on a determination, by the processor, that the hot spots represent potential body lesions that are located within a particular tissue region and / or that are of a particular lesion subtype represented by the particular lesion class); and determining, by the processor, values ​​of one or more patient indicators quantifying disease (e.g., tumor) burden within and / or associated with the particular lesion class based on the corresponding hotspot subset; and (d) presenting, by the processor, a graphical representation of the patient indicator value calculated for each of the plurality of lesion categories (e.g., a summary table listing each lesion category and the patient indicator value calculated for each lesion category), thereby providing a user with a graphical report summarizing tumor burden within a particular tissue region and / or associated with a particular lesion subtype.

65. The method of claim 64, wherein the plurality of lesion categories comprises one or more of: (i) a local tumor category (e.g., a "T" or "miT" category) that identifies potential lesions and / or portions thereof located within one or more local tumor-associated tissue regions associated with and / or adjacent to a local (e.g., primary) tumor site in the patient and represented by the corresponding hotspot subset [e.g., wherein the cancer is prostate cancer and the one or more local tumor-associated tissue regions comprise the prostate and optionally one or more adjacent structures (e.g., seminal vesicles, external sphincter, rectum, bladder, levator muscles and / or pelvic wall); e.g., wherein the cancer is breast cancer and the one or more local tumor-associated tissue regions comprise the breast; e.g., wherein the cancer is colorectal cancer and the one or more local tumor-associated tissue regions comprise the colon; e.g., wherein the cancer is lung cancer and the one or more local tumor-associated tissue regions comprise the lung]; (ii) a regional nodal category (e.g., an "N" or "miN" category) that identifies potential lesions located within regional lymph nodes that are adjacent to and / or proximal to the original (e.g., primary) tumor site and are represented by the corresponding hotspot subset [e.g., where the cancer is prostate cancer and the regional lymph node category identifies hotspots representing lesions located within one or more pelvic lymph nodes (e.g., internal iliac, external iliac, obturator, presacral nodes, or other pelvic lymph nodes)]; and (iii) one or more (e.g., distant) metastatic tumor categories (e.g., one or more "M" or "miM" categories) that identify potential metastases (e.g., lesions that have spread beyond the original (e.g., primary) tumor site) and / or subtypes thereof and are represented by the corresponding hotspot subset [e.g., wherein the cancer is prostate cancer and the one or more metastatic tumor categories identify hotspots representing potential metastatic lesions located outside the patient's pelvic region (e.g., defined by the pelvic brim, e.g., according to the American Joint Committee on Cancer staging manual)].

66. The method of claim 64 or 65, wherein the one or more metastatic tumor types include one or more of the following: a distant lymph node metastasis category (e.g., a "Ma" or "miMa" category) that identifies potential lesions that have metastasized to distant lymph nodes and are represented by the corresponding hotspot subset [e.g., wherein the cancer is prostate cancer and the distant lymph node region category identifies hotspots representing lesions located in lymph nodes outside the pelvis (e.g., outside the pelvic region) (e.g., common iliac, retroperitoneal, supra-diaphragmatic, inguinal, and other extra-pelvic lymph nodes)]; a distal bone metastasis category (e.g., a "Mb" or "miMb" category) that identifies potential lesions located within one or more bones (e.g., distal bones) of the patient and represented by the corresponding hotspot subset; and A visceral (also called distant soft tissue) metastasis category (e.g., a "Mc" or "miMc" category) that identifies potential lesions located in one or more organs or other non-lymphoid soft tissue areas outside of the local tumor-associated tissue area and represented by the corresponding hotspot subset (e.g., where the cancer is prostate cancer and the visceral metastasis category identifies hotspots representing potential lesions located in the patient's extra-pelvic organs, such as the brain, lungs, liver, spleen, and kidneys).

67. The method of any one of claims 64 to 66, wherein step (c) comprises determining, for each specific lesion category, a value for one or more of the following patient indicators: a lesion count that quantifies the number of (e.g., distinct) lesions represented by a subset of hotspots corresponding to the particular lesion class (e.g., calculated as the number of hotspots within the corresponding subset); a maximum absorption value quantifying the maximum absorption within the corresponding set of hotspots (e.g., calculated as the maximum individual voxel intensity of all voxels within the hotspot volume of the corresponding subset); a mean absorption value quantifying the overall average absorption within the corresponding hotspot subset (e.g., calculated as the overall average intensity of all voxels within the (total combined) hotspot volume of the corresponding subset); a total volume lesion volume, which quantifies the total volume of lesions belonging to the particular lesion class (e.g., calculated as the sum of the volumes of all individual lesions (e.g., hotspots) of the corresponding subset); and An intensity-weighted tumor volume (ILTV) score (e.g., an aPSMA score) is calculated as the weighted sum of all individual lesion volumes weighted (e.g., multiplied) by the measurements of their intensities [e.g., where the measurements of their intensities are lesion indicators that quantify hotspot intensities on a standardized scale based on comparison to one or more reference intensities indicative of physiological (e.g., normal, non-cancer-related) radiopharmaceutical uptake within one or more corresponding reference tissue regions, such as portions of the aorta and the liver].

68. A method according to claims 64 to 67, comprising determining, for each of the lesion categories, an alpha-numeric code classifying the overall burden within the specific lesion category (e.g., a miTNM staging code indicating (i) the specific lesion category and (ii) one or more numbers and / or numbers indicating the specific number, size, spatial extent, spatial pattern and / or sublocation of hotspots of the corresponding subset, and thereby the underlying body lesions they represent), and optionally in step (e), generating and / or displaying a representation of the alpha-numeric code for each specific lesion category.

