Systems and methods for interactive display of a medical image

The interactive display of medical images with color-coded cancerous lesions and user-adjustable filters addresses inefficiencies in cancer diagnosis by enhancing lesion identification and assessment, improving diagnostic accuracy and treatment planning.

WO2026241391A1PCT designated stage Publication Date: 2026-11-19EXINI DIAGNOSTICS +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/US2026/027997
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-14
Filing Date
2026-05-13
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

The existing processes for reviewing and analyzing multiple patient images for cancer diagnosis and treatment are inefficient and lack tools for improved accuracy and efficiency in identifying and assessing cancerous lesions.

Method used

Systems and methods for interactive display of medical images, such as 3D PET, CT, or PET/CT images, with graphical overlays that color-code suspected cancerous lesions (hotspots) and allow user-adjustable filtering based on criteria like SUV, lesion size, and lesion index, using supervised deep learning neural networks.

Benefits of technology

Enhances the ability to quickly and accurately identify and assess the severity of cancerous lesions by providing clear visual contrast and allowing users to filter and emphasize relevant hotspots, improving diagnostic accuracy and treatment planning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2026027997_19112026_PF_FP_ABST
    Figure US2026027997_19112026_PF_FP_ABST
Patent Text Reader

Abstract

Presented herein are systems and methods for the interactive display of a medical image, such as a 3D PET image, CT image, or PET / CT image, with a graphical overlay identifying regions of suspected cancerous lesions (hotspots), e.g., prostate cancer.
Need to check novelty before this filing date? Find Prior Art

Description

Attorney’s Docket No. 2010358-0388SYSTEMS AND METHODS FOR INTERACTIVE DISPLAY OF A MEDICAL IMAGECROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to and benefit of U. S. Provisional Application No. 63 / 805,641, filed May 14, 2025, the content of which is hereby incorporated by reference herein in its entirety.TECHNICAL FIELD

[0002] This invention relates generally to systems and methods for the interactive display of a medical image such as a 3D positron emission tomography (PET) image, a computed tomography (CT) image, or a 3D composite PET / CT image (positron emission tomography / computed tomography (CT) image).BACKGROUND

[0003] Nuclear medicine imaging involves the use of radiolabeled compounds, referred to as radiopharmaceuticals. Radiopharmaceuticals are administered to patients and accumulate in various regions in the body in manner that depends on, and is therefore indicative of, biophysical and / or biochemical properties of tissue therein, such as those influenced by presence and / or state of disease, such as cancer. For example, certain radiopharmaceuticals, following administration to a patient, accumulate in regions of abnormal osteogenesis associated with malignant bone lesions, which are indicative of metastases. Other radiopharmaceuticals may bind to specific receptors, enzymes, and proteins in the body that are altered during evolution of disease. After administration to a patient, these molecules circulate in the blood until they find their intended target. The bound radiopharmaceutical remains at the site of disease, while the rest of the agent clears from the body.Attorney’s Docket No. 2010358-0388

[0004] Nuclear medicine imaging techniques capture images by detecting radiation emitted from the radioactive portion of the radiopharmaceutical. The accumulated radiopharmaceutical serves as a beacon so that an image may be obtained depicting the disease location and concentration using commonly available nuclear medicine modalities. Examples of nuclear medicine imaging modalities include bone scan imaging (also referred to 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 throughout the world. Choice of a particular imaging modality depends on and / or dictates the particular radiopharmaceutical used. For example, technetium 99m (99mTc) labeled compounds are compatible with bone scan imaging and SPECT imaging, while PET imaging often uses fluorinated compounds labeled with 18F. The compound "mTc methylenediphosphonate (99niTc MDP) is a popular radiopharmaceutical used for bone scan imaging in order to detect metastatic cancer. Radiolabeled prostate-specific membrane antigen (PSMA) targeting compounds such as "mTc labeled 1404 and PyL™ (also referred to as [18F]DCFPyL) can be used with SPECT and PET imaging, respectively, and offer the potential for highly specific prostate cancer detection.

[0005] Accordingly, nuclear medicine imaging is a valuable technique for providing physicians with information that can be used to determine the presence and the extent of disease in a patient. The physician can use this information to provide a recommended course of treatment to the patient and to track the progression of disease.

[0006] For example, an oncologist may use nuclear medicine images from a study of a patient as input in her assessment of whether the patient has a particular disease, e.g., prostate cancer, what stage of the disease is evident, what the recommended course of treatment (if any) would be, whether surgical intervention is indicated, and likely prognosis. The oncologist may use aAttorney’s Docket No. 2010358-0388radiologist report in this assessment. A radiologist report is a technical evaluation of the nuclear medicine images prepared by a radiologist for a physician who requested the imaging study and includes, for example, the type of study performed, the clinical history, a comparison between images, the technique used to perform the study, the radiologist’s observations and findings, as well as overall impressions and recommendations the radiologist may have based on the imaging study results. A signed radiologist report is sent to the physician ordering the study for the physician’s review, followed by a discussion between the physician and patient about the results and recommendations for treatment.

[0007] Thus, the process involves having a radiologist perform an imaging study on the patient, analyzing the images obtained, creating a radiologist report, forwarding the report to the requesting physician, having the physician formulate an assessment and treatment recommendation, and having the physician communicate the results, recommendations, and risks to the patient. The process may also involve repeating the imaging study due to inconclusive results, or ordering further tests based on initial results. If an imaging study shows that the patient has a particular disease or condition (e.g., cancer), the physician discusses various treatment options, including surgery, as well as risks of doing nothing or adopting a watchful waiting or active surveillance approach, rather than having surgery.

[0008] Accordingly, the process of reviewing and analyzing multiple patient images, over time, plays a critical role in the diagnosis and treatment of cancer. There is a significant need for improved tools that facilitate and improve accuracy of image review and analysis for cancer diagnosis and treatment. Improving the toolkit utilized by physicians, radiologists, and other healthcare professionals provides for significant improvements in standard of care and patient experience.Attorney’s Docket No. 2010358-0388SUMMARY

[0009] Presented herein are systems and methods for the interactive display of a medical image, such as a 3D PET image, CT image, or PET / CT image, with a graphical overlay identifying regions of suspected cancerous lesions (hotspots), e.g., prostate cancer.

[0010] In certain embodiments, hotspots are color-coded based on lesion type, e g., TNM type (tumor, node, and metastasis type). The TNM types may include, for example, miT (local, e.g., within a prostate), miN (pelvil lymph node), miMa (distant lymph node), miMb (bone), and / or miMc (distant organ). In certain embodiments, the hotspots are identified and / or characterized (e.g., by TNM type) using a trained, supervised deep learning neural network as applied to one or more anatomical (e.g., CT) images in combination with one or more functional images (e.g., PET and / or single photon emission computed tomography (SPECT) images).

[0011] In certain embodiments, a medical image displayed via systems and methods described herein comprises a depiction of an anatomy of a subject. The medical image may be rendered in the interactive display such that depiction of the subject’s anatomy is shown in black-and-white (B& W), grayscale or otherwise monochrome. A graphical overlay may sumperimpose color renderings of the hotspots on the depiction of the anatomy. Among other things, sumperimposing colorized hotpots on a B& W, grayscale, or monochrome depiction of subject’s anaotomy can aid a user, such as a radiologist or physician, in identifying and locating cancerous lesions within the subject. Among other things, providing contrast between hotspots and the depiction of anatomy in the manner described above may aid the user to clearly and rapidly distinguish between hotspots and anatomical features.

[0012] In certain embodiments, the systems and methods receive user input via one or more graphical control elements (widgets) indicating a setting or adjustment of one or more thresholdsAttorney’s Docket No. 2010358-0388for application a graphical display filter to adjust the appearance of the color-coded overlay. For example, in certain embodiments, the filter criteria may be a range, a minimum, and / or a maximum of one or more of the following measurements associated with each of the hotspots: Standardized Uptake Value (SUV), for example, SUVmean, SUVmax, and / or SUVpeak; lesion size (e.g., volume); and / or a Lesion Index (LI).

[0013] Upon application of the filter, the color of one or more hotspots depicted in the overlay are adjusted, for example, to visually highlight or otherwise graphically emphasize one or more hotspots that satisfy the filter critera and / or to visually / graphically de-emphasize one or more hotspots that do not satisfy the filter criteria. For example, in certain embodiments, hotspots that do not meet the filter criteria are visually de-emphasized / dimmed in the overlay such that they have a different, e.g., darker hue (color) than the original hue. Alternatively, rather than changing the hue (color), application of the filter visually de-emphasizes hotspots in the overlay by rendering those hotspots with a different shade (e.g., an amount of black added to the original hue), a different tone (e.g., an amount of gray added to the original hue), or a different tint (e.g., an amount of white added to the original hue). In certain embodiments, the emphasized or de-emphasized hotspots are still recognizable by type via the color-coding shown in the legend. Providing user control and filtering of hotspots in this manner facilitates contextualizing and assessing severity of hotspots and / or their clinical significance in a quick and convenient way. For example, assessing the clinical significance of one or more, or a collection of, hotspots appearing in a medical image may require a reviewer, such as a radiologist or physician, to evaluate and weigh multiple parameters, such as TNM type, anatomical context, uptake values, lesion index, etc. Accordingly, the ability to visually filter and / or color-codeAttorney’s Docket No. 2010358-0388graphical overlays comprising color renderings of hotspots based on criteria relating to different parameters may enhance the process of hotspot assessment by the user.

[0014] In one aspect, the invention is directed to a method for interactive display of a medical image of a subject, the method comprising: rendering, by a processor of a computing device, a first instantiation of the medical image (e.g., a PET image, a CT image, a PET / CT image, or a PET / SPECT image, e.g., a 3-dimensional image (3D image)) for graphical display, said first instantiation comprising a black and white (B& W) and / or grayscale or otherwise monochrome depiction of an anatomy of the subject with a first overlay superimposed on the depiction of the anatomy, said first overlay comprising a first color rendering of regions (e.g., 3D regions) corresponding to a set of hotspots (e.g., where hotspots are regions of suspected cancerous lesions, e.g., prostate cancer, e.g., metastatic prostate cancer); causing, by the processor, display of one or more graphical control elements (widgets) allowing for user selection (e.g., setting or adjustment) of one or more criteria for application of a graphical display filter to adjust the first instantiation of the nuclear medicine image; receiving, by the processor, via the one or more widgets, a user selection of the one or more criteria for applying the graphical display filter to adjust the first instantiation of the nuclear medicine image; and rendering, by the processor, a second instantiation of the nuclear medicine image, said second instantiation reflecting a product of the adjustment of the first instantiation of the nuclear medicine image following application of the graphical display filter, wherein, the second instantiation comprises the B& W and / or grayscale or otherwise monochrome depiction of the anatomy of the subject with a second overlay superimposed on the depiction of the anatomy, said second overlay comprising a second color rendering of the regions corresponding to the set ofAttorney’s Docket No. 2010358-0388hotspots depicted in the first overlay, wherein the second color rendering adjusts a color of one or more of the regions according to the one or more thresholds of the graphical display filter.

[0015] In certan embodiments, the set of hotspots are color-coded based on lesion type (e.g., TNM type (tumor, node, and metastasis type), e.g., miT (local, e.g., within a prostate), miN (pelvic lymph node), miMa (distant lymph node), miMb (bone), and / or miMc (distant organ)).

[0016] In certain embodiments, the one or more criteria for applying the graphical display filter comprises a range and / or a minimum and / or a maximum of one or more of the following measurements associated with each of the hotspots: (i) a Standardized Uptake Value (SUV), for example, SUVmean, SUVmax, and / or SUVpeak; (ii) a lesion size (e.g., volume); and / or (iii) a Lesion Index (LI).

[0017] In certain embodiments, upon application of the filter, the color of one or more hotspots depicted in the first overlay are adjusted to (i) visually highlight or otherwise graphically emphasize one or more hotspots that satisfy the filter critera and / or (ii) visually / graphically de-emphasize one or more hotspots that do not satisfy the filter criteria, thereby creating the second overlay. For example, in certain embodiments, hotspots that do not meet the filter criteria are visually de-emphasized / dimmed in the overlay such that they have a different, e.g., darker hue (color) than the original hue. Alternatively, rather than changing the hue (color), application of the filter visually de-emphasizes hotspots in the overlay by rendering those hotspots with a different shade (e.g., an amount of black added to the original hue), a different tone (e.g., an amount of gray added to the original hue), or a different tint (e g., an amount of white added to the original hue). In certain embodiments, the emphasized or deemphasized hotspots are still recognizable by type via the color-coding shown in the legend.Attorney’s Docket No. 2010358-0388

[0018] In certain embodiments, at least a portion of the hotspots of the first (and second) overlay are determined using a machine learning algorithm (e.g., a supervised deep learning neural network, e.g., a convolutional neural network).

[0019] In certain embodiments, at least a portion of the hotspots of the first (and second) overlay are user-defined (e.g., a user may draw a hotspot, e.g., for which SUV and / or size data are determined and upon which filtering criteria are applied).

[0020] In another aspect, the invention is directed to a system for interactive display of a medical image of a subject, 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 any of the methods described herein.

[0021] In another aspect, the invention is directed to a method for interactive display of a medical image of a subject, the method comprising: rendering, by a processor of a computing device, a first instantiation of the medical image (e.g., a PET image, a CT image, a PET / CT image, or a PET / SPECT image, e.g., a 3-dimensional image (3D image)) for graphical display, said first instantiation comprising a black-and-white (B& W) and / or grayscale or otherwise monochrome depiction of an anatomy of the subject with a first overlay superimposed on the depiction of the anatomy, said first overlay comprising a first color rendering of regions (e.g., 3D regions) corresponding to a set of hotspots (e.g., where hotspots are regions of suspected cancerous lesions, e.g., prostate cancer, e.g., metastatic prostate cancer); causing, by the processor, display of one or more graphical control elements (widgets) allowing for user selection (e.g., setting or adjustment) of one or more criteria for application of a graphical display filter to adjust the first instantiation of the medical image; receiving, by the processor, via the one or more widgets, a user selection of the one or more criteria for applying the graphicalAttorney’s Docket No. 2010358-0388display filter to adjust the first instantiation of the medical image; and rendering, by the processor, a second instantiation of the medical image, said second instantiation reflecting a product of the adjustment of the first instantiation of the medical image following application of the graphical display filter, wherein the second instantiation comprises the B& W and / or grayscale or otherwise monochrome depiction of the anatomy of the subject with a second overlay superimposed on the depiction of the anatomy, said second overlay comprising a second color rendering of the regions corresponding to the set of hotspots depicted in the first overlay.

