Devices for monitoring side effects of treatment

Abstract

A system for monitoring organ health during treatment for cancer and the like that utilizes physiological imaging of the type used for treatment monitoring and organ-specific processing to provide a comprehensive assessment of treatment side effects.

Description

[0001] STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] Cross-reference to related applications

[0003] This application claims the benefit of U.S. Provisional Application No. 63 / 021,936, filed May 8, 2020, and U.S. Non-Provisional Application No. 17 / 236,715, filed April 21, 2021, which are incorporated herein by reference. Background Art

[0004] The present invention relates to medical imaging equipment that provides physiological imaging (eg, PET imaging, functional MRI), and in particular to an apparatus that uses physiological imaging to automatically assess toxic side effects during treatment of cancer or the like.

[0005] Medical treatments (e.g., those using immunotherapy) may have side effects severe enough to require discontinuation of treatment. In one example, when using immune checkpoint inhibitors (ICIs) to treat cancer, patients may experience immune-mediated colitis with symptoms including diarrhea. Grade 2 or 3 colitis requires postponing ICI treatment, while grade 4 colitis can be life-threatening and require permanent discontinuation of treatment.

[0006] The impact of such side effects on treatment can be reduced by carefully monitoring them during treatment. Unfortunately, the number of different possible side effects can be large, and many common side effects are difficult to monitor. For example, the diagnosis of immune-mediated colitis may require a colonoscopy or colon biopsy, which is impractical for routine repeated monitoring. Summary of the Invention

[0007] The present inventors have recognized that physiological imaging, commonly performed to monitor tumor regression during cancer treatment, can be used along with organ-specific toxicity rules to simultaneously provide insight into the health of non-involved organs.

[0008] In one example, you can use 18 F-FDG uptake (the same molecular imaging agent used to track tumors) is used to monitor intestinal health. The present invention uses data from structural imaging scanners (such as CT machines) that provide anatomical information to segment organs to focus analysis on molecular drug uptake data (such as from PET machines) or organ function data (such as from MRI machines) of the organ, applying toxicity rules to the organ to perform organ health assessment. Multiple organs can be analyzed simultaneously to provide a comprehensive overview of organ health.

[0009] In contrast to structural imaging (standard CT and MRI imaging) which provides anatomical images, physiological imaging as used herein refers to imaging techniques (e.g., PET imaging, functional MRI, or MRS) that detect molecular changes (e.g., altered metabolism), functional changes (e.g., altered blood flow), or chemical changes (e.g., local chemical composition and uptake).

[0010] In one embodiment, the present invention provides an apparatus for assessing organ health during patient treatment, the apparatus having an electronic computer that executes a stored program to: (a) receive a structural image of at least one organ of the patient; (b) receive a physiological image of at least one organ indicative of organ function; (c) process the structural image to create a mask describing the organ; (d) use the mask to select a portion of the physiological image relevant to the organ; (e) apply a toxicity rule specific to the organ to the portion of the physiological image to provide an assessment of organ health; and / or (f) output an organ health indication based on the assessment.

[0011] It is thus a feature of at least one embodiment of the invention to provide an automated system that can utilize often pre-existing physiological image information to characterize side effects that may affect treatment.

[0012] The stored program may also process portions of the physiological image to identify lesions and refine the organ mask to remove the lesions before the toxicity rules are applied.

[0013] It is therefore a feature of at least one embodiment of the invention to enable monitoring of organ health even when the organ includes lesions that are being treated.By segmenting and isolating these lesions, sensitivity to organ health is increased.

[0014] Processing of the structural image to create the mask may associate the mask with an organ type, and identification of lesions in the organ may be based on lesion identification rules associated with the organ type.

[0015] It is thus a feature of at least one embodiment of the invention to provide a sophisticated quantitative analysis of uptake information about organ health that is difficult to obtain simply by observing physiological images due to variations in how the physiological images relate to organ health.

[0016] The electronic computer executing the stored program can also process the structural image to identify a lymph node region mask, which is used to select a portion of the physiological scan related to the lymph nodes. The lymph node activity rule can then be applied to the portion of the physiological image related to the lymph nodes to assess the activity of the lymph nodes, and the output can provide a lymph node activity indication indicating activation and / or stress of the lymph node system.

[0017] It is thus a feature of at least one embodiment of the invention to provide an early measure of organ health by assessing whether lymph nodes are activated or overburdened.

[0018] More generally, the stored program can provide a set of toxicity rules associated with different organs, wherein different toxicity rules are applied to different organs to provide an assessment of organ health for a plurality of different organs. In this case, the output indicates the organ health of a plurality of organs (e.g., a first and a second organ; a first, a second, and a third organ; a first, a second, a third, and a fourth organ, etc.).