69. The method according to any one of claims 64 to 68 further comprises determining an overall disease stage (e.g., an alphanumeric code) of the patient based on the multiple lesion categories and their corresponding hotspot subsets, which indicates the patient's overall disease state and / or burden, and presenting, by the processor, a graphical representation of the overall disease stage (e.g., an alphanumeric code) included in the report.

70. The method of any one of claims 64 to 69, further comprising: determining, by the processor, one or more reference intensity values, each indicative of physiological (e.g., normal, non-cancer-related) uptake of a radiopharmaceutical within a particular reference tissue region (e.g., a portion of an aorta; e.g., a liver) within the patient and calculated based on intensities of image voxels within corresponding reference volumes identified within the medical image; and In step (d), a representation (eg, a graph) of the one or more reference intensity values ​​is presented by the processor for inclusion in the report.

71. A method for characterizing and reporting individual lesions detected based on imaging assessment of a patient having cancer and / or at risk for cancer, the method comprising: (a) obtaining, by a processor of a computing device, a medical image of the patient; (b) detecting, by the processor, a set of one or more (e.g., a plurality) of hot spots within the medical image, each hot spot of the set within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein voxels of the 3D hot spot volume have elevated intensity relative to their surroundings (e.g., and / or other indication or increased radiopharmaceutical uptake)] and representing a possible underlying physical pathology within the individual; (c) assigning, by the processor, one or more lesion category labels to each of the one or more hotspots in the set, each lesion category label category representing a specific tissue region and / or lesion subtype and identifying the hotspot as representing a potential lesion located in the specific tissue region and / or belonging to the lesion subtype; (d) calculating, by the processor, for each specific metric in the one or more individual hotspot quantitative metrics, a value of a specific individual hotspot quantitative metric for each hotspot of the set; and (e) displaying, by the processor, a graphical representation comprising, for each particular hotspot of at least a portion of the set of hotspots, an identification of the particular hotspot (e.g., a row in a table, and optionally an alphanumeric identification, such as an identification of the number of the particular hotspot), and one or more lesion category labels assigned to the particular hotspot and values ​​of the one or more individual hotspot quantitative metrics calculated for the particular hotspot [e.g., an overview table (e.g., a scrollable overview table) listing each hotspot in a row and the assigned lesion categories and hotspot quantitative metrics column-wise].

72. The method of claim 71, wherein the lesion category label comprises a label representing one or more of: (i) a local tumor category (e.g., a "T" or "miT" category) that identifies potential lesions and / or portions thereof located within one or more local tumor-associated tissue regions associated with and / or adjacent to a local (e.g., primary) tumor site in the patient and represented by the corresponding hotspot subset [e.g., wherein the cancer is prostate cancer and the one or more local tumor-associated tissue regions comprise the prostate and optionally one or more adjacent structures (e.g., seminal vesicles, external sphincter, rectum, bladder, levator muscles and / or pelvic wall); e.g., wherein the cancer is breast cancer and the one or more local tumor-associated tissue regions comprise the breast; e.g., wherein the cancer is colorectal cancer and the one or more local tumor-associated tissue regions comprise the colon; e.g., wherein the cancer is lung cancer and the one or more local tumor-associated tissue regions comprise the lung]; (ii) a regional nodal category (e.g., an "N" or "miN" category) that identifies potential lesions located within regional lymph nodes that are adjacent to and / or proximal to the original (e.g., primary) tumor site and are represented by the corresponding hotspot subset [e.g., where the cancer is prostate cancer and the regional lymph node category identifies hotspots representing lesions located within one or more pelvic lymph nodes (e.g., internal iliac, external iliac, obturator, presacral tubercles, or other pelvic lymph nodes)]; and (iii) one or more (e.g., distant) metastatic tumor categories (e.g., one or more "M" or "miM" categories) that identify potential metastases (e.g., lesions that have spread beyond the original (e.g., primary) tumor site) and / or subtypes thereof and are represented by the corresponding hotspot subset [e.g., wherein the cancer is prostate cancer and the one or more metastatic tumor categories identify hotspots representing potential metastatic lesions located outside the patient's pelvic region (e.g., defined by the pelvic brim, e.g., according to the American Joint Committee on Cancer Staging Manual)].

73. The method of claim 72, wherein the one or more metastatic tumor types include one or more of the following: a distant lymph node metastasis category (e.g., a "Ma" or "miMa" category) that identifies potential lesions that have metastasized to distant lymph nodes and are represented by the corresponding hotspot subset [e.g., wherein the cancer is prostate cancer and the distant lymph node region category identifies hotspots representing lesions located in extra-pelvic (e.g., extra-pelvic region) lymph nodes (e.g., common iliac, retroperitoneal, supra-diaphragmatic, inguinal, and other extra-pelvic lymph nodes)]; a distal bone metastasis category (e.g., a "Mb" or "miMb" category) that identifies potential lesions located within one or more bones (e.g., distal bones) of the patient and represented by the corresponding hotspot subset; and A visceral (also called distant soft tissue) metastasis category (e.g., a "Mc" or "miMc" category) that identifies potential lesions located in one or more organs or other non-lymphoid soft tissue areas outside of the local tumor-associated tissue area and represented by the corresponding hotspot subset (e.g., where the cancer is prostate cancer and the visceral metastasis category identifies hotspots representing potential lesions located in the patient's extra-pelvic organs, such as the brain, lungs, liver, spleen, and kidneys).