[0022] In certan embodiments, (e.g., in the first color rendering and / or the second color rendering) the (e.g., regions corresponding to the) set of hotspots is color-coded based on lesion type [e.g., each of the regions of the first and / or second color rendering corresponds to a particular hotspot of the set of hotspots, each hotspot of the set of hotspots is assigned a particular lesion type, and each region in the first and / or second color rendering is color-coded based on the particular lesion type assigned to the corresponding particular hotspot],

[0023] In certan embodiments, the lesion type comprises a TNM type (tumor, node, and metastasis type).

[0024] In certan embodiments, each hotspot of the set of hotspots is assigned a particular lesion type and, within the first color rendering and / or the second color rendering, each of the regions is color-coded according to the lesion type assigned to the corresponding hotspot (e.g., a given region corresponds to a particular hotspot of the set and is color-coded to reflect the lesion type assigned to the particular hotspot).

[0025] In certan embodiments, the particular lesion type assigned to each hotspot is selected from a list of possible lesion types, said list of possible lesion types comprising one or moreAttorney’s Docket No. 2010358-0388members selected from the group consisting of: miT (local, e.g., within a prostate), miN (pelvic lymph node), miMa (distant lymph node), miMb (bone), and miMc (distant organ).

[0026] In certan embodiments, the particular lesion type assigned to each hotspot is selected from the group consisting of: miT (local, e.g., within a prostate), miN (pelvic lymph node), miMa (distant lymph node), miMb (bone), and miMc (distant organ).

[0027] In certan embodiments, the one or more criteria for applying the graphical display filter comprises a (e.g., adjustable) range of one or more measurements associated with each hotspot of the set of hotspots.

[0028] In certan embodiments, the one or more measurements comprise one or more members selected from the group consisting of: (i) a Standardized Uptake Value (SUV) (e.g., quantifying an intensity or SUV of each hotspot, for example, SUVmean, SUVmax, and / or SUVpeak), (ii) a lesion size (e.g., volume), and (iii) a Lesion Index (LI).

[0029] In certan embodiments, the one or more criteria for applying the graphical display filter comprises a (e.g., adjustable) maximum and / or a minimum of one or more measurements associated with each hotspot of the set of hotspots.

[0030] In certan embodiments, the one or more measurements comprise one or more members selected from the group consisting of: (i) a Standardized Uptake Value (SUV) (e.g., quantifying an intensity or SUV of each hotspot, for example, SUVmean, SUVmax, and / or SUVpeak), (ii) a lesion size (e.g., volume), and (iii) a Lesion Index (LI).

[0031] In certan embodiments, upon application of the filter, color of the regions corresponding to the set of hotspots depicted in the first overlay is adjusted to (i) visually highlight or otherwise graphically emphasize one or more hotspots out of the set of hotspots that satisfy the one or more criteria and / or (ii) visually / graphically de-emphasize one or moreAttorney’s Docket No. 2010358-0388hotspots out of the set of hotspots that do not satisfy the one or more criteria, thereby creating the second overlay.

[0032] In certan embodiments, at least a portion of the set of hotspots (e.g., of the first and / or second overlay) are determined using a machine learning algorithm.

[0033] In certan embodiments, the machine learning algorithm comprises a supervised deep learning neural network or a convolutional neural network.

[0034] In certan embodiments, at least a portion of the set of hotspots of the first (e.g., and / or second) overlay is user-defined (e.g., a user may draw a hotspot).

[0035] In certan embodiments, at least a part of the one or more criteria is determined based on the user-defined hotspots.

[0036] In certan embodiments, the medical image comprises a Positron Emission Tomography (PET) image.

[0037] In certan embodiments, the medical image comprises a computed tomography (CT) image.

[0038] In certan embodiments, the medical image comprises a PET / CT image or a SPECT / CT image.

[0039] In certan embodiments, the medical image comprises a composite image, said composite image comprising an anatomical image (e.g., a CT image) and a corresponding functional image (e.g., a nuclear medicine image (e.g., a PET image; e.g., a SPECT image)).

[0040] In certan embodiments, the medical image comprises a 3-dimensional image (3D image).

[0041] In certan embodiments, the hotspots are regions of suspected cancerous lesions (e.g., prostate cancer, metastatic prostate cancer).Attorney’s Docket No. 2010358-0388

[0042] In certan embodiments, the second color rendering adjusts a color of one or more of the regions according to the one or more criteria of the graphical display fdter.

[0043] In another aspect, the invention is directed to a system for interactive display of a medical image of a subject, 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: render a first instantiation of the medical image (e.g., a PET image, a CT image, a PET / CT image, or a PET / SPECT image, e.g., a 3-dimensional image (3D image)) for graphical display, said first instantiation comprising a black-and-white (B& W) and / or grayscale or otherwise monochrome depiction of an anatomy of the subject with a first overlay superimposed on the depiction of the anatomy, said first overlay comprising a first color rendering of regions (e.g., 3D regions) corresponding to a set of hotspots (e.g., where hotspots are regions of suspected cancerous lesions, e.g., prostate cancer, e.g., metastatic prostate cancer); cause display of one or more graphical control elements (widgets) allowing for user selection (e.g., setting or adjustment) of one or more criteria for application of a graphical display filter to adjust the first instantiation of the medical image; receive, via the one or more widgets, a user selection of the one or more criteria for applying the graphical display filter to adjust the first instantiation of the medical image; and render a second instantiation of the medical image, said second instantiation reflecting a product of the adjustment of the first instantiation of the medical image following application of the graphical display filter, wherein the second instantiation comprises the B& W and / or grayscale or otherwise monochrome depiction of the anatomy of the subject with a second overlay superimposed on the depiction of the anatomy, said second overlay comprising a second color rendering of the regions corresponding to the set of hotspots depicted in the first overlay.Attorney’s Docket No. 2010358-0388

[0044] Features of 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

[0045] 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 taken in conjunction with the accompanying drawings, in which:

[0046] FIG. l is a set of three images showing corresponding slices of a CT image, a PET image, and a PET / CT fusion, obtained from a 3D PET / CT scan, according to an illustrative embodiment.

[0047] FIG. 2 is a diagram illustrating an example process for segmenting an anatomical image and identifying anatomical boundaries in a co-aligned functional image, according to an illustrative embodiment.

[0048] FIG. 3 is a diagram illustrating an example process for segmenting and classifying hotspots, according to an illustrative embodiment.

[0049] FIG. 4 is a block flow diagram of an example process for interactive display of a medical image of a subject, according to an illustrative embodiment.

[0050] FIG. 5 is a schematic showing an approach for computing lesion index values, according to an illustrative embodiment.

[0051] FIG. 6 is a schematic showing another approach for computing lesion index values, according to an illustrative embodiment.Attorney’s Docket No. 2010358-0388

[0052] FIG. 7 is a block diagram of an exemplary cloud computing environment, according to an illustrative embodiment.

[0053] FIG. 8 is a block diagram of an example computing device and an example mobile computing device, according to an illustrative embodiment.

[0054] FIG. 9 is a screenshot of a graphical user interface depicting an enhanced hotspot outlining feature, according to an illustrative embodiment.

[0055] FIG. 10A is a screenshot of a graphical user interface (GUI) depicting an enhanced color-coded Maximum Intensity Projection (MIP) view of a medical image (here, a PET image showing the anatomy of the subject) with hotspot map overlay, shown before application of a fdter, according to an illustrative embodiment.

[0056] FIG. 10B is a screenshot of the GUI and enhanced color-coded MIP view of FIG.10 A, after application of the fdter, according to an illustrative embodiment.

[0057] FIG. 11 is a screenshot of a GUI depicting an enhanced color-coded MIP view of a medical image (here, a PET image showing the anatomy of the subject) with hotspot map overlay, shown before application of a fdter, according to an illustrative embodiment.

[0058] FIG. 12 is a screenshot of the GUI and MIP of FIG. 11, after application of the fdter, according to an illustrative embodiment.

[0059] FIG. 13-1 is a screenshot of a graphical user interface that permits parallel study review in a single window, according to an illustrative embodiment.

[0060] FIG. 13-2 is a screenshot of a graphical user interface that permits parallel study review in a single window, according to an illustrative embodiment.

[0061] FIG. 13-3 is a screenshot of a graphical user interface that permits parallel study review in a single window, according to an illustrative embodiment.Attorney’s Docket No. 2010358-0388

[0062] FIG. 13-4 is a screenshot of a graphical user interface that permits parallel study review in a single window, according to an illustrative embodiment.

[0063] FIG. 14-1 is a screenshot of a graphical user interface depicting improved hotspot segmentation, according to an illustrative embodiment.

[0064] FIG. 14-2 is a screenshot of a graphical user interface depicting improved hotspot segmentation, according to an illustrative embodiment.

[0065] FIG. 15 is a screenshot of a graphical user interface comprising a review toolbar, according to an illustrative embodiment.

[0066] The features and advantages of the present disclosure will become more apparent from the detailed description set forth below when taken in conjunction with the drawings, in which like reference characters identify corresponding elements throughout. In the drawings, like reference numbers generally indicate identical, functionally similar, and / or structurally similar elements.CERTAIN DEFINITIONS

[0067] In order for the present disclosure to be more readily understood, certain terms are first defined below. Additional definitions for the following terms and other terms are set forth throughout the specification.

[0068] A, an. The articles “a” and “an” are used herein to refer to one or to more than one (i.e., at least one) of the grammatical object of the article. By way of example, “an element” 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.Attorney’s Docket No. 2010358-0388

[0069] First, second, etc.: It should be understood that any reference to an element herein using a designation such as "first," "second," and so forth does not limit the quantity or order of those elements, unless such limitation is explicitly stated. Rather, these designations may be used herein as a convenient method of distinguishing between two or more elements or instances of an element. Thus, a reference to first and second elements does not mean that only two elements may be employed or that the first element must precede the second element in some manner. In addition, unless stated otherwise, a set of elements may comprise one or more elements.

[0070] Image". As used herein, an “image” - for example, a three-dimensional (3D) image of a subject, includes any visual representation, such as a photo, a video frame, streaming video, as well as any electronic, digital or mathematical analogue of a photo (e.g., a digital image), video frame, or streaming video, displayed or stored in memory (e.g., a digital image may, but need not be displayed for visual inspection). Any apparatus described herein, in certain embodiments, includes a display for displaying an image or any other result produced by the processor. Any method described herein, in certain embodiments, includes a step of displaying an image or any other result produced via the method. In certain embodiments, an image is a 3D image, conveying information that varies with position within a 3D volume. Such images may, for example, be represented digitally as a 3D matrix (e.g., a N x M x L matrix) with each voxel of a 3D image represented by an element of a 3D matrix. Other representations are also contemplated and included, for example, a 3D matrix may be reshaped as a vector (e.g., a 1 x K size vector, where K is a total number of voxels) by stitching each row or column end to end. Examples of images include, for example, medical images, such as bone-scan images (also referred to as scintigraphy images), computed tomography (CT) images, magnetic resonance images (MRIs), optical images (e.g., bright-field microscopy images, fluorescence images,Attorney’s Docket No. 2010358-0388reflection or transmission images, etc ), positron emission tomography (PET) images, single-photon emission tomography (SPECT) images, ultrasound images, x-ray images, and the like. In certain embodiments, a medical image is or comprises a nuclear medicine image, produced from radiation emitted from within a subject being imaged. In certain embodiments, a medical image is or comprises an anatomical image (e.g., a 3D anatomical image) conveying information regarding location and extent of anatomical structures such as internal organs, bones, soft-tissue, and blood vessels, within a subject. Examples of anatomical images include, without limitation, x-ray images, CT images, MRIs, and ultrasound images. In certain embodiments, a medical image is or comprises a functional image (e.g., a 3D functional image) conveying information relating to physiological activities within specific organs and / or tissue, such as metabolism, blood flow, regional chemical composition, absorption, etc. Examples of functional images include, without limitation, nuclear medicine images, such as PET images, SPECT images, as well as other functional imaging modalities, such as functional MRI (fMRI), which measures small changes in blood flow for use in assessing brain activity. 3D images may be rendered in two dimensions (2D) (e.g., for display on a screen) in any of a variety of ways, as understood by one of skill in the art.

[0071] Map: As used herein, the term “map” is understood to mean a visual display, or any data representation that may be interpreted for visual display, which contains spatially-correlated information. For example, a three-dimensional map of a given volume may include a dataset of values of a given quantity that varies in three spatial dimensions throughout the volume. A three-dimensional map may be displayed in two-dimensions (e.g., on a two-dimensional screen, or on a two-dimensional printout).Attorney’s Docket No. 2010358-0388

[0072] 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, allowing them to be individually and distinguishably accessed and operated upon and / or used for operating on, for example, one or more images.

[0073] 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 rendered as a dataset in three dimensions and / or may be displayed as a set 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.

[0074] Whole body. As used herein, the terms “full body” and “whole body” used (interchangeably) in the context of segmentation and other manners of identification of regions within an image of a subject refer to approaches that evaluate a majority (e.g., greater than 50%) of a graphical representation of a subject’s body in a 3D anatomical image to identify target tissue regions of interest. In certain embodiments, full body and whole body segmentation refers to identification of target tissue regions within at least an entire torso of a subject. In certain embodiments, portions of limbs are also included, along with a head of the subject

[0075] Radionuclide. As used herein, “radionuclide” refers to a moiety comprising a radioactive isotope of at least one element. Exemplary suitable radionuclides include but are not limited to those described herein. In some embodiments, a radionuclide is one used in positron emission tomography (PET). In some embodiments, a radionuclide is one used in single-photon emission computed tomography (SPECT). In some embodiments, a non-limiting list of radionuclides includes99mTc,111In,64Cu,67Ga,68Ga,186Re,188Re,153Sm,177Lu,67Cu,123I,124I,Attorney’s Docket No. 2010358-0388125I,126I,131I,nC,13N,150,18F,153Sm,166HO,177LU,149Pm,90Y,213Bi,103Pd,109Pd,159Gd,140La,198Au,199Au,169Yb,175Yb,165Dy,166Dy,105Rh,niAg,89Zr,225Ac,82Rb,75Br,76Br,77Br,80Br, 8°mBr, 82Br, 83Br, 211At and192If

[0076] Radiopharmaceutical. As used herein, the term “radiopharmaceutical” refers to a compound comprising a radionuclide. In certain embodiments, radiopharmaceuticals are used for diagnostic and / or therapeutic purposes. In certain embodiments, radiopharmaceuticals include small molecules that are labeled with one or more radionuclide(s), antibodies that are labeled with one or more radionuclide(s), and antigen-binding portions of antibodies that are labeled with one or more radionuclide(s).