[0019] Thus, it is a feature of at least one embodiment of the present invention to provide healthcare professionals with a comprehensive overview of organ health that would otherwise require a daunting battery of tests. By automatically segmenting organs and using organ-specific rules, it becomes feasible to assess organ health of multiple organs simultaneously.

[0020] The output may provide a composite measure of organ health for multiple organs.

[0021] It is thus a feature of at least one embodiment of the invention to provide a healthcare provider with an immediate indication of whether the health of an organ is significantly affected.

[0022] Alternatively or additionally, the output may provide a structural image-based image augmented with organ health data.

[0023] It is thus a feature of at least one embodiment of the invention to utilize image data for segmentation to provide a framework for communicating the system's analysis to a healthcare provider.

[0024] The stored program may retain previous output indicating organ health associated with previous structural images and previous physiological images to provide a display of organ health trends for a plurality of organs.

[0025] It is thus a feature of at least one embodiment of the invention to provide important trending information that enables early response to symptoms of toxicity and that would be difficult or impossible to assess by simply reviewing individual uptake scans.

[0026] The toxicity model may include thresholds or other characteristics extracted from the physiological images to predict organ health for a plurality of different organs, and the output may indicate the predicted organ health for the plurality of organs.

[0027] It is thus a feature of at least one embodiment of the invention to incorporate empirical understanding of current organ health to estimate a likely prognosis.

[0028] The electronic computer may also analyze the physiological scans to assess changes in cancerous lesions, and the output may provide an indication of changes in cancerous lesions.

[0029] It is thus a feature of at least one embodiment of the invention to provide an integrated display of treatment efficacy and side effects to provide improved guidance during treatment.

[0030] These specific objects and advantages may apply only to some embodiments falling within the claims and therefore do not limit the scope of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a block diagram of an example embodiment of the present invention showing a CT and PET scanner in communication with a central processor having a display terminal and executing a stored program to provide organ health information;

[0032] Figure 2 is a flow chart illustrating the processing of data from CT and PET scanners by a central processor to provide an assessment of organ health;

[0033] Figure 3 Is the CPU about Figure 2 A flowchart of a program implemented by the flowchart of

[0034] Figure 4 is used Figure 2 and Figure 3 The processing in Figure 1 The example display produced on the display terminal is as follows. DETAILED DESCRIPTION

[0035] Now refer to Figure 1 A system 10 for assessing organ health during disease treatment may provide a structural imaging scanner 12, such as a kilovoltage or megavoltage CT (computed tomography) scanner, an MRI (magnetic resonance imaging) scanner, or the like, which may provide a high-resolution structural image scan 14 presenting anatomical information about a patient 15.

[0036] Additionally, the system 10 may provide a physiological imaging scanner 16 that may scan the patient 15 after, for example, introduction of a molecular imaging uptake agent 17 to measure uptake of the uptake agent 17 .

[0037] In one example, the physiological imaging scanner 16 can be a PET (positron emission tomography) scanner. As is generally understood in the art, PET is a nuclear medicine imaging technique that produces physiological scans 18 that reveal molecular processes within the patient 15 as reflected by the preferential migration of the uptake agent 17 to tumor tissue. An example molecular imaging uptake agent 17 is 2-deoxy-2-[fluoro-18]fluoro-D-glucose ( 18F-FDG PET / CT). However, the PET scanner is only one example of a physiological imaging scanner 16, and the present invention contemplates the use of a physiological imaging scanner 16 that includes a functional CT or functional MRI machine or other similar equipment that measures the underlying metabolism of tissue.

[0038] Typically, the physiological imaging scanner 16 will produce a physiological scan 18 having a lower spatial resolution than the high-resolution structural image scan 14 from the structural imaging scanner 12. In each case, the structural image scan 14 and the pharmacological physiological scan 18 will present a dimension of information associated with volume elements (voxels) distributed three-dimensionally within the volumetric region of interest in the patient 15.

[0039] In the present invention, patient 15 may be scanned simultaneously in both structural imaging scanner 12 and physiological imaging scanner 16, which in some cases may be the same machine using different hardware or protocols. These scans will be repeated at different times throughout the course of patient 15's treatment, for example, between treatment sessions of patient 15 with chemotherapy, radiation therapy, etc.