74. A method according to any one of claims 71 to 73, wherein the lesion category label comprises one or more tissue markers that identify a specific organ or bone, wherein (the lesion represented by a hotspot) is determined (e.g., based on comparison of the hotspot with an anatomical segmentation map) to be located in the specific organ or bone (e.g., one or more of the organ or bone regions listed in Table 1).

75. The method of any one of claims 71 to 74, wherein the one or more individual hotspot quantification metrics include one or more of: maximum intensity, peak intensity, average intensity (e.g., SUV average), lesion volume, and lesion index.

76. A method of quantifying and reporting disease (e.g., tumor) progression and / or risk over time in a patient having cancer and / or at risk of cancer, the method comprising: (a) obtaining, by a processor of a computing device, a plurality of medical images of the patient, each medical image representing a scan of the patient obtained at a particular time (e.g., a longitudinal dataset); (b) for each particular image of the plurality of medical images, detecting, by the processor, a corresponding set of one or more (e.g., a plurality) of hot spots within the particular medical image, each hot spot within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein voxels of the 3D hot spot volume have elevated intensity relative to their surroundings (e.g., and / or other indication or increased radiopharmaceutical uptake)] and representing possible underlying physical pathology within the individual; (c) for each specific indicator of the one or more (e.g., overall) patient indicators that measure (e.g., quantify) the overall disease (e.g., tumor) load in the patient at a specific time, determining, by the processor, a value of the specific (e.g., overall) patient indicator for each specific medical image of the plurality of medical images based on the corresponding set of hot spots detected for the specific medical image, thereby determining, for each specific indicator of the one or more patient indicators, a set of values ​​that tracks changes in disease load by measuring the specific patient indicator values ​​over time; and (d) displaying, by the processor, a graphical representation of a set of values ​​of at least a portion (e.g., a specific one, a specific subset) of the one or more patient indicator values, thereby conveying a measure of disease progression of the patient over time.

77. The method of claim 76, wherein the one or more patient indicators comprise: a lesion count quantifying the number of (e.g., distinct) lesions represented by a set of hotspots corresponding to a particular medical image and detected within the particular medical image (e.g., at a particular point in time) (e.g., calculated as the number of hotspots within the corresponding set of hotspots); a maximum absorption value quantifying the maximum absorption within a corresponding hotspot set of a particular medical image (e.g., calculated as the maximum individual voxel intensity of all voxels within a hotspot volume of the corresponding hotspot set of the particular medical image); a mean absorption value quantifying the overall average absorption within the corresponding set of hotspots (e.g., calculated as the overall average intensity of all voxels within the (total combined) hotspot volume of the corresponding set); a total volumetric lesion volume, which quantifies the total volume of lesions detected within the individual at a particular point in time (e.g., calculated as the sum of all individual hotspot volumes of the corresponding set of hotspots detected within a particular medical image); and An intensity-weighted tumor volume (ILTV) score (e.g., an aPSMA score) is calculated as a weighted sum of all individual lesion volumes, each lesion volume being weighted (e.g., multiplied) by a measure of its intensity [e.g., where the measure of hotspot intensity is a lesion indicator, the hotspot intensity is quantified on a standardized scale based on comparison to one or more reference intensities indicative of physiological (e.g., normal, non-cancer-related) radiopharmaceutical uptake within one or more corresponding reference tissue regions, such as portions of the aorta and the liver].

78. The method according to any one of claims 76 to 77 further comprises, for each specific medical image of the multiple medical images, determining an overall disease stage (e.g., an alphanumeric code) based on the corresponding hotspot set and indicating the overall disease state and / or burden of the patient at a specific time point, and presenting a graphical representation of the overall disease stage (e.g., an alphanumeric code) at each time point by the processor.

79. The method of any one of claims 76 to 78, further comprising: determining, by the processor, for each of the plurality of medical images, a set of one or more reference intensity values, each indicative of physiological (e.g., normal, non-cancer related) uptake of a radiopharmaceutical within a particular reference tissue region (e.g., a portion of an aorta; e.g., a liver) within the patient and calculated based on intensities of image voxels within a corresponding reference volume identified within the medical image; and A representation (eg, a table; eg, traces in a graph) of the one or more reference intensity values ​​is presented by the processor.

80. A method for automatically processing a 3D image of a subject to determine one or more patient indicator values ​​measuring the subject's (e.g., overall) disease burden and / or risk, the method comprising: (a) receiving, by a processor of a computing device, a 3D functional image of the individual obtained using a functional imaging modality; (b) segmenting, by the processor, a plurality of 3D hotspot volumes within the 3D functional image, each 3D hotspot volume corresponding to a local region having an increased intensity relative to its surroundings and representing a potential cancerous lesion within the individual, thereby obtaining a set of 3D hotspot volumes; (c) calculating, by the processor, for each specific metric in one or more individual hotspot quantification metrics, a value of a specific individual hotspot quantification metric for each 3D hotspot volume of the set, wherein for a specific individual 3D hotspot volume, each hotspot quantification metric quantifies a characteristic (e.g., intensity, volume, etc.) of the specific 3D hotspot volume and is (e.g., calculated as) a specific function of the intensity and / or number of individual stereo pixels within the specific 3D hotspot volume; and (d) determining, by the processor, the one or more patient indicator values, wherein at least a portion of the patient indicators are each associated with one or more specific individual hotspot quantification metrics and are calculated using (e.g., the same) specific function of the intensity and / or number of stereoscopic pixels within a combined hotspot volume, wherein the combined hotspot volume includes at least a portion (e.g., substantially all; e.g., a specific subset) of a set of 3D hotspot volumes (e.g., forming a union thereof).