[0077] Machine learning module: Certain embodiments described herein make use of (e.g., include) software instructions that include one or more machine learning module s), also referred to herein as artificial intelligence software. As used herein, the term “machine learning module refers to a computer implemented process (e.g., function) that implements one or more specific machine learning algorithms in order to determine, for a given input (such as an image (e.g., a 2D image; e.g., a 3D image), dataset, and the like) one or more output values. For example, a machine learning module may receive as input a 3D image of a subject (e.g., a CT image; e.g., an MRI), and for each voxel of the image, determine a value that represents a likelihood that the voxel lies within a region of the 3D image that corresponds to a representation of a particular organ or tissue of the subject. In certain embodiments, two or more machine learning modules may be combined and implemented as a single module and / or a single software application. In certain embodiments, two or more machine learning modules may also be implemented separately, e.g., as separate software applications. A machine learning module may be software and / or hardware. For example, a machine learning module may be implementedAttorney’s Docket No. 2010358-0388entirely as software, or certain functions of a convolutional neural network (CNN) module may be carried out via specialized hardware (e.g., via an application specific integrated circuit (ASIC)).

[0078] Subject: As used herein, a “subject” means a human or other mammal (e.g., rodent (mouse, rat, hamster), pig, cat, dog, horse, primate, rabbit, and the like). The term “subject” is used herein interchangeably with the term “patient”.

[0079] Administering; As used herein, “administering” an agent means introducing a substance (e.g., an imaging agent) into a subject. In general, any route of administration may be utilized including, for example, parenteral (e.g., intravenous), oral, topical, subcutaneous, peritoneal, intraarterial, inhalation, vaginal, rectal, nasal, introduction into the cerebrospinal fluid, or instillation into body compartments.

[0080] Tissue; As used herein, the term “tissue” refers to bone (osseous tissue) as well as soft-tissue.DETAILED DESCRIPTION

[0081] It is contemplated that systems, architectures, devices, methods, and processes of the claimed invention encompass variations and adaptations developed using information from the embodiments described herein. Adaptation and / or modification of the systems, architectures, devices, methods, and processes described herein may be performed, as contemplated by this description.

[0082] Throughout the description, where articles, devices, systems, and architectures are described as having, including, or comprising specific components, or where processes and methods are described as having, including, or comprising specific steps, it is contemplated that,Attorney’s Docket No. 2010358-0388additionally, there are articles, devices, systems, and architectures of the present invention that consist essentially of, or consist of, the recited components, and that there are processes and methods according to the present invention that consist essentially of, or consist of, the recited processing steps.

[0083] It should be understood that the order of steps or order for performing certain action is immaterial so long as the invention remains operable. Moreover, two or more steps or actions may be conducted simultaneously.

[0084] Documents are incorporated herein by reference as noted. Where there is any discrepancy in the meaning of a particular term, the meaning provided in the Definition section above is controlling.

[0085] Headers are provided for the convenience of the reader - the presence and / or placement of a header is not intended to limit the scope of the subject matter described herein.

[0086] As described in further detail herein, and, for example, in (i) PCT / EP2021 / 068337, filed on July 2, 2021, entitled, “SYSTEMS AND METHODS FOR ARTIFICIAL INTELLIGENCE-BASED IMAGE ANALYSIS FOR DETECTION ANDCHARACTERIZATION OF LESIONS,” and published on January 13, 2022, as International (PCT) Publication No. WO / 2022 / 008374; and (ii) U. S. Application No. 18 / 207,246, filed July 8, 2023, entitled, “SYSTEMS AND METHODS FOR ASSESSING DISEASE BURDEN AND PROGRESSION,” the contents of each of which are incorporated by reference herein in their entirety, in certain embodiments, automated (including semi -automated) analysis of three dimensional (3D) medical images can be performed to detect or evaluate cancer or risk thereof in a patient, and / or to monitor disease progression and treatment efficacy. These 3D medical images may include, for example, positron emission tomography (PET) images, single-photonAttorney’s Docket No. 2010358-0388emission computerized tomography (SPECT) images, whole-body bone images, combined PET / CT images (CT = computed tomography), and / or combined SPECT / CT images. In certain embodiments, a PET image of a patient is obtained using the prostate-specific membrane antigen (PSMA) binding agent PyL™ (also referred to as 18F-DCFPyL, [18F]DCFPyL, or DCFPyL-18F, with chemical structure as shown in the above-referenced WO / 2022 / 008374) and is overlaid (or otherwise used in combination) with a CT image of the patient. Examples of the automated analysis of these composite 18F-DCFPyL PET / CT images are described in the abovereferenced WO / 2022 / 008374 and U. S. Application No. 18 / 207,246.

[0087] Analysis of medical images may be used to identify and delineate various regions of interest (RO Is) that represent potential underlying physical lesions within a subject. These RO Is may, for example, be localized regions of elevated intensity (e.g., bright spots) - referred to as hotspots, which appear in nuclear medicine images due to accumulation of imaging agents as described herein - or may be associated with other characteristic image features, e.g., depending on the particular imaging modality and collection protocol. Identified ROIs may be analyzed to determine overall patient risk and survival / prognostic metrics, indicative of disease state, progression, and / or treatment efficacy.A. Nuclear Medicine Images

[0088] In certain embodiments, technologies of the present disclosure analyze nuclear medicine images may be obtained using a nuclear medicine imaging modality such as bone scan imaging (also referred to as scintigraphy), Positron Emission Tomography (PET) imaging, and Single-Photon Emission Tomography (SPECT) imaging.Attorney’s Docket No. 2010358-0388

[0089] In certain embodiments, nuclear medicine images are obtained using imaging agents comprising radiopharmaceuticals. Nuclear medicine images may be obtained following administration of a radiopharmaceutical to a patient (e.g., a human subject), and provide information regarding the distribution of the radiopharmaceutical within the patient.

[0090] Nuclear medicine imaging techniques detect radiation emitted from the radionuclides of radiopharmaceuticals to form an image. The distribution of a particular radiopharmaceutical within a patient may be influenced and / or dictated by biological mechanisms, such as blood flow or perfusion, as well as by specific enzymatic or receptor binding interactions. Different radiopharmaceuticals may be designed to take advantage of different biological mechanisms and / or particular specific enzymatic or receptor binding interactions and thus, when administered to a patient, selectively concentrate within particular types of tissue and / or regions within the patient. Greater amounts of radiation are emitted from regions within the patient that have higher concentrations of radiopharmaceutical than other regions, such that these regions appear brighter in nuclear medicine images. Accordingly, intensity variations within a nuclear medicine image can be used to map the distribution of a radiopharmaceutical within the patient. This mapped distribution of radiopharmaceutical within the patient can be used to, for example, infer the presence of cancerous tissue within various regions of the patient’s body. In certain embodiments, intensities of voxels of a nuclear medicine image, for example, a PET image, represent standardized uptake values (SUVs) (e.g., having been calibrated for injected radiopharmaceutical dose and / or patient weight).

[0091] For example, upon administration to a patient, technetium 99m methylenediphosphonate ("mTc MDP) selectively accumulates within the skeletal region of the patient, in particular at sites with abnormal osteogenesis associated with malignant bone lesions.Attorney’s Docket No. 2010358-0388The selective concentration of radiopharmaceutical at these sites produces identifiable hotspots -localized regions of high intensity - in nuclear medicine images. Accordingly, presence of malignant bone lesions associated with metastatic prostate cancer can be inferred by identifying such hotspots within a whole-body scan of the patient. In certain embodiments, analyzing intensity variations in whole-body scans obtained following administration of "mTc MDP to a patient, such as by detecting and evaluating features of hotspots, can be used to compute, risk indices that correlate with patient overall survival and other prognostic metrics indicative of disease state, progression, treatment efficacy, and the like. In certain embodiments, other radiopharmaceuticals can also be used in a similar fashion to99mTc MDP.

[0092] In certain embodiments, a choice of a particular radiopharmaceutical depends on a particular nuclear medicine imaging modality used. For example, 18F sodium fluoride (NaF) also accumulates in bone lesions, similar to "mTc MDP but can be used with PET imaging. In certain embodiments, PET imaging may also utilize a radioactive form of the vitamin choline, which is readily absorbed by prostate cancer cells. In certain embodiments, PET imaging may be performed with [18F]Fluorodeoxy glucose, abbreviated [18F]FDG, which is an 18F labeled glucose analog. Without wishing to be bound to any particular theory, uptake of 18F FDG is believed to be a marker for glucose uptake in tissue, which is correlated with metabolism and, accordingly, may serve as a marker for cancer.

[0093] In certain embodiments, radiopharmaceuticals that selectively bind to particular proteins or receptors of interest - particularly those whose expression is increased in cancerous tissue - may be used. Such proteins or receptors of interest include, but are not limited to tumor antigens, such as CEA, which is expressed in colorectal carcinomas, Her2 / neu, which isAttorney’s Docket No. 2010358-0388expressed in multiple cancers, BRCA 1 and BRCA 2, expressed in breast and ovarian cancers; and TRP-1 and -2, expressed in melanoma.

[0094] For example, human prostate-specific membrane antigen (PSMA) is upregulated in prostate cancer, including metastatic disease. PSMA is expressed by virtually all prostate cancers and its expression is further increased in poorly differentiated, metastatic and hormone refractory carcinomas. Accordingly, radiopharmaceuticals that comprise PSMA binding agents (e.g., compounds that a high affinity to PSMA) labelled with one or more radionuclide(s) can be used to obtain nuclear medicine images of a patient from which the presence and / or state of prostate cancer within a variety of regions (e.g., including, but not limited to skeletal regions) of the patient can be assessed. In certain embodiments, nuclear medicine images obtained using PSMA binding agents are used to identify the presence of cancerous tissue within the prostate, when the disease is in a localized state. In certain embodiments, nuclear medicine images obtained using radiopharmaceuticals comprising PSMA binding agents are used to identify the presence of cancerous tissue within a variety of regions that include not only the prostate, but also other organs and tissue regions such as lungs, lymph nodes, and bones, as is relevant when the disease is metastatic.

[0095] In particular, upon administration to a patient, radionuclide labelled PSMA binding agents selectively accumulate within cancerous tissue, based on their affinity to PSMA. In a similar manner to that described above with regard to "mTc MDP, the selective concentration of radionuclide labelled PSMA binding agents at particular sites within the patient produces detectable hotspots in nuclear medicine images. As PSMA binding agents concentrate within a variety of cancerous tissues and regions of the body expressing PSMA, localized cancer within a prostate of the patient and / or metastatic cancer in various regions of the patient’s body can beAttorney’s Docket No. 2010358-0388detected and evaluated. Various metrics that are indicative of and / or quantify severity (e.g., likely malignancy) of individual lesions, overall disease burden and risk for a patient, and the like, can be computed based on automated analysis of intensity variations in nuclear medicine images obtained following administration of a PSMA binding agent radiopharmaceutical to a patient. These disease burden and / or risk metrics may be used to stage disease and make assessments regarding patient overall survival and other prognostic metrics indicative of disease state, progression, treatment efficacy.

[0096] A variety of radionuclide labelled PSMA binding agents may be used as radiopharmaceutical imaging agents for nuclear medicine imaging to detect and evaluate prostate cancer. In certain embodiments, the particular radionuclide labelled PSMA binding agent that is used depends on factors such as the particular imaging modality (e.g., PET; e.g., SPECT) and the particular regions (e.g., organs) of the patient to be imaged. For example, certain radionuclide labelled PSMA binding agents are suited for PET imaging, while others are suited for SPECT imaging. For example, certain radionuclide labelled PSMA binding agents facilitate imaging a prostate of the patient, and are used primarily when the disease is localized, while others facilitate imaging organs and regions throughout the patient’s body, and are useful for evaluating metastatic prostate cancer.

[0097] Several exemplary PSMA binding agents and radionuclide labelled versions thereof are described in further detail in Section G (“Imaging Agents”) herein, as well as in U. S. Patent Nos. 8,778,305, 8,211,401, and 8,962,799, and in U. S. Patent Publication No. US 2021 / 0032206 Al, the content of each of which are incorporated herein by reference in their entireties.B. Image Segmentation in Nuclear Medicine ImagingAttorney’s Docket No. 2010358-0388

[0098] Nuclear medicine images are examples of functional images. Functional images convey information relating to physiological activities within specific organs and / or tissue, 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 regarding location and extent of anatomical structures such as internal organs, bones, soft-tissue, and blood vessels, within a subject. Examples of anatomical images include, without limitation, x-ray images, CT images, magnetic resonance images, and ultrasound images.

[0099] Accordingly, in certain embodiments, anatomical images can be analyzed together with nuclear medicine images in order to provide anatomical context for the functional information that they (nuclear medicine images) convey. For example, while nuclear medicine images, such as PET and SPECT convey a three-dimensional distribution of radiopharmaceutical within a subject, adding anatomical context from an anatomical imaging modality, such as CT imaging, allows one to determine the particular organs, soft-tissue regions, bones, etc. that radiopharmaceutical has accumulated in.

[0100] For example, a functional image may be aligned with an anatomical image so that locations within each image that correspond to a same physical location - and therefore correspond to each other - can be identified. For example, coordinates and / or pixels / voxels within a functional image and an anatomical image may be defined with respect 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 established. In this manner, one or more voxels within an anatomical image and one or more voxels within a functional image that represent a same physical location or volume can be identified as corresponding to each other.Attorney’s Docket No. 2010358-0388

[0101] For example, FIG. 1 shows axial slices of a 3D CT image 102 and a 3D PET image 104, along with a fused image 106 in which the slice of the 3D CT image is displayed in grayscale and with the PET image is displayed as a semitransparent overlay. By virtue of the alignment between the CT and PET images, a location of a hotspot within the PET image, indicative of accumulated radiopharmaceutical and, accordingly a potential lesion, can be identified in the corresponding CT image, and viewed in anatomical context, for example, within a particular location in the pelvic region (e.g., within a prostate).