[0040] Still refer to Figure 1 , the structural image scan 14 and the pharmacological physiological scan 18 are received by an electronic computer 22 for processing, as will be described in more detail below. Generally, the electronic computer 22 includes one or more processing units 24 in communication with a memory 26 that stores data and stored programs 28 for implementing portions of the present invention. The computer 22 can be in communication with a graphics display 30 for displaying color output images based on the structural image scan 14 and the pharmacological physiological scan 18, and can also be in communication with user input devices 32, such as a keyboard, mouse, etc., each of which allows a user to enter data. As will be discussed in further detail below, the present invention provides an output on the display 30 that indicates organ health and lesion progression or regression based on measurements of pharmacological uptake within multiple tumors and multiple organ locations of the patient 15.

[0041] Now also refer to Figure 2 and Figure 3 At a first step, indicated by process block 34, the stored program 28 receives scan data obtained from the structural imaging scanner 12 and the physiological imaging scanner 16 as discussed above.

[0042] according to Figure 3At process block 36 in FIG. 1 , the structural image scan 14 from the structural imaging scanner 12 is transmitted to a set of segmenters 38 a to 38 b, which operate to generate masks 40 a to 40 c that isolate portions of the structural image scan 14 associated with specific organs. In this regard, each of the segmenters 38 can be tuned to a specific organ, such as the heart, lungs, brain, liver, kidneys, bones, spleen, stomach, pancreas, pharynx, larynx, blood vessels, muscles, gallbladder, intestines, lymph nodes, bone marrow, bladder, etc., so that the mask 40 defines a specific organ volume for each organ of a particular patient 15, but also allows for easy identification of these organs. For clarity, only three segmenters 38 are shown; the number of segmenters 38 can be increased or decreased as needed or desired to provide coverage of all organs or organ regions of interest.

[0043] In one embodiment, the segmenter 38 can be implemented as a convolutional neural network (CNN), which is trained, preprocessed (e.g., normalized), and then segmented to identify organs of these patients using a training set of structural image scans 14 acquired from different patients. This training set is then used to train the neural network to generate a set of weights 42a to 42c specific to different organ types, which, as is generally understood in the art, can generate segmentations for these organs. In one embodiment, the neural network architecture can be the Deep Medic architecture described by Kamnitsas K., Ledig C, Newcombe VF et al. in "Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation," Medical Image Analysis 2017; 36: 61-78, which is incorporated herein by reference. The present invention contemplates that other segmentation systems that provide automatic segmentation of organ volumes can be used. In addition, although a separate segmenter 38 is shown, it should be understood that common hardware can be used and the process can be performed sequentially by changing specific weights 42.

[0044] The information for the masks 40 for the various organs is then received by the corresponding lesion identifiers 44a to 44c, which also receive the physiological scan 18 and / or the structural scan 14 (indicated by the dashed lines). The lesion identifiers 44a to 44c are also organ-specific and have lesion identification rules 45a to 45c associated with the specific organs, describing, for example, uptake characteristics and lesion size and shape characteristics associated with lesions in those organs to help identify the lesions according to the organ's size and shape. Figure 3 Process block 48 automatically segments and identifies lesions within the received mask 40. Techniques for automatic lesion identification can look for localized areas of excessive uptake, for example, as described in U.S. patent application 2016 / 0100795, assigned to the assignee of the present invention and incorporated herein by reference.

[0045] The output of the lesion identifier 44 is a set of lesion volumes 46a to 46c, which can be subtracted (on a spatial basis) from the mask 40 to produce modified masks 40'a to 40'c that do not include lesion volumes.

[0046] The physiological scan 18 and the improved mask 40' are then passed to toxicity evaluators 50a to 50c that are associated with organ-specific rules 52a to 52c for the specific organ of the mask 40'. Generally, the toxicity evaluator 50 analyzes the data of the physiological scan 18 restricted to the region of the enhanced mask 40' to provide greater sensitivity and specificity. In one embodiment, the organ-specific rules 52 may extract a histogram of standardized uptake values ​​(SUVs) from the physiological scan 18 within the mask 40' and store these values ​​in a history record 54 that records these values ​​during consecutive scans for a particular patient 15. Trends in these historical SUV values ​​from the record 54 may then be correlated with organ health. For example, the inventors have determined that for intestinal organs, a defined uptake agent may be more likely to be toxic to the organ in patients who later experience colitis than in patients who do not experience colitis. 18 The SUV 95 at the 95th percentile of the intestinal SUV histogram of F-FDG PET / CT has a significantly higher increase from the baseline scan before the patient is treated. In this case, an organ-specific rule 52 for the intestine can specify that an increase in SUV 95 from baseline of more than 40% indicates that the patient is in the pre-colitic stage (meaning that colitis is likely to occur), thereby allowing adjustment of treatment or taking other measures. The inventor's studies indicate that the above-mentioned signs of colitis can be seen a median of 115 days before the clinical diagnosis of colitis, providing a useful prediction of loss of organ health. This measurement has a sensitivity of 75% and a specificity of 88%. Additional rules 52 can be prepared and validated for other organs through similar empirical studies.