81. A method for automatically determining the prognosis of an individual with prostate cancer from one or more medical images [e.g., one or more PSMA PET images (PET images obtained when a PSMA targeting compound is administered to the individual) and / or one or more anatomical (e.g., CT) images] of the individual, the method comprising: (a) receiving and / or accessing, by a processor of a computing device, one or more images of the individual; (b) automatically determining, by the processor, from the one or more images, a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) molecular imaging TNM (miTNM) lesion type classifications (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) for localized (T), pelvic node (N), and / or extra-pelvic (M) disease; (ii) an indication of lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) a standard physiological uptake value (SUV) (e.g., maximum SUV, peak SUV, average SUV); (iv) total lesion volume; (v) a change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein); and (c) automatically determining a prognosis for the individual from the quantitative assessment in (b), wherein the prognosis comprises one or more of the following for the individual: (I) expected survival (e.g., number of months), (II) expected time to disease progression, (III) expected time to radiographic progression, (IV) risk of simultaneous (synchronous) metastasis, and (V) risk of future (metachronous) metastasis.

82. The method of claim 81, wherein the quantitative assessment of the one or more prostate cancer lesions determined in step (b) comprises one or more of: (A) total tumor volume, (B) change in tumor volume, (C) total SUV, and (D) PSMA score, and wherein the prognosis of the individual determined in step (c) comprises one or more of: (E) expected survival (e.g., months), (F) time to progression, and (G) radiographic time to progression.

83. The method of claim 81, wherein the quantitative assessment of the one or more prostate cancer lesions determined in step (b) comprises one or more characteristics of PSMA expression in the prostate, and wherein the prognosis of the individual determined in step (c) comprises the risk of simultaneous (synchronous) metastasis and / or the risk of future (metachronous) metastasis.

84. A method for automatically determining from a plurality of medical images of the individual [e.g., one or more PSMA PET images (PET images obtained when a PSMA-targeted compound is administered to the individual) and / or one or more anatomical (e.g., CT) images] that an individual with prostate cancer is responsive to treatment, the method comprising: (a) receiving and / or accessing, by a processor of a computing device, a plurality of images of the individual, wherein at least a first image of the plurality of images is obtained before administration of the treatment and at least a second image of the plurality of images is obtained after administration of the treatment (e.g., after a period of time); (b) automatically determining, by the processor, from the image, a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) a molecular imaging TNM (miTNM) lesion type classification (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) of localized (T), pelvic node (N), and / or extrapelvic (M) disease; (ii) an indication of lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) a standard physiological uptake value (SUV) (e.g., SUV maximum, SUV peak, SUV mean); (iv) total lesion volume; (v) a change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) a calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein) (e.g., wherein the quantitative assessment comprises a response evaluation criteria in PSMA imaging (Response Evaluation Criteria in PSMA-imaging, RECIP) criteria and / or PSMA PET progression (PPP) criteria); and (c) automatically determining from the quantitative assessment in (b) whether the individual responds to the treatment (e.g., responder / non-responder) and / or the extent to which the individual responds to the treatment (e.g., numerical value or classification).

85. A method of automatically identifying whether an individual with prostate cancer (e.g., metastatic prostate cancer) will benefit from a particular treatment for prostate cancer using a plurality of medical images of the individual [e.g., one or more PSMA PET images (PET images obtained when a targeted PSMA compound is administered to the individual) and / or one or more anatomical (e.g., CT) images], the method comprising: (a) receiving and / or accessing, by a processor of a computing device, a plurality of images of the individual; (b) automatically determining, by the processor, from the image, a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) a molecular imaging TNM (miTNM) lesion type classification (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) of localized (T), pelvic node (N), and / or extrapelvic (M) disease; (ii) an indication of lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) a standard physiological uptake value (SUV) (e.g., SUV maximum, SUV peak, SUV mean); (iv) total lesion volume; (v) a change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) a calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein) (e.g., wherein the quantitative assessment comprises a response assessment criteria in the PSMA imaging (RECIP) criteria and / or the PSMA PET progression (PPP) criteria); and (c) automatically determining from the quantitative assessment in (b) whether the individual would benefit from a particular treatment for prostate cancer [e.g., determining whether the individual is likely to benefit from one or more particular treatments and / or classes of treatments, such as a particular radioligand therapy, such as leuprolide eligibility score].

86. A system for automatically processing a 3D image of a subject to determine one or more patient indicator values ​​measuring the subject's (e.g., overall) disease burden and / or risk, the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receiving a 3D functional image of the individual obtained using a functional imaging modality; (b) segmenting a plurality of 3D hotspot volumes within the 3D functional image, each 3D hotspot volume corresponding to a local region having an increased intensity relative to its surroundings and representing a potential cancerous lesion within the individual, thereby obtaining a set of 3D hotspot volumes; (c) for each specific metric of the one or more individual hotspot quantification metrics, calculating a value of the specific individual hotspot quantification metric for each 3D hotspot volume of the set; and (d) determining values ​​of the one or more patient indicators, wherein at least a portion of the patient indicators are each associated with one or more specific individual hotspot quantification metrics and are a function of at least a portion (e.g., substantially all; e.g., a specific subset) of the one or more specific individual hotspot quantification metric values ​​calculated for the 3D hotspot volume set.