[0102] In certain embodiments, the aligned pair are a composite image, such as a PET / CT or SPECT / CT. For example, a composite medical image may comprise a functional image or nuclear medicine image portion (e.g., a first channel), such as a PET image or a SPECT image, and an anatomical image portion (e.g., a second channel), such as a CT or MR image. Different image portions of a composite correspond with each other, in the sense that they represent - i.e., are images of - substantially similar portions of a subject including, for example, similar fields of view in relation to the subject’s body, similar positioning of the subject, similar magnifications, etc. A functional image portion and a corresponding anatomical image portion of a composite image may be aligned such that a mapping or relationship between voxels of each image portion can be established. In certain embodiments, an anatomical image (e.g., a 3D anatomical image, such as a CT image) and a functional image (e.g., a 3D functional image, such as a PET or SPECT image) are acquired using separate anatomical and functional imaging modalities, respectively. In certain embodiments, an anatomical image (e g., a 3D anatomical image, such as a CT image) and a functional image (e.g., a 3D functional image, such as a PET or SPECT image) are acquired using a single multimodality imaging system. A functional image and an anatomical image may, for example, be acquired via two scans using a single multimodalAttorney’s Docket No. 2010358-0388imaging system - for example first performing a CT scan and then, second, performing a PET scan - during which a subject remains in a substantially fixed position.

[0103] In certain embodiments, 3D boundaries of particular tissue regions of interest can be accurately identified by analyzing 3D anatomical images. For example, automated segmentation of 3D anatomical images can be performed to segment 3D boundaries of regions such as particular organs, organ sub-regions and soft-tissue regions, as well as bone. In certain embodiments, organs such as a prostate, urinary bladder, liver, aorta (e.g., portions of an aorta, such as a thoracic aorta), a parotid gland, etc., are segmented. In certain embodiments, one or more particular bones are segmented. In certain embodiments, an overall skeleton is segmented.

[0104] In certain embodiments, automated segmentation of 3D anatomical images may 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 particular regions of interest, producing 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 July 16, 2020, the contents of which are incorporated herein by reference in their entirety, multiple machine learning modules implementing convolutional neural networks (CNNs) may be used to segment 3D anatomical images, such as CT images, of a whole body of a subject and thereby create a 3D segmentation map that identifies multiple target tissue regions across a subject’s body.

[0105] In certain embodiments, for example to segment certain organs where functional images are believed to provide additional useful information that facilitate segmentation, a machine learning module may receive both an anatomical image and a functional image as input, for example as two different channels of input (e.g., analogous to multiple color channels in aAttorney’s Docket No. 2010358-0388color, RGB, image) and use these two inputs to determine an anatomical segmentation. This, multi-channel, approach is described in further detail in U. S. Patent Publication No. US 2021 / 0334974 Al, entitled “Systems and Methods for Deep-Learning-Based Segmentation of Composite Images,” and published October 28, 2021, the contents of which is hereby incorporated by reference in its entirety.

[0106] In certain embodiments, as illustrated FIG. 2, 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) may be aligned with (e.g., co-registered to) each other, for example as in a composite image 202 such as a PET / CT image. Anatomical image 204 may be segmented 208 to create a segmentation map 210 (e.g., a 3D segmentation map) that distinguishably identifies one or more tissue regions and / or sub-regions of interest, such as one or more particular organs and / or bones. Segmentation map 210, having been created from anatomical image 204, is aligned with anatomical image 204, which, in turn, is aligned with functional image 206. Accordingly, boundaries of particular regions (e.g., segmentation masks), such as particular organs and / or bones, identified via segmentation map 210 can be transferred to and / or overlaid 212 upon functional image 206 to identify volumes within functional image 206 for purposes of classifying hotspots, and determining useful indices that serve as measures and / or predictions of cancer status, progression, and response to treatment. Segmentation maps and masks may also be displayed, for example as a graphical representation overlaid on a medical image to guide physicians and other medical practitioners.C. Lesion Detection and Characterization

[0107] In certain embodiments, approaches described herein include techniques for detecting and characterizing lesions within a subject via (e.g., automated) analysis of medical images, suchAttorney’s Docket No. 2010358-0388as nuclear medicine images. Regions of interest (ROIs) in medical images that represent potential lesions may be identified based on, for example, differences in intensity values relative to surroundings, or other characteristic features (e.g., abnormal shapes, texture, spatial frequencies, etc., depending on particular imaging modality). That is, in certain embodiments, ROIs representing potential lesions may appear in medical images as spatial regions having intensity values and / or patterns deviating from an established baseline pattern, e.g., background intensities. In certain embodiments, ROIs are uptake regions that are indicative of anomalous, e.g., increased, uptake of an imaging agent within a particular localized region within a patient. In particular, as described herein, in certain embodiments, uptake regions may be hotspots. In certain embodiments, hotspots are localized (e.g., contiguous) regions of high intensity, relative to their surroundings, within images, such as 3D functional images and may be indicative of a potential cancerous lesion present within a subject.

[0108] A variety of approaches may be used for detecting, segmenting, and classifying uptake regions such as hotspots. In certain embodiments, hotspots are detected and segmented using analytical methods, such as filtering techniques including, but not limited to, a difference of Gaussians (DoG) filter and a Laplacian of Gaussians (LoG) filter. In certain embodiments, hotspots are segmented using a machine learning module that receives, as input, a 3D functional image, such as a PET image, and generates, as output a hotspot segmentation map (a “hotspot map”) that differentiates boundaries of identified hotspots from background. In certain embodiments, each segmented hotspot within a hotspot map is individually identifiable (e.g., individually labelled). In certain embodiments, a machine learning module used for segmenting hotspots may take as input, in addition to a 3D functional image, one or both of a 3D anatomical image (e.g., a CT image) and a 3D anatomical segmentation map. The 3D anatomicalAttorney’s Docket No. 2010358-0388segmentation map may be generated via automated segmentation (e.g., as described herein) of the 3D anatomical image.

[0109] In certain embodiments, segmented hotspots may be classified according to an anatomical region in which they are located. For example, in certain embodiments, locations of individual segmented hotspots within a hotspot 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 segmentation map and labeled according to their location, e.g., based on proximity to and / or overlap with particular organs. In certain embodiments, a machine learning module may be used to classify hotspots. For example, in certain embodiments, a machine learning module may generate, as output, a hotspot map in which segmented hotspots are not only individually labeled and identifiable (e.g., distinguishable from each other), but are also labeled, for example, as corresponding to one of a bone, lymph, or prostate lesion. In certain 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 in sequence to create a final labeled hotspot map.

[0110] Various approaches for performing detailed segmentation of 3D anatomical images and identification of hotspots representing lesions in 3D functional images, which may be used with various approaches described herein, are described in PCT publication WO / 2020 / 144134, entitled “Systems and Methods for Platform Agnostic Whole Body Segmentation,” and published July 16, 2020, U. S. Patent Publication No. US 2021 / 0334974 Al, entitled “Systems and Methods for Deep-Learning-Based Segmentation of Composite Images,” and published October 28, 2021, and PCT publication WO / 2022 / 008374, entitled “Systems and Methods forAttorney’s Docket No. 2010358-0388Artificial Intelligence-Based Image Analysis for Detection and Characterization of Lesions,” and published January 13, 2022, the contents of each of which is incorporated herein in its entirety, [oni] FIG. 3 shows an example process 300 for segmenting and classifying hotspots, based on an example approach described in further detail in PCT publication WO / 2022 / 008374, entitled “Systems and Methods for Artificial Intelligence-Based Image Analysis for Detection and Characterization of Lesions,” and published January 13, 2022. The approach illustrated in FIG. 3 uses two machine learning modules, each of which receives, as input, 3D functional image 306, 3D anatomical image 304, and 3D anatomical segmentation map 310. Machine learning module 312a is a binary classifier that generates a single-class hotspot map 320a, by labeling voxels as hotspot or background (not a hotspot). Machine learning module 312b performs multi-class segmentation, and generates multi-class hotspot map 320b, in which hotspots are both segmented and labeled as one of three classes - prostate, lymph, or bone. Among other things, classifying hotspots in this manner - via a machine learning module 312b (e.g., as opposed to directly comparing hotspot locations with segmented boundaries from segmentation map 310) - obviates a need to segment certain regions. For example, in certain embodiments, machine learning module 312b may classify hotspots as belonging to prostate, lymph, or bone, without a prostate region having be identified and segmented from 3D anatomical image 304 (e.g., in certain embodiments, 3D anatomical segmentation map 310 does not comprise a prostate region). In certain embodiments, hotspot maps 320a and 320b are merged, for example, by transferring labels from multi-class hotspot map 320b to the hotspot segmentations identified in single-class hotspot map 320a (e.g., based on overlap). Without wishing to be bound to any particular theory, it is believed that this approach combines improved segmentation and detection of hotspots from single class machine learning module 312a withAttorney’s Docket No. 2010358-0388classification results from multi-class machine learning module 312b. In certain embodiments, hotspot regions identified via this final, merged, hotspot map are further refined, using an analytical technique such as an adaptive thresholding technique described in PCT publication WO / 2022 / 008374, entitled “Systems and Methods for Artificial Intelligence-Based Image Analysis for Detection and Characterization of Lesions,” and published January 13, 2022.

[0112] In certain embodiments, once detected and segmented, hotspots may be identified and assigned labels according to a particular anatomical (e.g., tissue) region in which they are located and / or a particular lesion sub-type that they are likely to represent. For example, in certain embodiments, hotspots may be assigned an anatomical location that identifies them as representing locations with a one of a set of tissue regions, such as the listed in Table 1, below. In certain embodiments, a list of tissue regions may include those in Table 1 as well as a gluteus maximus (e.g., left and right) and a gallbladder. In certain embodiments, hotspots are assigned to and / or labeled as belonging to a particular tissue region based on a machine learning classification and / or via comparison of their 3D hotspot volume’s location and / or overlap with various tissue volumes identified via masks in an anatomical segmentation map. In certain embodiments, a prostate is not segmented. For example, as described above, in certain embodiments, machine learning module 312b may classify hotspots as belonging to prostate, lymph, or bone, without a prostate region having be identified and segmented from 3D anatomical image 304. In certain embodiments, regions such as a head and / or one or more arm(s) of a patient, may be segmented. In certain embodiments, one or more arm(s) comprises upper arm(s) and / or at least a portion of one or both humerus / humeri. In certain embodiments, a head comprises a skull, a mandible, a brain, a parotid gland, and cervical curve, and a cervical vertebra. In certain embodiments, segmentation processes of the present disclosure may utilize aAttorney’s Docket No. 2010358-0388cutoff rule to determine whether to segment a particular region based on an extent of a subject that is imaged, and / or is graphically represented in a medical image. For example, in certain embodiments, if a medical image does not cover (e.g., certain percentage of a region of interest, reach set anatomical landmarks), segmentation of a particular region (e.g., a skull, a parotid gland) is not performed, or may be performed, but discarded or flagged (e.g., as low confidence / potenti ally erroneous).

[0113] In certain embodiments, technologies of the present disclosure provide users with tools that include segmentation procedures that utilize particular processes for detecting and / or segmenting hotspots that are tailored to different regions (e.g., organs). For example, a user may seek to locate hotspots in an organ with high physiological uptake, such as a liver that an approach used for (detecting and / or segmenting hotspots within) other organs may struggle with. Accordingly, a tailored approach for detecting and / or segmenting hotspots in high uptake organs may be used.

[0114] In certain embodiments, a hotspot segmentation procedure for organ with high physiological uptake may include one or both of: (1) detecting hotspots by applying a model (e.g., statistical, machine-learning) to determine and remove (e.g., filter out) intensities corresponding to normal, background, uptake within the organ; and (2) segmenting hotspots by applying a model (e.g., statistical, machine-learning) to identify and delineate 3D hotspot volume boundaries using the organ boundaries (e.g., which may be identified via one or more segmentation masks determined via approaches described herein, for example, in Section B above) and / or determined local background intensity value associated with and representing normal background uptake within the organ (e.g., by detecting abnormally high intensity peaks).Attorney’s Docket No. 2010358-0388

[0115] In certain embodiments, a local background intensity value associated with background uptake in a high-uptake organ comprises is determined as a mean and / or a standard deviation of the uptake value in the organ. In certain embodiments, a local background intensity value is determined by fitting a multi-component mixture model to intensities of voxels within a VOI corresponding to the high uptake organ. For example, a mean and a standard deviation of local background intensity in an organ is determined by fitting a two component Gaussian Mixture model to the organ intensity values. For example, a two component Gaussian mixture model may have the following form:where is the mean, tr, is the standard deviation and w, is the weight of the zthcomponent and G is the Gaussian function. In certain embodiments, a mean M) and a standard deviation (a) of a component with the largest weight, w,, (e.g., a major mode) is chosen as an estimate of a mean and a standard deviation, respectively, of local background intensity corresponding to normal background uptake within the organ. A smaller component may be assumed to encapsulate abnormally low uptake (e.g., due to misalignment between a medical image and organ segmentation, and / or due to the cysts in the organ causing abnormally low uptake).

[0116] In certain embodiments, sub-regions within the high-uptake organ that represent locations of potential hotspots are detected by thresholding organ intensity values and selecting regions with intensities above a detection threshold detection value. In certain embodiments, a detection threshold detection value is determined based on (e g., as a function of) the local background intensity value, for example based on a mean and standard deviation of a major mode determined via fitting a multi-component mixture model as described herein (e.g., as th = uAttorney’s Docket No. 2010358-0388+4 ). In certain embodiments, an initial set of preliminary sub-regions are detected, and filtered via one or more selection criteria. One or more selection criteria may include (1) a degree of overlap (e.g., absolute volume, volumetric percentage) between a given one of the one or more preliminary sub-regions and a region (e.g., 3D volume of interest) of the 3D functional image corresponding to the high-uptake organ; (2) a degree of overlap (e.g., absolute volume, volumetric percentage) between a given one of the one or more preliminary sub-regions and a region (e.g., a 3D volume) of the 3D functional image corresponding to at least one other organ or tissue region [e.g., a neighboring organ, an organ with high physiological uptake, an organ known to “bleed over” (e.g., distort uptake values of a neighboring organ due to a misalignment between a functional image and a segmentation mask) into the high uptake organ, e.g., kidney]; (3 a minimum (e.g., hotspot) volume (e.g., a minimum number of voxels; e.g., a minimum corresponding physical volume) [e.g., such that only preliminary sub-regions having volumes greater than or equal to the minimum (e.g., hotspot) volume are selected] [e.g., wherein the minimum hotspot volume is at least 2 (e.g., 1, 5, 10) voxels] (e.g., to avoid noise induced hotspots).