87. A system for automatically analyzing a time series of medical images [e.g., three-dimensional images, e.g., nuclear medicine images (e.g., bone scan (scintigraphy), PET and / or SPECT), e.g., anatomical images (e.g., CT, X-ray, MRI), e.g., combined nuclear medicine and anatomical images (e.g., overlays)] of an individual, the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receiving and / or accessing a time series of medical images of the individual; and (b) identifying a plurality of hot spots within each of the medical images, and determining, by the processor, one, two or all three of (i), (ii) and (iii): (i) a change in the number of identified lesions, (ii) a change in the overall volume of the identified lesions (e.g., a change in the sum of the volumes of each identified lesion), and (iii) a change in the total volume weighted by PSMA (e.g., a lesion index) (e.g., the sum of the product of the lesion index and the lesion volume for all lesions in the region of interest) [e.g., wherein the changes identified in step (b) are used to identify (1) disease status [e.g., progression, regression or no change], (2) make a treatment management decision [e.g., active surveillance, prostatectomy, anti-androgen therapy, prednisone, radiation, radiotherapy, radioPSMA therapy or chemotherapy], or (3) treatment efficacy (e.g., wherein the individual has started treatment or has continued treatment with a drug or other therapy according to an initial set of images in the time series of medical images)] [e.g., wherein step (b) comprises using a machine learning module / model].

88. A system for analyzing a plurality of medical images of an individual (e.g., to assess disease status and / or progression in the individual), the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receiving and / or accessing a plurality of medical images of the individual and obtaining, by the processor, a plurality of 3D heat maps, each corresponding to a particular medical image (of the plurality of medical images) and identifying one or more hot spots within the particular medical image (e.g., representing possible underlying body lesions within the individual); (b) for each specific image (medical image) of the plurality of medical images, determining a corresponding 3D anatomical segmentation map using a machine learning module [e.g., a deep learning network (e.g., a convolutional neural network (CNN))], which identifies a set of organ regions within the specific medical image [e.g., representing soft tissue and / or bone structures within the individual (e.g., one or more cervical vertebrae; thoracic vertebrae; lumbar vertebrae; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and mandible)], thereby generating a plurality of 3D anatomical segmentation maps; (c) using (i) the plurality of 3D hotspot maps and (ii) the plurality of 3D anatomical segmentation maps to determine identification of one or more lesion correspondences, each (lesion correspondence) identifying two or more corresponding hotspots within different medical images and determined (e.g., by the processor) to represent the same underlying physical lesion within the individual; and (d) determining one or more metrics {e.g., one or more hotspot quantification metrics and / or changes therein [e.g., quantifying the characteristics of individual hotspots and / or the potential body lesions they represent (e.g., over time / between multiple medical images), such as changes in volume, radiopharmaceutical absorption, shape, etc.] based on the multiple 3D hotspot maps and the identification of the one or more lesions corresponding thereto; such as patient indicators (e.g., measuring an individual's overall disease burden and / or condition and / or risk) and / or changes therein; such as values ​​for classifying a patient (e.g., belonging to and / or suffering from a specific disease condition, progression, etc. category), such as prognostic metrics [e.g., indicating and / or quantifying the likelihood (e.g., overall survival) of one or more clinical outcomes (e.g., disease condition, progression, possible survival, treatment efficacy, etc.); such as values ​​for predictive metrics (e.g., indicating predicted response to therapy and / or other clinical outcomes)} based on the identification of the multiple 3D hotspot maps and the one or more lesions corresponding thereto.

89. A system for analyzing a plurality of medical images of an individual, the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) obtaining (e.g., receiving and / or accessing, and / or generating) a first 3D heat map of the individual; (b) obtaining (e.g., receiving and / or accessing, and / or generating) a first 3D anatomical segmentation map associated with the first 3D heat map; (c) obtaining (e.g., receiving and / or accessing, and / or generating) a second 3D heat map of the individual; (d) obtaining (e.g., receiving and / or accessing, and / or generating) a second 3D anatomical segmentation map associated with the second 3D heat map; (e) determining a registration field (e.g., 3D registration field and / or point-by-point registration) using / based on the first 3D anatomical segmented map and the second 3D anatomical segmented map; (f) registering the first 3D heat map with the second 3D heat map using the registration field, thereby generating a co-registered pair of 3D heat maps; (g) determining the identity of one or more lesions using co-registration of the 3D heat map; and (h) storing and / or providing the identification corresponding to the one or more lesions for display and / or further processing.

90. A system for analyzing a plurality of medical images of an individual (e.g., to assess disease status and / or progression in the individual), the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receiving and / or accessing a plurality of medical images of the individual; (b) for each specific image (medical image) of the plurality of medical images, determining, using a machine learning module [e.g., a deep learning network (e.g., a convolutional neural network (CNN))], a corresponding 3D anatomical segmentation map that identifies a set of organ regions within the specific medical image [e.g., representing soft tissue and / or bone structures within the individual (e.g., one or more cervical vertebrae; thoracic vertebrae; lumbar vertebrae; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain and mandible)], thereby generating multiple 3D anatomical segmented images; (c) determining one or more registration fields (e.g., full 3D registration fields; e.g., point-by-point registration) using the plurality of 3D anatomical segmentation maps and applying the one or more registration fields to register the plurality of medical images, thereby generating a plurality of registered medical images; (d) for each particular image in the plurality of registered medical images, determining a corresponding registered 3D heat map identifying one or more hot spots (e.g., representing possible underlying physical lesions in the individual) within the particular registered medical image, thereby generating a plurality of registered 3D heat maps; (e) determining, using the plurality of 3D registered heat maps, identifications of one or more lesion correspondences, each (lesion correspondence) identifying two or more corresponding hot spots within different medical images and determined (e.g., by the processor) to represent the same underlying body lesion within the individual; and (f) determining one or more metrics {e.g., one or more hotspot quantification metrics and / or changes therein [e.g., quantifying the characteristics of individual hotspots and / or the potential body lesions they represent (e.g., over time / between multiple medical images), such as changes in volume, radiopharmaceutical absorption, shape, etc.] based on the multiple 3D hotspot maps and the identification of the one or more lesions corresponding thereto; such as patient indicators (e.g., measuring an individual's overall disease burden and / or condition and / or risk) and / or changes therein; such as values ​​for classifying a patient (e.g., belonging to and / or suffering from a specific disease condition, progression, etc.), such as prognostic metrics [e.g., indicating and / or quantifying one or more clinical outcomes (e.g., disease condition, progression, possible survival, treatment efficacy, etc.) (e.g., overall survival); such as values ​​for predictive metrics (e.g., indicating predicted response to therapy and / or other clinical outcomes)} based on the multiple 3D hotspot maps and the identification of the one or more lesions corresponding thereto; 91. A system for analyzing a plurality of medical images of an individual, the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) obtaining (e.g., receiving and / or accessing, and / or generating) a first 3D anatomical image (e.g., CT, X-ray, MRI, etc.) and a first 3D functional image [e.g., a nuclear medicine image (e.g., PET, SPECT, etc.)] of the individual; (b) obtaining (e.g., receiving and / or accessing, and / or generating) a second 3D anatomical image and a second 3D functional image of the individual; (c) obtaining (e.g., receiving and / or accessing, and / or generating) a first 3D anatomical segmentation map based on (e.g., using) the first 3D anatomical image; (d) obtaining (e.g., receiving and / or accessing, and / or generating) a second 3D anatomical segmentation map based on (e.g., using) the second 3D anatomical image; (e) determining a registration field (e.g., 3D registration field and / or point-by-point registration) using / based on the first 3D anatomical segmented map and the second 3D anatomical segmented map; (f) registering (aligning) the second 3D functional image with the first 3D functional image using the 3D registration field, thereby producing a registered version of the second 3D functional image; (g) obtaining a first 3D heat map associated with the first functional image; (h) determining a second 3D heat map using the registered version of the second 3D functional image, whereby the second 3D heat map is registered with the first 3D heat map; (i) determining the identity of one or more lesions using the first 3D heat map and the second 3D heat map registered therewith; and (j) storing and / or providing the identification corresponding to the one or more lesions for display and / or further processing.

92. A system for automatically or semi-automatically systemically evaluating an individual with metastatic prostate cancer [e.g., metastatic castration-resistant prostate cancer (mCRPC) or metastatic hormone-sensitive prostate cancer (mHSPC)] to assess disease progression and / or treatment efficacy, the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receiving a first positron emission tomography (PET) image of the individual that targets prostate specific membrane antigen (PSMA) (a first PSMA-PET image) and a first 3D anatomical image [e.g., a computed tomography (CT) image; e.g., a magnetic resonance image (MRI)] of the individual, wherein the first 3D anatomical image of the individual was acquired simultaneously with, immediately after, or immediately before (e.g., on the same date as) the first PSMA PET image, such that the first 3D anatomical image and the first PSMA PET image correspond to a first date, and wherein the images depict a sufficiently large area of ​​the individual's body to encompass areas of the body to which metastatic prostate cancer has spread (e.g., a full torso image or a full body image encompassing multiple organs) {e.g., wherein the PSMA-PET image was acquired using F-18piflufolastatPSMA (i.e., 2-(3-{1-carboxy-5-[(6-[18F]fluoro-pyridine-3-carbonyl)amino]-pentyl}ureido)-glutaric acid, also known as [18F]F-DCFPyL), or Ga-68 PSMA-11, or other radiolabeled prostate specific membrane antigen inhibitor imaging agents}; (b) receiving a second PSMA-PET image of the individual and a second 3D anatomical image of the individual, both obtained on a second date after the first date; (c) automatically determining a registration field (e.g., a full 3D registration field; e.g., point-by-point registration) using automatically identified landmarks within the first 3D anatomical image and the second 3D anatomical image (e.g., the identified regions represent one or more of the cervical spine; thoracic spine; lumbar spine; left and right hip bones, sacrum, and coccyx; left ribs and left scapula; right ribs and right scapula; left femur; right femur; skull, brain, and mandible), and aligning the first PSMA-PET image and the second PSMA-PET image using the determined registration field, by the processor [e.g., before or after segmenting the CT and / or PSMA-PET images to identify boundaries of organs and / or bones, and before or after automatic hotspot (e.g., lesion) detection from the PSMA-PET images]; and (d) using the first PSMA-PET image and the second PSMA-PET image thus aligned to automatically detect (e.g., stage and / or quantify) a change (e.g., progression or remission) in the disease from the first date to the second date [e.g., automatically identifying and / or identifying as such (e.g., marking, labeling) one or both of the following (i) and (ii): (i) a change in the number of lesions {e.g., one or more new lesions (e.g., organ-specific lesions), or elimination of one or more lesions (e.g., organ-specific)}, and (ii) a change in tumor size {e.g., an increase in tumor size (PSMA-VOL increase / decrease), e.g., total tumor size, or a decrease in tumor size (PSMA-VOL decrease)} {e.g., a change in the volume of each of one or more specific lesions, or a change in the overall volume of a specific type of lesion (e.g., organ-specific tumor), or a change in the total volume of identified lesions}.