[0117] In certain embodiments, hotspots are segmented by sub-dividing a VOI within the functional image corresponding to the high uptake organ into one or more sub-segments, for example using a watershed algorithm. Intensities within the VOI may be smoothed prior to subdividing it and peaks detected and used as basins in the watershed algorithm. Smoothing may avoid detecting noise as peaks. These determined sub-segments may then be compared with the detected sub-regions representing potential hotspots, to identify, for each preliminary sub-region, a subsegment that it overlaps with. A 3D hotspot volume corresponding to a particular subregion (e.g., a jthsub-region and corresponding hotspot volume) may be segmented using anAttorney’s Docket No. 2010358-0388individual segmentation threshold value (ts), determined based on (e.g., as a maximum of) the detection threshold value and a measure of intensity within the particular sub-region (e.g., a particular percentage, e.g., 60%, of the SUV max). For example, for a jthsub-region and corresponding 3D hotspot volume to be segmented, a corresponding individual segmentation threshold may be determined via,tSj = max z + 4< J, 0.6 • SUVmctXj)In certain embodiments, a 3D hotspot volume is segmented using a flood fill algorithm originating in the SUVmax of the detected hotspot, including values that are above the threshold tSj, and are located in the subsegment(s) that the original sub-region (that represents the detected hotspot) overlaps with (e.g., the hotspot segmentation is not allowed to spread into subsegments that were not previously overlapped when the hotspot was detected).Table 1: Certain Tissue Regions (^Prostate may, optionally, be segmented if present - may be absent if patient has, e.g., undergone radical prostatectomy, or may not segmented in any case, in certain embodiments).Attorney’s Docket No. 2010358-0388

[0118] In certain embodiments, additionally or alternatively, hotspots may be classified as belonging to one or more lesion sub-types. In certain embodiments, lesion sub-type classifications may be made by comparing hotspot locations with classes of anatomical regions. For example, in certain embodiments a miTNM classification scheme may be used, where hotspots are labeled as belonging to one of three classes - miT, miN, or miM - based on whether they represent lesions located within a prostate (miT), pelvic lymph node (miN), or distant metastases (miM). In certain embodiments, a five-class version of the miTNM scheme may be used, with distant metastases further divided into three sub classes - miMb for bone metastases, miMa for lymph metastases, and miMc for other soft tissue metastases.

[0119] For example, in certain embodiments, hotspots located within a prostate are labeled as belonging to class “T” or “miT”, e.g., representing local tumor. In certain embodiments, hotspots located outside a prostate, but within a pelvic region are labeled as class “N” or “miN”. In certain embodiments, for example as described in U. S. Application No. 17 / 959,357, filed October 4, 2022, entitled “Systems and Methods for Automated Identification and ClassificationAttorney’s Docket No. 2010358-0388of Lesions in Local Lymph and Distant Metastases,” published as U. S. 2023 / 0115732 Al on April 13, 2023, the content of which is incorporated herein by reference in its entirety, a pelvic atlas may be registered to identify boundaries of a pelvic region and / or various sub-regions therein, for purposes of identifying pelvic lymph node lesions. A pelvic atlas may, for example, include boundaries of a pelvic region and / or a planar reference (e.g., a plane passing through an aorta-bifurcation) which hotspot locations can be compared to (e.g., such that hotspots located outside the pelvic region and / or above the planar reference passing through an aorta bifurcation are labeled as “M” or “miM” - e.g., distant metastases). In certain embodiments, distant metastases may be classified as lymph (miMa), bone (miMb), or visceral (miMc) based on a comparison of hotspot locations with an anatomical segmentation map. For example, hotspots located within one or more bones (e.g., and outside a pelvic region) may be labeled as distant metastases, hotspots located within one or more segmented organs or a subset of organs (e.g., brain, lung, liver, spleen, kidneys) may be labeled as visceral (miMc) distant metastases, and remaining hotspots located outside a pelvic region labeled as distant lymph metastases (miMa).

[0120] Additionally or alternatively, in certain embodiments, hotspots may be assigned an miTNM class based on a determination that they are located within a particular anatomical region, for example based on a table such as Table 2, where each column corresponds to a particular miTNM label (first row indicating the particular miTNM class) and includes, in rows two and below, particular anatomical regions associated with each miTNM class. In certain embodiments, a hotspot can be assigned as being located within a particular tissue region listed in Table 2 based on a comparison of the hotspot’s location with an anatomical segmentation map, allowing for an automated miTNM class assignment.Attorney’s Docket No. 2010358-0388Table 2. An Example List of Tissue Regions Corresponding to Five Classes in a Lesion Anatomical Labeling Approach.

[0121] In certain embodiments, hotspots may be further classified in terms of their anatomical location and / or lesion sub-type. For example, in certain embodiments, hotspots identified as located in pelvic lymph (miN) may be identified as belonging to a particular pelvic lymph node sub-region, such as one of a left / right internal iliac, a left or right external iliac, a left or right common iliac, a left or right obturator, a presacral region, or other pelvic region. InAttorney’s Docket No. 2010358-0388certain embodiments, distant lymph node metastases (miMa) may be classified as retroperitoneal (RP), supradiaphragmatic (SD), or other extrapelvic (OE). Approaches for regional (miN) and distant (miMa) lymph metastases classifications may include registration of pelvic atlas images and / or identification of various whole body landmarks, which are described in further detail in U. S. Application No. 17 / 959,357, filed October 4, 2022, entitled “Systems and Methods for Automated Identification and Classification of Lesions in Local Lymph and Distant Metastases,” published as U. S. 2023 / 0115732 Al on April 13, 2023, the content of which is incorporated herein by reference in its entirety.D. Interactive Display and Filtering of Hotspot Overlays

[0122] The present disclosure also presents systems and methods for interactive display of a medical image of a subject as shown, for example, in FIG. 4. In certain embodiments, a method for interactive display of a medical image 400 comprises rendering a first instantiation of the medical image for graphical display 402. A first instantiation may comprise a black-and-white (B& W) and / or grayscale or otherwise monochrome depiction of an anatomy of the subject with a first overlay superimposed on the depiction of the anatomy. A first overlay may comprise a first color rendering of regions (e.g., 3D regions) corresponding to a set of hotspots. The color difference between the hotspot rendering and the anatomy depiction may create visual contrast such that a user can more easily and / or quickly distinguish between the hotspot rendering and the anatomy depiction, aiding them to rapidly and easily identify hotspots and their relative positions with respect to the underlying anatomy of the subject.

[0123] In certain embodiments, the regions corresponding to the set of hotspots in the first color rendering and / or the second color rendering are color coded based on lesion type, such as a TNM type (tumor, node, and metastasis type). For example, each hotspot may be assigned aAttorney’s Docket No. 2010358-0388lesion type selected from a TNM category, such as miT (e.g., identifying the hotspot as representing a local lesion, e.g., within a prostate), miN (e.g., identifying the hotspot as representing a lesion in a pelvic lymph node), miMa (e.g., identifying the hotspot as representing a lesion in distant lymph node), miMb (e.g., identifying the hotspot as representing a bone lesion), and / or miMc (e.g., identifying the hotspot as representing a lesion in a distant organ), e.g., as described herein. Then, in a color rendering, each region corresponding to a hotspot can be color coded to reflect the TNM type of the corresponding hotspot, for example with regions corresponding to miT hotspots shown in a first color, regions corresponding to miN hotspots shown in a second color, and so on. A legend conveying the color-coding scheme may be displayed, e.g., showing color categories and associated TNM types. The legend may comprise (e.g., next to each hotspot category) a number reflecting the number of hotspots belonging to each color category.

[0124] In certain embodiments, at least some of the regions corresponding to the set of hotspots (e.g., in the first color rendering and / or the second color rendering) have an outline. For a given hotspot, an outline may be represented by a contour line along a perimeter of the associated region. The contour line may be of a different color as compared to the color of the region. The contour line may be of the same color as the color of the region. The contour line may be of the same color but have a different transparency, different hue, different contrast, or different saturation as compared to a color of the region. The contour line and the region may be visually distinct. The contour line may aid a user in discerning and / or easily assessing the edge of the region. A user may adjust the thickness of the outline.

[0125] In certain embodiments, at least some hotspots are segmented using a machine learning algorithm. The machine learning algorithm may comprise deep learning neural networkAttorney’s Docket No. 2010358-0388and / or a convolutional neural network. In certain embodiments, at least some hotspots (e.g., in the liver and spleen) are segmented using an adaptive threshold. The adaptive threshold may be based on hotspot SUVmax. The adaptive threshold may be based on the background uptake of an organ.

[0126] In certain embodiments, method 400 comprises causing display of one or more graphical control elements (widgets) 404, allowing for user selection (e.g., setting or adjustment) of one or more criteria for application of a graphical display filter to adjust a first instantiation of the medicine image. The criteria may comprise a range and / or a minimum and / or a maximum of measurements associated with each of the hotspots. The measurements may comprise a Standardized Uptake Value (SUV), for example, SUVmean, SUVmax, and / or SUVpeak. The measurements may comprise a lesion size (e.g., volume). The measurements may comprise a Lesion Index (LI). A user may select and / or deselect displayed hotspots by interacting with one or more graphical control elements, e.g., by inputting, setting, or adjusting one or more criteria. A user may select separate criteria for different lesion types. A user may select separate criteria for different organs. In certain embodiments, the criteria may be recommended to a user. The recommendations may be based on pre-defined parameters. The recommendations may be (e.g., automatically) determined based on hotspots of a medical image, e.g., using a machine learning algorithm, e.g., by removing hotspot with SUV below a certain threshold, e.g., in a specific organ.

[0127] A user may select and / or de-select a color category in the legend (e.g., by clicking with a mouse pointer on the category) such that corresponding hotspots are selected / de-selected, respectively. A user may select and / or deselect displayed regions, e.g., by clicking with a mouse pointer on individual regions, e.g., by selecting or defining an area in the first instantiation. TheAttorney’s Docket No. 2010358-0388associated measurements of selected hotspots may be displayed. The criteria may be determined based on the associated measurements of the selected hotspots, e.g., minimum and maximum values of SUV of the selected hotspots, a range of LI as determined by maximum and minimum values of LI of the selected hotspots. A user may have an option to apply the criteria to all hotspots to select a portion of hotspots that satisfy the criteria determined based on the user-defined hotspots.

[0128] In certain embodiments, the criteria may be applied to hotspots by default, also known as presets. In other words, the first instantiation may comprise a first portion of selected hotspots and a second portion of unselected hotspots. Various organs may be associated with different presets. The presets may be user defined. A user may adjust presets. The presets may be (e.g., automatically) determined based on hotspots, e.g., using a machine learning algorithm.

[0129] In certain embodiments, a method for interactive display of a medical image 400 comprises receiving, via one or more widgets, a user selection of one or more criteria 406 for applying a graphical display filter to adjust a first instantiation of the medicine image. In certain embodiments, a method for interactive display of a medical image 400 comprises rendering a second instantiation of a medicine image 408. A second instantiation may reflect a product of the adjustment of a first instantiation of the medicine image following application of a graphical display filter. A second instantiation may comprise a B& W and / or grayscale or otherwise monochrome depiction of the anatomy of the subject with a second overlay superimposed on the depiction of the anatomy. A second overlay may comprise a second color rendering of the regions corresponding to a set of hotspots depicted in a first overlay. A second color rendering may comprise a color of one or more of the regions according to one or more thresholds of a graphical display filter. The color difference between the first instantiation and the secondAttorney’s Docket No. 2010358-0388instantiation may create visual contrast such that a user more easily and / or faster distinguishes between selected and non-selected hotspots, aiding to rapidly and easily assess hotspots that satisfy one or more criteria.

[0130] In certain embodiments, appearance (e.g., color) of a set of hotspots is adjusted to visually highlight or otherwise graphically emphasize one or more hotspots from the set that satisfy filter criteria. In certain embodiments, appearance (e.g., color) of a set of hotspots is adjusted to visually / graphically de-emphasize one or more hotspots from the set that do not satisfy a filter criteria. The hotspots deselected by a user may change their appearance such that the user may again select them. The change in appearance may comprise a change in transparency, a change in color, a change in hue, a change in contrast, a change in saturation. The hotspots deselected by a user may become completely transparent. A legend may be updated (e.g., automatically) to display a number of hotspots based on user selection process. A legend may be updated to display, for each color category, a number of selected hotspots and a total number of hotspots in the category.

[0131] In certain embodiments, two parallel studies of medical images are displayed within a single window. Such parallel display may aid a user in faster and more direct visual comparison between two studies to assess development, differences, or changes in hotspots. A user may apply same presets to the two studies. A user may apply same criteria to the two studies. A pair of medical images may be automatically selected out of a series of images. For example, the pair may be selected based on timing (e.g., first and last), quality (e g., contrast between hotspots and background), progress (e.g., number of hotspots). Such automatic selection may aid a user in rapidly detecting most drastic, relevant, or important developments or changes of hotspots.E. Hotspot and Uptake Region Quantification MetricsAttorney’s Docket No. 2010358-0388

[0132] In certain embodiments, detected - e.g., identified and segmented - uptake regions, such as hotspots, may be characterized via various individual quantification metrics. In particular, for a particular individual uptake region, such as a hotspot, uptake region quantification metrics can be used to quantify a measure of size (e.g., 3D volume) and / or intensity of the particular uptake region in a manner that is indicative of a size and / or level of radiopharmaceutical uptake within the (e.g., potential) underlying physical lesion that the particular uptake region (e.g., hotspot) represents. Accordingly, individual uptake region quantification metrics may convey, for example, to a physician or radiologist, a likelihood that an uptake region (e.g., hotspot) appearing in an image represents a true underlying physical lesion and / or convey a likelihood or level of malignancy thereof (e.g., allowing to differentiate between benign and malignant lesions).

[0133] In certain embodiments, image segmentation, lesion detection, and characterization techniques as described herein are used to determine, for each of one or more medical images, a corresponding set of one or more uptake regions, such as a hotspot set. For example, as described herein, image segmentation techniques may be used to determine, for each hotspot detected in a particular image, a particular 3D volume - a 3D hotspot volume - representing and / or indicative of a volume (e.g., 3D location and extent) of a potential underlying physical lesion within the subject. Each uptake region, in turn, comprises a set of image voxels, each having a particular intensity value.

[0134] Once determined, a set of uptake regions may be used to compute one or more quantification metrics for each individual uptake region. Individual uptake region quantification metrics (e.g., hotspot quantification metrics) may be computed according to various methods and formulae described herein, for example below. In the description below, the variable L is used toAttorney’s Docket No. 2010358-0388refer to a set of uptake regions detected with a particular image, with L = { 1, 2,, AL} representing a set of NL (i.e., N being the number of uptake regions, such as hotspots) uptake regions detected within an image and the variable / indexing the 7thuptake region. As described herein, each uptake region corresponds to a particular 3D volume within an image, with 7? / denoting the volume of the 7thuptake region.