93. A system for quantifying and reporting disease (e.g., tumor) burden in a patient having cancer and / or at risk for cancer, the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) obtaining a medical image of the patient; (b) detecting one or more (e.g., multiple) hot spots within the medical image, each hot spot within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein voxels of the 3D hot spot volume have elevated intensity relative to their surroundings (e.g., and / or other indication or increased radiopharmaceutical uptake)] and representing a possible underlying physical pathology within the individual; (c) for each specific lesion category of a plurality of lesion categories representing specific tissue regions and / or lesion subtypes: identifying a corresponding subset of the one or more hot spots as belonging to the particular lesion class (e.g., based on a determination by the processor that the hot spots represent potential body lesions that are located within a particular tissue region and / or that belong to a particular lesion subtype represented by the particular lesion class); and determining values ​​of one or more patient indicators quantifying disease (e.g., tumor) burden within and / or associated with the particular lesion class based on the corresponding subset of the hot spots; and (d) presenting a graphical representation of the patient indicator value calculated for each of the plurality of lesion categories (e.g., a summary table listing each lesion category and the patient indicator value calculated for each lesion category), thereby providing a user with a graphical report summarizing tumor burden within a particular tissue region and / or associated with a particular lesion subtype.

94. A system for characterizing and reporting detected individual lesions based on imaging assessment of a patient having cancer and / or at risk for cancer, the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) obtaining a medical image of the patient; (b) detecting a set of one or more (e.g., a plurality of) hot spots within the medical image, each hot spot of the set within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein voxels of the 3D hot spot volume have elevated intensity relative to their surroundings (e.g., and / or other indication or increased radiopharmaceutical uptake)] and representing a possible underlying physical pathology within the individual; (c) assigning one or more lesion category labels to each of the one or more hotspots in the set, each lesion category label class representing a specific tissue region and / or lesion subtype and identifying the hotspot as representing a potential lesion located in the specific tissue region and / or belonging to the lesion subtype; (d) calculating, for each specific metric in the one or more individual hotspot quantification metrics, a value of a specific individual hotspot quantification metric for each hotspot of the set; and (e) displaying a graphical representation comprising, for each particular hotspot of at least a portion of the set of hotspots, an identification of the particular hotspot (e.g., a row in a table, and optionally an alphanumeric identification, such as an identification of the number of the particular hotspot), and one or more lesion category labels assigned to the particular hotspot and values ​​of the one or more individual hotspot quantitative metrics calculated for the particular hotspot [e.g., an overview table (e.g., a scrollable overview table) listing each hotspot in a row and the assigned lesion categories and hotspot quantitative metrics in columns].

95. A system for quantifying and reporting disease (e.g., tumor) progression and / or risk over time in a patient having cancer and / or at risk of cancer, the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) obtaining a plurality of medical images of the patient, each medical image representing a scan of the patient obtained at a particular time (e.g., a longitudinal dataset); (b) for each particular image of the plurality of medical images, detecting a corresponding set of one or more (e.g., multiple) hot spots within the particular medical image, each hot spot within the medical image corresponding to (e.g., being or comprising) a particular 3D volume [e.g., a 3D hot spot volume; e.g., wherein voxels of the 3D hot spot volume have elevated intensity relative to their surroundings (e.g., and / or other indication or increased radiopharmaceutical uptake)] and representing a possible underlying body pathology within the individual; (c) for each specific indicator of the one or more (e.g., overall) patient indicators that measure (e.g., quantify) the overall disease (e.g., tumor) load in the patient at a specific time, determining a value of the specific (e.g., overall) patient indicator for each specific medical image of the plurality of medical images based on the corresponding set of hot spots detected for the specific medical image, thereby determining a set of values ​​for each specific indicator of the one or more patient indicators that tracks changes in disease load by measuring the specific patient indicator values ​​over time; and (d) displaying a graphical representation of a set of values ​​of at least a portion (eg, a specific one, a specific subset) of the one or more patient indicator values, thereby conveying a measure of disease progression of the patient over time.

96. A system for automatically processing a 3D image of a subject to determine one or more patient indicator values ​​measuring the subject's (e.g., overall) disease burden and / or risk, the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receiving a 3D functional image of the individual obtained using a functional imaging modality; (b) segmenting a plurality of 3D hotspot volumes within the 3D functional image, each 3D hotspot volume corresponding to a local region having an increased intensity relative to its surrounding and representing a potential cancerous lesion within the individual, thereby obtaining a set of 3D hotspot volumes; (c) for each specific metric of the one or more individual hotspot quantification metrics, calculating a value of the specific individual hotspot quantification metric for each 3D hotspot volume of the set, wherein for a specific individual 3D hotspot volume, each hotspot quantification metric quantifies a characteristic (e.g., intensity, volume, etc.) of the specific 3D hotspot volume and is (e.g., calculated as) a specific function of the intensity and / or number of individual voxels within the specific 3D hotspot volume; and (d) determining values ​​of the one or more patient indicators, wherein at least a portion of the patient indicators are each associated with one or more specific individual hotspot quantification metrics and are calculated using a (e.g., the same) specific function of the intensity and / or number of stereo pixels within a combined hotspot volume, wherein the combined hotspot volume includes at least a portion (e.g., substantially all; e.g., a specific subset) of a set of 3D hotspot volumes (e.g., forming a union thereof).

97. A system for automatically determining the prognosis of an individual with prostate cancer from one or more medical images of the individual [e.g., one or more PSMA PET images (PET images obtained when a targeted PSMA compound is administered to the individual) and / or one or more anatomical (e.g., CT) images], the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receiving and / or accessing one or more images of the individual; (b) automatically determining from the one or more images a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) molecular imaging TNM (miTNM) lesion type classifications (e.g., miT, miN, miMa (lymph), miMb (bone), miMc (other)) of localized (T), pelvic node (N), and / or extrapelvic (M) disease; (ii) indicating lesion location (e.g., prostate, ilium, pelvic bone, rib cage, etc.); (iii) standard uptake value (SUV) (e.g., SUV maximum, SUV peak, SUV mean); (iv) total lesion volume; (v) change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein); and (c) automatically determining a prognosis for the individual from the quantitative assessment in (b), wherein the prognosis comprises one or more of the following for the individual: (I) expected survival (e.g., number of months), (II) expected time to disease progression, (III) expected time to radiographic progression, (IV) risk of simultaneous (synchronous) metastasis, and (V) risk of future (metachronous) metastasis.

98. A system according to claim 97, wherein the quantitative assessment of the one or more prostate cancer lesions determined in step (b) comprises one or more of the following: (A) total tumor volume, (B) change in tumor volume, (C) total SUV and (D) PSMA score, and wherein the prognosis of the individual determined in step (c) comprises one or more of the following: (E) expected survival (e.g., months), (F) time to progression and (G) radiographic time to progression.

99. A system according to claim 97, wherein the quantitative assessment of the one or more prostate cancer lesions determined in step (b) comprises one or more characteristics of PSMA expression in the prostate, and wherein the prognosis of the individual determined in step (c) comprises the risk of simultaneous (synchronous) metastasis and / or the risk of future (metachronous) metastasis.

100. A system for automatically determining a response to treatment in an individual with prostate cancer from a plurality of medical images of the individual [e.g., one or more PSMA PET images (PET images obtained when a targeted PSMA compound is administered to the individual) and / or one or more anatomical (e.g., CT) images], the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receiving and / or accessing, by a processor of a computing device, a plurality of images of the individual, wherein at least a first image of the plurality of images is obtained before administration of the treatment and at least a second image of the plurality of images is obtained after administration of the treatment (e.g., after a period of time); (b) automatically determining from the image a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) molecular imaging TNM (miTNM) lesion type classification (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) of localized (T), pelvic nodal (N), and / or extrapelvic (M) disease; (ii) indication of lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) standard physiological uptake value (SUV) (e.g., SUV maximum, SUV peak, SUV mean); (iv) total lesion volume; (v) change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein) (e.g., wherein the quantitative assessment is included in the response assessment criteria of PSMA imaging (RECIP) criteria and / or PSMA PET progression (PPP) criteria); and (c) automatically determining from the quantitative assessment in (b) whether the individual responds to the treatment (e.g., responds / does not respond) and / or the extent to which the individual responds to the treatment (e.g., numerical value or classification).

101. A system for automatically identifying whether an individual with prostate cancer (e.g., metastatic prostate cancer) will benefit from a particular treatment for prostate cancer using a plurality of medical images of the individual [e.g., one or more PSMA PET images (PET images obtained while a targeted PSMA compound is being administered to the individual) and / or one or more anatomical (e.g., CT) images], the system comprising: a processor of the computing device; and A memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to: (a) receiving and / or accessing a plurality of images of the individual; (b) automatically determining from the image a quantitative assessment of one or more prostate cancer lesions (e.g., metastatic prostate cancer lesions) [e.g., wherein the quantitative assessment comprises one or more members selected from the group consisting of: (i) molecular imaging TNM (miTNM) lesion type classification (e.g., miT, miN, miMa (lymphatic), miMb (bone), miMc (other)) of localized (T), pelvic nodal (N), and / or extrapelvic (M) disease; (ii) indication of lesion location (e.g., prostate, iliac, pelvic bone, rib cage, etc.); (iii) standard physiological uptake value (SUV) (e.g., SUV maximum, SUV peak, SUV mean); (iv) total lesion volume; (v) change in lesion volume (e.g., individual lesions and / or total lesions); and (vi) calculated PSMA (aPSMA) score] (e.g., using one or more of the methods described herein) (e.g., wherein the quantitative assessment is included in the response assessment criteria of PSMA imaging (RECIP) criteria and / or PSMA PET progression (PPP) criteria); and (c) automatically determining from the quantitative assessment in (b) whether the individual would benefit from a specific treatment for prostate cancer [e.g., determining one or more specific treatments and / or classes of treatments for the individual, such as a specific radioligand therapy, such as leuprolide eligibility score].

102. A therapeutic agent for treating (e.g., through multiple cycles of the therapeutic agent) an individual having and / or at risk of a particular disease (e.g., prostate cancer (e.g., metastatic castration-resistant prostate cancer)), wherein the individual has (i) been administered a first cycle of the therapeutic agent and imaged (e.g., before and / or during and / or after the first cycle of the therapeutic agent), and (ii) has been identified as a responder to the therapeutic agent using a method according to any one of claims 1 to 52 (e.g., the individual has been identified / classified as a responder based on the values ​​of the one or more risk indicators determined using a method according to any one of claims 1 to 52).

103. A second (e.g., second-line) therapeutic agent for treating an individual having and / or at risk for a particular disease (e.g., prostate cancer (e.g., metastatic castration-resistant prostate cancer)), who has (i) been administered an initial, cycle of a first therapeutic agent and imaged (e.g., before and / or during and / or after a cycle of the first therapeutic agent), and (ii) has been identified as a non-responder to the first therapeutic agent using the method of any one of claims 1 to 52 (e.g., based on the values ​​of the one or more risk indicators determined using the method of any one of claims 1 to 52, the individual has been identified / classified as a non-responder) (e.g., thereby subjecting the individual to a potentially more effective therapy).

Images