[0135] Uptake region quantification metrics may be presented to a user via a GUI and / or a (e.g., automatically or semi-automatically) generated report. As described in further detail herein, individual uptake region quantification metrics may include uptake region intensity metrics (e.g., hotspot intensity metrics) and uptake region volume metrics (e.g., hotspot volume metrics) (e.g., lesion volume) that quantify an intensity and size, respectively of a particular uptake region (such as a particular hotspot) and / or underlying lesion it represents. Uptake region intensity and size may, in turn, be indicative of a level of radiopharmaceutical uptake within, and size of, respectively, an underlying physical lesion within the subject.E. i. Uptake Region Intensity Metrics

[0136] In certain embodiments, an uptake quantification metric is or comprises an uptake region (e.g., hotspot) intensity metric or uptake (e.g., hotspot) intensity metric that quantifies an intensity of an individual uptake region, such as a 3D hotspot volume. Uptake intensity metrics may be computed based on individual voxel intensities within identified uptake regions. For example, for a particular uptake region, a value of an uptake intensity metric may be computed as a function of at least a portion (e.g., a particular subset, e.g., all) of that hotspot’s voxel intensities. Uptake intensity metrics may include, without limitation, metrics such as a maximum uptake intensity, a mean uptake intensity, and peak uptake intensity, and the like. AsAttorney’s Docket No. 2010358-0388with voxel intensities in nuclear medicine images, in certain embodiments uptake intensity metrics may represent (e.g., be in units of) SUV values.

[0137] In certain embodiments, a value of a particular uptake intensity metric is computed, for a subject uptake region, based on (e.g., as a function of) that subject uptake region’s voxel intensities alone, e.g., and not based on intensities of other image voxels outside the subject uptake region.

[0138] For example, an uptake intensity metric may be a maximum uptake intensity (e.g., SUV), or “SUV-max,” computed as a maximum voxel intensity (e.g., SUV or uptake) within an uptake region (e.g., within a 3D hotspot volume). In certain embodiments, a maximum uptake intensity may be computed according to equations (la), (lb), or (1c), below.(la) Qmax ) = max(Qj)lERi(lc) SUV = max (Uptake InVoxel G lesion volume)where, in equations (la) and (lb) / represents a particular (e.g., 7th) uptake region, as described above, qt is the intensity of voxel i and i G Rtis the set of voxels within the particular uptake region, Ri. In equation (lb), SUVt indicates a particular unit - standardized uptake value (SUV) -of voxel intensity, as described herein.

[0139] In certain embodiments, an uptake intensity metric may be a mean uptake region intensity (e.g., SUV), or “SUV-mean,” and may be computed as a mean over all voxel intensities (e.g., SUV or uptake) within an uptake region. In certain embodiments, a mean uptake region intensity may be computed according to equations (2a), (2b), or (2c) below.Attorney’s Docket No. 2010358-0388(2a)(2b)SUVmean(l = meaner)ieRl(2c) UptakelnVoxelswmean= yi lesion volumeniwhere m is the number of individual voxels within a particular uptake region (e.g., a 3D hotspot volume).

[0140] In certain embodiments, an uptake region intensity metric may be a peak uptake region intensity (e.g., SUV), or “SUV-peak,” and may be computed as a mean over intensities of the voxels (e.g., SUV or uptake) whose midpoints are located within a (e.g., pre-defined) particular distance (e.g., within 5 mm) of the midpoint of the uptake region voxel where the maximum intensity (e.g., SUV-max) is located within an uptake region, and, accordingly, may be computed according to equations (3a)-(3c) below.Attorney’s Docket No. 2010358-0388where i: dist(imax, i) < d is the set of (uptake region) voxels having a mid-point within a distance, d, from voxel imax, which is the maximum intensity voxel within the uptake region (e.g., Qmax(J)=<7i-max).E. U. Lesion Index Metrics

[0141] In certain embodiments, an uptake region intensity metric is individual lesion index value that maps an intensity of voxels within a particular 3D hotspot volume to a value on a standardized scale. Such lesion index values are described in further detail in PCT / EP2020 / 050132, filed January 6, 2020, and PCT / EP2021 / 068337, filed July 2, 2021, the content of each of which is hereby incorporated by reference in its entirety. Calculation of lesion index values may include calculation of reference intensity values within particular reference tissue regions, such as an aorta portion (also referred to as blood pool) and / or a liver.

[0142] For example, in one particular implementation, a first, blood-pool, reference intensity value is determined based on a measure of intensity (e.g., a mean SUV) within an aorta region and a second, liver, reference intensity value is determined based on a measure of intensity (e.g., a mean SUV) within a liver region. As described in further detail, for example in PCT / EP2021 / 068337, filed July 2, 2021, the content of which is incorporated herein by reference in its entirety, calculation of reference intensities may include approaches such as identifying reference volumes (e.g., an aorta or portion thereof; e.g., a liver volume) within a functional image, such as a PET or SPECT image, eroding and / or dilating certain reference volumes, e.g., to avoid include voxels on the edge of a reference volume, and selecting subsets of reference voxel intensities, based on modeling approaches, e.g., to account for anomalous tissue features, such as cysts and lesions, within a liver. In certain embodiments, a third reference intensityAttorney’s Docket No. 2010358-0388value may be determined, either as a multiple (e.g., twice) of a liver reference intensity value, or based on an intensity of another reference tissue region, such as a parotid gland.

[0143] In certain embodiments, uptake region intensities may be compared with one or more reference intensity values to determine a lesion index as a value on a standardized scale, which facilitates comparison across different images. For example, FIGs. 5 and 6 illustrate approaches for assigning uptake regions a lesion index value ranging from 0 to 3. In the approach shown in FIGs. 5 and 6, a blood-pool (aorta) intensity value is assigned a lesion index of 1, a liver intensity value is assigned a lesion 2, and a value of twice the liver intensity is assigned a lesion index of 3. A lesion index for a particular uptake region can be determined by first computing a value of an initial uptake region intensity metric for the particular uptake region, such as a mean uptake region intensity (e.g., Qmean(Z) or SUl mean) and comparing the value of the initial uptake region intensity metric with the reference intensity values. For example, the value of the initial hotspot intensity metric may fall within one of four ranges - [0, GFbiood], (StTFbiood, 5( / Fiiver], (5(7Fiiver, 2 x STTFiiver], and greater than 2x57 / Iijvcr(e.g., (2x5t7Piiver, co)). In certain embodiments, an uptake region may be assigned a lesion index value based on a step function, e.g., as shown in FIG. 5. In certain embodiments, as shown in FIG. 6 a lesion index value can then be computed for the particular uptake region based on (i) the value of the initial uptake region intensity metric and (ii) a linear interpolation according to the particular range in which the value of the initial uptake region intensity metric falls, as illustrated in FIG. 6, where the fdled and open dots on the horizontal (SUV) and vertical (LI) axes illustrate example values of initial hotspot intensity metrics and resultant lesion index values, respectively. In certain embodiments, if SUV references for either liver or aorta cannot be calculated, or if the aortaAttorney’s Docket No. 2010358-0388value is higher than the liver value, the lesion index will not be calculated and will be displayed as

[0144] A lesion index value according to the mapping scheme described above and illustrated in FIG. 5B may, for example, be computed as shown in equation (4), below.and / ) are linear interpolations between the respective spans in equation (4).E.iii. Uptake Region, Lesion Volume

[0145] In certain embodiments, an uptake region quantification metric may be a volume metric, such as a lesion volume, (Aoi, which provides a measure of size (e.g., volume) of an underlying physical lesion that an uptake region represents. A lesion volume may, in certain embodiments, computed as shown in equations (5a) and (5b), below.(5b) Qvoi = v x ntwhere in equation (5a), v, is a volume of an zthvoxel, and equation (5b) assumes a uniform voxel volume, v, and as before « / is a number of voxels in a particular hotspot volume, / . In certain embodiments, a voxel volume is computed as v — 8x x 8y x 8z, where dx, 8y, and 8z are grid spacing (e.g., in millimeters, mm) in x, y, and z. In certain embodiments, a lesion volume has units of milliliters (ml).Attorney’s Docket No. 2010358-0388F. Computer System and Network Architecture

[0146] Certain embodiments described herein make use of 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., a software function) that implements one or more specific machine learning techniques, e.g., artificial neural networks (ANNs), e.g., convolutional neural networks (CNNs), e.g., recursive neural networks, e.g., recurrent neural networks such as long short-term memory (LSTM) or Bilateral long short-term memory (Bi-LSTM), random forest, decision trees, transformers, support vector machines, and the like, in order to determine, for a given input, one or more output values. In certain embodiments, the machine learning module comprises or is a supervised deep learning neural network.

[0147] In certain embodiments, machine learning modules implementing machine learning techniques are trained, for example using datasets that include categories of data described herein (e.g., CT images, MRI images, PET images, SPECT images). Such training may be used to determine various parameters of machine learning algorithms implemented by a machine learning module, such as weights associated with layers in neural networks. In certain embodiments, once a machine learning module is trained, e.g., to accomplish a specific task such as segmenting anatomical regions, segmenting and / or classifying hotspots, or determining values for prognostic, treatment response, and / or predictive metrics, values of determined parameters are fixed and the (e.g., unchanging, static) machine learning module is used to process new data (e.g., different from the training data) and accomplish its trained task without further updates to its parameters (e.g., the machine learning module does not receive feedback and / or updates). InAttorney’s Docket No. 2010358-0388certain embodiments, machine learning modules may receive feedback, e.g., based on user review of accuracy, and such feedback may be used as additional training data, to dynamically update the machine learning module. In certain embodiments, two or more machine learning modules may be combined and implemented as a single module and / or a single software application. In certain embodiments, two or more machine learning modules may also be implemented separately, e.g., as separate software applications. A machine learning module may be software and / or hardware. For example, a machine learning module may be implemented entirely as software, or certain functions of an ANN module may be carried out via specialized hardware (e.g., via an application specific integrated circuit (ASIC)).

[0148] In certain embodiments, the software comprises one or more graphical control elements (widgets) designed to accept user input, for example, graphical control elements that allow for user selection (e.g., setting or adjustment) of one or more thresholds for application of a graphical display filter to adjust nuclear medicine images. The one or more graphical control elements (widgets) may include, for example, one or more windows, text boxes, dialog boxes, buttons, radio buttons, dropdown menus, scroll bars, sliders, and / or the like. The user interaction with the one or more widgets may comprise entry of alphanumeric characters, mouse click(s) (e.g., mouse clicks on one or more portions of the nuclear medicine image), drag-and-drop movement(s), and / or the like.

[0149] As shown in FIG. 7, an implementation of a network environment 700 for use in providing systems, methods, and architectures as described herein is shown and described. In brief overview, referring now to FIG. 7, a block diagram of an exemplary cloud computing environment 700 is shown and described. The cloud computing environment 700 may include one or more resource providers 702a, 702b, 702c (collectively, 702). Each resource provider 702Attorney’s Docket No. 2010358-0388may include computing resources. Tn some implementations, computing resources may include any hardware and / or software used to process 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 702 may be connected to any other resource provider 702 in the cloud computing environment 700. In some implementations, the resource providers 702 may be connected over a computer network 708. Each resource provider 702 may be connected to one or more computing devices 704a, 704b, 704c (collectively, 704), over the computer network 708.

[0150] The cloud computing environment 700 may include a resource manager 706. The resource manager 706 may be connected to the resource providers 702 and the computing devices 704 over the computer network 708. In some implementations, the resource manager 706 may facilitate the provision of computing resources by one or more resource providers 702 to one or more computing devices 704. The resource manager 706 may receive a request for a computing resource from a particular computing device 704. The resource manager 706 may identify one or more resource providers 702 capable of providing the computing resource requested by the computing device 704. The resource manager 706 may select a resource provider 702 to provide the computing resource. The resource manager 706 may facilitate a connection between the resource provider 702 and a particular computing device 704. In some implementations, the resource manager 706 may establish a connection between a particular resource provider 702 and a particular computing device 704. In some implementations, the resource manager 706 may redirect a particular computing device 704 to a particular resource provider 702 with the requested computing resource.Attorney’s Docket No. 2010358-0388

[0151] FIG. 8 shows an example of a computing device 800 and a mobile computing device 850 that can be used to implement the techniques described in this disclosure. The computing device 800 is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The mobile computing device 850 is intended to represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smart-phones, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to be limiting.

[0152] The computing device 800 includes a processor 802, a memory 804, a storage device 806, a high-speed interface 808 connecting to the memory 804 and multiple high-speed expansion ports 810, and a low-speed interface 812 connecting to a low-speed expansion port 814 and the storage device 806. Each of the processor 802, the memory 804, the storage device 806, the high-speed interface 808, the high-speed expansion ports 810, and the low-speed interface 812, are interconnected using various busses, and may be mounted on a common motherboard or in other manners as appropriate. The processor 802 can process instructions for execution within the computing device 800, including instructions stored in the memory 804 or on the storage device 806 to display graphical information for a GUI on an external input / output device, such as a display 816 coupled to the high-speed interface 808. In other implementations, multiple processors and / or multiple buses may be used, as appropriate, along with multiple memories and types of memory. Also, multiple computing devices may be connected, with each device providing portions of the necessary operations (e.g., as a server bank, a group of blade servers, or a multi-processor system). Thus, as the term is used herein, where a plurality ofAttorney’s Docket No. 2010358-0388functions are described as being performed by “a processor”, this encompasses embodiments wherein the plurality of functions are performed by any number of processors (one or more) of any number of computing devices (one or more). Furthermore, where a function is described as being performed by “a processor”, this encompasses embodiments wherein the function is performed by any number of processors (one or more) of any number of computing devices (one or more) (e.g., in a distributed computing system).

[0153] The memory 804 stores information within the computing device 800. In some implementations, the memory 804 is a volatile memory unit or units. In some implementations, the memory 804 is a non-volatile memory unit or units. The memory 804 may also be another form of computer-readable medium, such as a magnetic or optical disk.

[0154] The storage device 806 is capable of providing mass storage for the computing device 800. In some implementations, the storage device 806 may be or contain a computer-readable medium, such as a hard disk device, an optical disk device, a flash memory or other similar solid state memory device, or an array of devices, including devices in a storage area network or other configurations. Instructions can be stored in an information carrier. The instructions, when executed by one or more processing devices (for example, processor 802), perform one or more methods, such as those described above. The instructions can also be stored by one or more storage devices such as computer- or machine-readable mediums (for example, the memory 804, the storage device 806, or memory on the processor 802).

[0155] The high-speed interface 808 manages bandwidth-intensive operations for the computing device 800, while the low-speed interface 812 manages lower bandwidth-intensive operations. Such allocation of functions is an example only. In some implementations, the highspeed interface 808 is coupled to the memory 804, the display 816 (e.g., through a graphicsAttorney’s Docket No. 2010358-0388processor or accelerator), and to the high-speed expansion ports 810, which may accept various expansion cards (not shown). In the implementation, the low-speed interface 812 is coupled to the storage device 806 and the low-speed expansion port 814. The low-speed expansion port 814, which may include various communication ports (e.g., USB, Bluetooth®, Ethernet, wireless Ethernet) may be coupled to one or more input / output devices, such as a keyboard, a pointing device, a scanner, or a networking device such as a switch or router, e.g., through a network adapter.

[0156] The computing device 800 may be implemented in a number of different forms, as shown in the figure. For example, it may be implemented as a standard server 820, or multiple times in a group of such servers. In addition, it may be implemented in a personal computer such as a laptop computer 822. It may also be implemented as part of a rack server system 824. Alternatively, components from the computing device 800 may be combined with other components in a mobile device (not shown), such as a mobile computing device 850. Each of such devices may contain one or more of the computing device 800 and the mobile computing device 850, and an entire system may be made up of multiple computing devices communicating with each other.

[0157] The mobile computing device 850 includes a processor 852, a memory 864, an input / output device such as a display 854, a communication interface 866, and a transceiver 868, among other components. The mobile computing device 850 may also be provided with a storage device, such as a micro-drive or other device, to provide additional storage. Each of the processor 852, the memory 864, the display 854, the communication interface 866, and the transceiver 868, are interconnected using various buses, and several of the components may be mounted on a common motherboard or in other manners as appropriate.Attorney’s Docket No. 2010358-0388

[0158] The processor 852 can execute instructions within the mobile computing device 850, including instructions stored in the memory 864. The processor 852 may be implemented as a chipset of chips that include separate and multiple analog and digital processors. The processor 852 may provide, for example, for coordination of the other components of the mobile computing device 850, such as control of user interfaces, applications run by the mobile computing device 850, and wireless communication by the mobile computing device 850.

[0159] The processor 852 may communicate with a user through a control interface 858 and a display interface 856 coupled to the display 854. The display 854 may be, for example, a TFT (Thin-Film-Transistor Liquid Crystal Display) display or an OLED (Organic Light Emitting Diode) display, or other appropriate display technology. The display interface 856 may comprise appropriate circuitry for driving the display 854 to present graphical and other information to a user. The control interface 858 may receive commands from a user and convert them for submission to the processor 852. In addition, an external interface 862 may provide communication with the processor 852, so as to enable near area communication of the mobile computing device 850 with other devices. The external interface 862 may provide, for example, for wired communication in some implementations, or for wireless communication in other implementations, and multiple interfaces may also be used.

[0160] The memory 864 stores information within the mobile computing device 850. The memory 864 can be implemented as one or more of a computer-readable medium or media, a volatile memory unit or units, or a non-volatile memory unit or units. An expansion memory 874 may also be provided and connected to the mobile computing device 850 through an expansion interface 872, which may include, for example, a SIMM (Single In Line Memory Module) card interface. The expansion memory 874 may provide extra storage space for theAttorney’s Docket No. 2010358-0388mobile computing device 850, or may also store applications or other information for the mobile computing device 850. Specifically, the expansion memory 874 may include instructions to carry out or supplement the processes described above, and may include secure information also. Thus, for example, the expansion memory 874 may be provided as a security module for the mobile computing device 850, and may be programmed with instructions that permit secure use of the mobile computing device 850. In addition, secure applications may be provided via the SIMM cards, along with additional information, such as placing identifying information on the SIMM card in a non-hackable manner.

[0161] The memory may include, for example, flash memory and / or NVRAM memory (nonvolatile random access memory), as discussed below. In some implementations, instructions are stored in an information carrier. The instructions, when executed by one or more processing devices (for example, processor 852), perform one or more methods, such as those described above. The instructions can also be stored by one or more storage devices, such as one or more computer- or machine-readable mediums (for example, the memory 864, the expansion memory 874, or memory on the processor 852). In some implementations, the instructions can be received in a propagated signal, for example, over the transceiver 868 or the external interface 862.

[0162] The mobile computing device 850 may communicate wirelessly through the communication interface 866, which may include digital signal processing circuitry where necessary. The communication interface 866 may provide for communications under various modes or protocols, such as GSM voice calls (Global System for Mobile communications), SMS (Short Message Service), EMS (Enhanced Messaging Service), or MMS messaging (Multimedia Messaging Service), CDMA (code division multiple access), TDMA (time division multipleAttorney’s Docket No. 2010358-0388access), PDC (Personal Digital Cellular), WCDMA (Wideband Code Division Multiple Access), CDMA2000, or GPRS (General Packet Radio Service), among others. Such communication may occur, for example, through the transceiver 868 using a radio-frequency. In addition, short-range communication may occur, such as using a Bluetooth®, Wi-Fi™, or other such transceiver (not shown). In addition, a GPS (Global Positioning System) receiver module 870 may provide additional navigation- and location-related wireless data to the mobile computing device 850, which may be used as appropriate by applications running on the mobile computing device 850.

[0163] The mobile computing device 850 may also communicate audibly using an audio codec 860, which may receive spoken information from a user and convert it to usable digital information. The audio codec 860 may likewise generate audible sound for a user, such as through a speaker, e.g., in a handset of the mobile computing device 850. Such sound may include sound from voice telephone calls, may include recorded sound (e.g., voice messages, music files, etc.) and may also include sound generated by applications operating on the mobile computing device 850.

[0164] The mobile computing device 850 may be implemented in a number of different forms, as shown in the figure. For example, it may be implemented as a cellular telephone 880. It may also be implemented as part of a smart-phone 882, personal digital assistant, or other similar mobile device.

[0165] Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, specially designed ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at leastAttorney’s Docket No. 2010358-0388one programmable processor, which may be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0166] These computer programs (also known as programs, software, software applications or code) include machine instructions for a programmable processor, and can be implemented in a high-level procedural and / or object-oriented programming language, and / or in assembly / 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., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and / or data to a 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 used to provide machine instructions and / or data to a programmable processor.

[0167] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0168] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middlewareAttorney’s Docket No. 2010358-0388component (e.g., an application server), or that 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 here), or any combination of such back end, middleware, or front end components. The components of the system can 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.

[0169] The computing system can include clients and servers. A client and server are generally 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.

[0170] In some implementations, various modules described herein can be separated, combined or incorporated into single or combined modules. Modules depicted in the figures are not intended to limit the systems described herein to the software architectures shown therein.

[0171] Elements of different implementations described herein may be combined to form other implementations not specifically set forth above. Elements may be left out of the processes, computer programs, databases, etc. described herein without adversely affecting their operation. In addition, the logic flows depicted in the figures do not require the particular order shown, or sequential order, to achieve desirable results. Various separate elements may be combined into one or more individual elements to perform the functions described herein.Attorney’s Docket No. 2010358-0388G. Imaging Agents

[0172] As described herein, a variety of radionuclide labelled PSMA binding agents may be used as radiopharmaceutical imaging agents for nuclear medicine imaging to detect and evaluate prostate cancer. In certain embodiments, certain radionuclide labelled PSMA binding agents are appropriate for PET imaging, while others are suited for SPECT imaging.G.i. PET imaging radionuclide labelled PSMA binding agents

[0173] In certain embodiments, a radionuclide labelled PSMA binding agent is a radionuclide labelled PSMA binding agent appropriate for PET imaging.

[0174] In certain embodiments, a radionuclide labelled PSMA binding agent comprises[18F]DCFPyL (also referred to as PyL™; also referred to as DCFPyL-18F):[18F]DCFPyL,or a pharmaceutically acceptable salt thereof.

[0175] In certain embodiments, a radionuclide labelled PSMA binding agent comprises[18F]DCFBC:Attorney’s Docket No. 2010358-0388[18F]DCFBC,or a pharmaceutically acceptable salt thereof.

[0176] In certain embodiments, a radionuclide labelled PSMA binding agent comprises68Ga- PSMA-HBED-CC (also referred to as68Ga-PSMA-l 1):68Ga-PSMA-HBED-CC,or a pharmaceutically acceptable salt thereof.

[0177] In certain embodiments, a radionuclide labelled PSMA binding agent comprisesPSMA-617:Attorney’s Docket No. 2010358-0388PSMA-617,or a pharmaceutically acceptable salt thereof. In certain embodiments, the radionuclide labelled PSMA binding agent comprises68Ga-PSMA-617, which is PSMA-617 labelled with68Ga, or a pharmaceutically acceptable salt thereof. In certain embodiments, a radionuclide labelled PSMA binding agent comprises177Lu-PSMA-617, which is PSMA-617 labelled with177Lu, or a pharmaceutically acceptable salt thereof.

[0178] In certain embodiments, a radionuclide labelled PSMA binding agent comprisesPSMA-I& T:Attorney’s Docket No. 2010358-0388PSMA-I& T,or a pharmaceutically acceptable salt thereof. In certain embodiments, a radionuclide labelledPSMA binding agent comprises68Ga-PSMA-I& T, which is PSMA-I& T labelled with68Ga, or a pharmaceutically acceptable salt thereof.

[0179] In certain embodiments, a radionuclide labelled PSMA binding agent comprisesPSMA- 1007:Attorney’s Docket No. 2010358-0388or a pharmaceutically acceptable salt thereof. Tn certain embodiments, a radionuclide labelledPSMA binding agent comprises 18F-PSMA-1007, which is PSMA-1007 labelled with 18F, or a pharmaceutically acceptable salt thereof.

[0180] In certain embodiments, a radionuclide labeled PSMA binding agent comprises 18F- JK-PSMA-7:18F-JK-PSMA-7,or a pharmaceutically acceptable salt thereof.

[0181] In certain embodiments, a radionuclide labeled PSMA binding agent comprises (18F) rhPSMA-7.3 (e.g., POSLUMA®, also described at https: / / www.posluma.com / prescribing-information.pdf):Attorney’s Docket No. 2010358-0388(18F) rhPSMA-7.3,or a pharmaceutically acceptable salt thereof.G.ii. SPECT imaging radionuclide labelled PSMA binding agents

[0182] In certain embodiments, a radionuclide labelled PSMA binding agent is a radionuclide labelled PSMA binding agent appropriate for SPECT imaging.

[0183] In certain embodiments, a radionuclide labelled PSMA binding agent comprises 1404 (also referred to as MIP-1404):1404,or a pharmaceutically acceptable salt thereof.

[0184] In certain embodiments, a radionuclide labelled PSMA binding agent comprises 1405(also referred to as MIP-1405):Attorney’s Docket No. 2010358-03881405,or a pharmaceutically acceptable salt thereof.

[0185] In certain embodiments, a radionuclide labelled PSMA binding agent comprises 1427(also referred to as MIP-1427):1427,or a pharmaceutically acceptable salt thereof.

[0186] In certain embodiments, a radionuclide labelled PSMA binding agent comprises 1428(also referred to as MIP-1428):Attorney’s Docket No. 2010358-03881428,or a pharmaceutically acceptable salt thereof.

[0187] In certain embodiments, a PSMA binding agent is labelled with a radionuclide by chelating it to a radioisotope of a metal [e.g., a radioisotope of technetium (Tc) (e.g., technetium-99m (99mTc)); e.g., a radioisotope of rhenium (Re) (e.g., rhenium-188 (188Re); e.g., rhenium-186 (186Re)); e g., a radioisotope of yttrium (Y) (e.g., 90Y); e.g., a radioisotope of lutetium (Lu)(e.g., 177Lu); e.g., a radioisotope of gallium (Ga) (e.g., 68Ga; e.g., 67Ga); e.g., a radioisotope of indium (e.g., 11 Un); e.g., a radioisotope of copper (Cu) (e.g., 67Cu)].

[0188] In certain embodiments, 1404 is labelled with a radionuclide (e.g., chelated to a radioisotope of a metal). In certain embodiments, a radionuclide labelled PSMA binding agent comprises "mTc-MIP-1404, which is 1404 labelled with (e.g., chelated to) "mTc:Attorney’s Docket No. 2010358-038899mTc-MIP-1404,or a pharmaceutically acceptable salt thereof. In certain embodiments, 1404 may be chelated to other metal radioisotopes [e.g., a radioisotope of rhenium (Re) (e.g., rhenium-188 (188Re); e.g., rhenium-186 (186Re)); e.g., a radioisotope of yttrium (Y) (e.g., 90 Y); e.g., a radioisotope of lutetium (Lu)(e.g., 177Lu); e.g., a radioisotope of gallium (Ga) (e.g., 68Ga; e.g., 67Ga); e.g., a radioisotope of indium (e.g., 11 Un); e.g., a radioisotope of copper (Cu) (e.g., 67Cu)] to form a compound having a structure similar to the structure shown above for 99mTc-MIP-1404, with the other metal radioisotope substituted for 99mTc.

[0189] In certain embodiments, 1405 is labelled with a radionuclide (e.g., chelated to a radioisotope of a metal). In certain embodiments, a radionuclide labelled PSMA binding agent comprises "mTc-MIP-1405, which is 1405 labelled with (e.g., chelated to) "mTc:Attorney’s Docket No. 2010358-038899mTc-MIP-1405,or a pharmaceutically acceptable salt thereof. In certain embodiments, 1405 may be chelated to other metal radioisotopes [e.g., a radioisotope of rhenium (Re) (e.g., rhenium-188 (188Re); e.g., rhenium-186 (186Re)); e.g., a radioisotope of yttrium (Y) (e.g.,90Y); e.g., a radioisotope of lutetium (Lu)(e.g., 177Lu); e.g., a radioisotope of gallium (Ga) (e.g., 68Ga; e.g., 67Ga); e.g., a radioisotope of indium (e.g., 11 Un); e.g., a radioisotope of copper (Cu) (e.g., 67Cu)] to form a compound having a structure similar to the structure shown above for 99mTc-MIP-1405, with the other metal radioisotope substituted for 99mTc.

[0190] In certain embodiments, 1427 is labelled with (e.g., chelated to) a radioisotope of a metal, to form a compound according to the formula below:Attorney’s Docket No. 2010358-0388or a pharmaceutically acceptable salt thereof, wherein M is a metal radioisotope [e.g., a radioisotope of technetium (Tc) (e.g., technetium-99m (99mTc)); e.g., a radioisotope of rhenium (Re) (e.g., rhenium-188 (188Re); e.g., rhenium-186 (186Re)); e.g., a radioisotope of yttrium (Y) (e.g., 90Y); e.g., a radioisotope of lutetium (Lu)(e.g., 177Lu); e.g., a radioisotope of gallium (Ga) (e.g., 68Ga; e.g., 67Ga); e.g., a radioisotope of indium (e g., 11 Un); e.g., a radioisotope of copper (Cu) (e.g., 67Cu)] with which 1427 is labelled.

[0191] In certain embodiments, 1428 is labelled with (e.g., chelated to) a radioisotope of a metal, to form a compound according to the formula below:1428 chelated to a metal,or a pharmaceutically acceptable salt thereof, wherein M is a metal radioisotope [e.g., a radioisotope of technetium (Tc) (e.g., technetium-99m (99mTc)); e.g., a radioisotope of rhenium (Re) (e.g., rhenium-188 (188Re); e.g., rhenium-186 (186Re)); e.g., a radioisotope of yttrium (Y) (e.g., 90Y); e.g., a radioisotope of lutetium (Lu)(e.g., 177Lu); e.g., a radioisotope of gallium (Ga) (e.g., 68Ga; e.g., 67Ga); e.g., a radioisotope of indium (e g., 11 Un); e.g., a radioisotope of copper (Cu) (e.g., 67Cu)] with which 1428 is labelled.Attorney’s Docket No. 2010358-0388

[0192] In certain embodiments, a radionuclide labelled PSMA binding agent comprisesPSMA I& S:or a pharmaceutically acceptable salt thereof. In certain embodiments, a radionuclide labelled PSMA binding agent comprises 99mTc-PSMA I& S, which is PSMA I& S labelled with 99mTc, or a pharmaceutically acceptable salt thereof.

[0193] Throughout the description, where apparatus and systems are described as having, including, or comprising specific components, or where processes and methods are described as having, including, or comprising specific steps, it is contemplated that, additionally, there are apparatus, and systems of the present invention that consist essentially of, or consist of, the recited components, and that there are processes and methods according to the present invention that consist essentially of, or consist of, the recited processing steps.

[0194] It should be understood that the order of steps or order for performing certain action is immaterial so long as the invention remains operable. Moreover, two or more steps or actions may be conducted simultaneously.Attorney’s Docket No. 2010358-0388EXAMPLESExample 1. New features and improvements of PYLCLARI Al® Version 2.5.0

[0195] This example describes new features and improvements of the nuclear medicine imaging software, PYLCLARI Al® Version 2.5.0, in accordance with certain embodiments described herein. PYLCLARI Al® is an artificial intelligence-powered assistant for whole body automated prostate-specific membrane antigen (PSMA) quantification and standardized reporting.

[0196] Outlined hotspots. This feature is shown in FIG. 9. Hotspots in planar views (901) now have a clear outline (902), making it easier to see the underlying uptake without needing to hide the hotspot itself.

[0197] Color-coded MIP. This feature is shown in FIGs. 10A and 10B. In the MIP, selected hotspots are now color-coded based on their TNM type (for example, 1001, 1002, 1003, 1051, 1052, 1053), giving the user a quick visual overview of the distribution. When using filters, any hotspots outside the user’s set criteria will appear dimmed, making it easier to focus on your selection. For example, FIG. 10A depicts a composite grayscale anatomical image with color-coded overlay showing identified lesions that are color-coded depending on the TNM type, shown in the legend at lower left. The color and number of hotspots per type is displayed in the legend next to the MIP image. In this example, application of a filter results in certain of the orange-coded miMb lesions in the overlay of original FIG. 10A that do not meet the filter criteria to appear as brown (de-emphasized) in the overlay shown in FIG. 10B. Similarly, certain of the purple miN lesions in FIG. 10A that do not meet the filter criteria appear as a darker hue in FIG.10B.Attorney’s Docket No. 2010358-0388

[0198] This feature is also demonstrated in FIGs. 11 and 12. The feature improves the visual distinction between tissue types of the selected hotspots by applying a color-coded colormap to the Maximum Intensity Projection (MIP) view. When a specific tissue type is selected or changed, corresponding colored voxels are instantly updated and overlaid onto the grayscale PET images, enabling users to clearly differentiate between hotspot type categories in real time. A legend is shown in the MIP view, indicating the color assigned to each tissue type. FIG. 11 shows a front and side view of the color-coded MIP with no filter applied. To focus on a specific set of hotspots, a user can apply a filter to the hotspot list. Hotspots that do not meet the filter criteria are visually dimmed, enhancing clarity by making the selected hotspots stand out more clearly. FIG. 12 shows a front and side view of the color-coded MIP with a filter applied.Lesions that do not meet the filter criteria are depicted in darker corresponding hues. In this example, the available hotspot types are miT (local tumor), miN (regional nodes), miMa (distant metastasis, extrapelvic lymph nodes), miMb (distant metastasis, bones), and miMc (distant metastasis, other sites).

[0199] Parallel study review in single window. This feature is shown in FIGs. 13-1, 13-2, 13-3, and 13-4 (collectively referred to as FIG. 13). Reviewing two studies within a single window is now supported and set as the default view. This layout is ideal when working on a single monitor, as both studies are displayed layered on top of each other. To focus on one study at a time, Study A or B can be selected from the toolbar. If reviewing studies in separate windows is still preferred, this setting can be adjusted under Settings (top-right comer) — > Default Review Settings, by enabling the Review studies in separate windows option.

[0200] Improved hotspot segmentation. This feature is shown in FIGs. 14-1 and 14-2 (collectively referred to as FIG. 14). Hotspots in the liver and spleen are segmented using anAttorney’s Docket No. 2010358-0388adaptive threshold based on their SUVmax and the background uptake of the organ, improving their delineation.

[0201] Review toolbar. This feature is shown in FIG. 15. A toolbar has been added to the top of the review page, which contains the following information and controls:• Patient name and ID ( 1501 )• Study view selector (only present in parallel study review in single window) (1502)• Sync studies toggle (only present in parallel study review) (1503)• Help page (1504)• Settings page (1505)

[0202] Toggle full-screen on double-click. Double-clicking within any viewport now toggles between full-screen and split-screen modes, allowing for a quicker and more flexible viewing experience.

[0203] Liver windowing preset. A new CT windowing preset optimized for liver imaging has been added to the available options.

[0204] Automatic series selection. When a study containing multiple CT and PT series is sent to the application, the most suitable series pair is now selected automatically. Preprocessing begins immediately, streamlining the workflow. This feature is particularly useful when multiple series are sent from the PACS.

[0205] Keyboard layout support. The application now supports multiple keyboard layouts.This update resolves issues with incorrect shortcut mappings, particularly on AZERTY keyboards.

[0206] Improvements to local installations and updates. The local installation process has been optimized with a reduced storage footprint, making deployments lighter and more efficient.Attorney’s Docket No. 2010358-0388Local update handling and PACS configuration has also been improved, simplifying maintenance and reducing overhead for IT administrators.EQUIVALENTS

[0207] It is to be understood that while the disclosure has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the claims. Other aspects, advantages, and modifications are within the scope of the claims.

[0208] This written description uses examples to disclose the invention, including the best mode, and also to enable any person skilled in the art to practice the present embodiments, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the present embodiments is defined by the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they include structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal languages of the claims.

Claims

Attorney’s Docket No. 2010358-0388What is claimed is:

1. A method for interactive display of a medical image of a subject, the method comprising:rendering, by a processor of a computing device, a first instantiation of the medical image (e.g., a PET image, a CT image, a PET / CT image, or aPET / SPECT image, e.g., a 3-dimensional image (3D image)) for graphical display, said first instantiation comprising a black-and-white (B& W) and / or grayscale or otherwise monochrome depiction of an anatomy of the subject with a first overlay superimposed on the depiction of the anatomy, said first overlay comprising a first color rendering of regions (e g., 3D regions) corresponding to a set of hotspots (e g., where hotspots are regions of suspected cancerous lesions, e.g., prostate cancer, e.g., metastatic prostate cancer);causing, by the processor, display of one or more graphical control elements (widgets) allowing for user selection (e.g., setting or adjustment) of one or more criteria for application of a graphical display filter to adjust the first instantiation of the medical image;receiving, by the processor, via the one or more widgets, a user selection of the one or more criteria for applying the graphical display filter to adjust the first instantiation of the medical image; andrendering, by the processor, a second instantiation of the medical image, said second instantiation reflecting a product of the adjustment of the first instantiation of the medical image following application of the graphical display filter,wherein the second instantiation comprises the B& W and / or grayscale or otherwise monochrome depiction of the anatomy of the subject with a second overlay superimposed on the depiction of the anatomy, said second overlay comprising a second color rendering of the regions corresponding to the set of hotspots depicted in the first overlay.Attorney’s Docket No. 2010358-0388. The method of claim 1, wherein (e.g., in the first color rendering and / or the second color rendering) the (e.g., regions corresponding to the) set of hotspots is color-coded based on lesion type [e.g., wherein each of the regions of the first and / or second color rendering corresponds to a particular hotspot of the set of hotspots, each hotspot of the set of hotspots is assigned a particular lesion type, and each region in the first and / or second color rendering is color-coded based on the particular lesion type assigned to the corresponding particular hotspot],3. The method of claim 2, wherein the lesion type comprises a TNM type (tumor, node, and metastasis type).

4. The method of claim 2 or 3, wherein each hotspot of the set of hotspots is assigned a particular lesion type and, within the first color rendering and / or the second color rendering, each of the regions is color-coded according to the lesion type assigned to the corresponding hotspot (e.g., a given region corresponds to a particular hotspot of the set and is color-coded to reflect the lesion type assigned to the particular hotspot).

5. The method of claim 4, wherein the particular lesion type assigned to each hotspot is selected from a list of possible lesion types, said list of possible lesion types comprising one or more members selected from the group consisting of: miT (local, e g., within a prostate), miN (pelvic lymph node), miMa (distant lymph node), miMb (bone), and miMc (distant organ).Attorney’s Docket No. 2010358-03886. The method of claim 4, wherein the particular lesion type assigned to each hotspot is selected from the group consisting of miT (local, e.g., within a prostate), miN (pelvic lymph node), miMa (distant lymph node), miMb (bone), and miMc (distant organ).

7. The method of any one of the preceding claims, wherein the one or more criteria for applying the graphical display filter comprises a (e.g., adjustable) range of one or more measurements associated with each hotspot of the set of hotspots.

8. The method of claim 7, wherein the one or more measurements comprise one or more members selected from the group consisting of: (i) a Standardized Uptake Value (SUV) (e.g., quantifying an intensity or SUV of each hotspot, for example, SUVmean, SUVmax, and / or SUVpeak), (ii) a lesion size (e.g., volume), and (iii) a Lesion Index (LI).

9. The method of any one of the preceding claims, wherein the one or more criteria for applying the graphical display filter comprises a (e.g., adjustable) maximum and / or a minimum of one or more measurements associated with each hotspot of the set of hotspots.

10. The method of claim 9, wherein the one or more measurements comprise one or more members selected from the group consisting of (i) a Standardized Uptake Value (SUV) (e.g., quantifying an intensity or SUV of each hotspot, for example, SUVmean, SUVmax, and / or SUVpeak), (ii) a lesion size (e.g., volume), and (iii) a Lesion Index (LI).Attorney’s Docket No. 2010358-038811. The method of any one of the preceding claims, wherein, upon application of the filter, color of the regions corresponding to the set of hotspots depicted in the first overlay is adjusted to (i) visually highlight or otherwise graphically emphasize one or more hotspots out of the set of hotspots that satisfy the one or more criteria and / or (ii) visually / graphically de-emphasize one or more hotspots out of the set of hotspots that do not satisfy the one or more criteria, thereby creating the second overlay.

12. The method of any one of the preceding claims, wherein at least a portion of the set of hotspots (e.g., of the first and / or second overlay) are determined using a machine learning algorithm.

13. The method of claim 12, wherein the machine learning algorithm comprises a supervised deep learning neural network or a convolutional neural network.

14. The method of any one of the preceding claims, wherein at least a portion of the set of hotspots of the first (e.g., and / or second) overlay is user-defined (e.g., a user may draw a hotspot).

15. The method of claim 14, wherein at least a part of the one or more criteria is determined based on the user-defined hotspots.

16. The method of any one of the preceding claims, wherein the medical image comprises a Positron Emission Tomography (PET) image.Attorney’s Docket No. 2010358-038817. The method of any one of the preceding claims, wherein the medical image comprises a computed tomography (CT) image.

18. The method of any one of the preceding claims, wherein the medical image comprises a PET / CT image or a SPECT / CT image.

19. The method of any one of the preceding claims, wherein the medical image comprises a composite image, said composite image comprising an anatomical image (e.g., a CT image) and a corresponding functional image (e.g., a nuclear medicine image (e.g., a PET image; e.g., a SPECT image)).

20. The method of any one of the preceding claims, wherein the medical image comprises a 3 -dimensional image (3D image).

21. The method of any one of the preceding claims, wherein the hotspots are regions of suspected cancerous lesions (e.g., prostate cancer, metastatic prostate cancer).

22. The method of any one of the preceding claims, wherein the second color rendering adjusts a color of one or more of the regions according to the one or more criteria of the graphical display filter.

23. A system for interactive display of a medical image of a subject, the system comprising:a processor of a computing device; andAttorney’s Docket No. 2010358-0388a memory having instructions stored thereon, wherein the instructions, when executed by the processor, cause the processor to:render a first instantiation of the medical image (e.g., a PET image, a CT image, a PET / CT image, or a PET / SPECT image, e.g., a 3-dimensional image (3D image)) for graphical display, said first instantiation comprising a black-and-white (B& W) and / or grayscale or otherwise monochrome depiction of an anatomy of the subject with a first overlay superimposed on the depiction of the anatomy, said first overlay comprising a first color rendering of regions (e.g., 3D regions) corresponding to a set of hotspots (e.g., where hotspots are regions of suspected cancerous lesions, e.g., prostate cancer, e.g., metastatic prostate cancer);cause display of one or more graphical control elements (widgets) allowing for user selection (e.g., setting or adjustment) of one or more criteria for application of a graphical display filter to adjust the first instantiation of the medical image;receive, via the one or more widgets, a user selection of the one or more criteria for applying the graphical display filter to adjust the first instantiation of the medical image; andrender a second instantiation of the medical image, said second instantiation reflecting a product of the adjustment of the first instantiation of the medical image following application of the graphical display filter,wherein the second instantiation comprises the B& W and / or grayscale or otherwise monochrome depiction of the anatomy of the subject with a second overlay superimposed on the depiction of the anatomy, said second overlay comprising a second color rendering of the regions corresponding to the set of hotspots depicted in the first overlay.