[0047] Each toxicity evaluator 50 using the organ-specific rules 52 can then output an organ health value 56 to a synthesizer 58. Example organ health values ​​56 can provide relevant changes in values ​​such as the SUV 95 discussed above and interpretive values ​​such as a threshold value (40%) for determining pre-colitis. These values ​​are then used by the synthesizer 58, as will be discussed in more detail below. More generally, the synthesizer 58 combines information from each of the toxicity calculators 50 and the organ-specific rules 52. The synthesizer 58 also receives the original CT structural image scan 14, as will be discussed below.

[0048] Still refer to Figure 2 and Figure 3The structural image scan 14 may also be provided to a lymph node neighborhood segmenter 60 that operates in a manner similar to the segmenter 38 but identifies regions of lymph node clusters in the patient. In one embodiment, the segmenter 60 may also utilize a convolutional neural network as described above that is trained using a set of manually segmented structural scans to identify neighborhoods of lymph nodes, such as around the neck and base of the neck, thereby generating a set of weights 62 for the neural network. Then, based on Figure 3 At process block 65 , the segmenter 60 , operating with the weights 62 , produces a neighborhood mask 64 that defines the lymph node cluster region.

[0049] This information of the neighborhood mask 64 is sent to a lymph node evaluator 66 which uses the physiological scan 18 to determine the lymph node Figure 3 Process block 67 identifies the lymph nodes (e.g., by a thresholding process or as discussed in the patents referenced above) and determines whether they have become enlarged, e.g., indicating possible stress on the lymph node system, e.g., due to detectable or undetectable organ toxicity in treatment involving immunotherapy.

[0050] according to Figure 3 At process block 69, the lymph node assessor 66 may generate an indication 68 of lymph node enlargement, which, as an alternative to lymph node stress, is also provided to the synthesizer 58. For example, the indication 68 may be a percentage increase in lymph node size as indicated by drug uptake since the baseline scan.

[0051] Additionally, the structural image scans 14 and physiological scans 18 may also be used by disease lesion tracking circuitry 73, which may provide longitudinal monitoring of disease state and, therefore, an indication of primary treatment efficacy. This tracking may implement the methods described in the above-referenced application 2016 / 0100795 and provide treatment information to the synthesizer 58.

[0052] Now refer to Figure 2 、 Figure 3 and Figure 4, the synthesizer 58 can provide an output display 70 according to process block 71, providing a comprehensive overview of organ health, which is difficult or impossible for individuals who rely on analyzing many different organs using different analysis criteria dictated by organ-specific rules 52 and collected from various empirical studies. In one embodiment, the display 70 can provide a scan image 72, for example, derived from a CT structural image scan 14 and depicting each organ 78 being analyzed highlighted using previously developed masks 40' and 64 (with diseased areas 75 removed). Each of these organ images can include annotations 74 in the form of text and / or charts identifying the organ and the measurements 79 that were taken, and a chart 76 can be provided showing trends in the measurements 79 of organ health, as well as thresholds 77 for health values ​​56 according to the organ-specific rules 52, indicating empirically derived thresholds at which side effects may prove to limit treatment. Thus, for example, for the intestine, changes in SUV 95 values ​​can provide the measurement 79, and a 40% threshold can be displayed as the threshold 77.

[0053] Each of these different health values ​​56 can be combined and weighted or normalized to provide a single summary indication 80 of the patient's organ health, intended to alert a physician of a potentially deteriorating condition. In the event of such an alert, the physician will carefully review the underlying data.

[0054] In addition to monitoring the health of a patient's organs, the present invention can also provide an assessment of disease changes through a graph 82 and a scanned image 84 that indicates the location of the disease using data from the disease lesion tracking circuit 73 as discussed above and with respect to the referenced application 2016 / 0100795. In this version, the present invention provides a more comprehensive view of the treatment and its effects and side effects.

[0055] It should be understood that the present invention can provide a user interface that provides the user with the ability to select and change the system parameters explicitly described above or implicitly required. These can include parameters required to select an organ or multiple organs, as well as parameters to add filters to control the amount and type of information provided.

[0056] The functionality of the present invention can be accessed or used remotely, and the data collected by the imaging system can be transmitted via a communications network (e.g., secure, HIPAA-compliant) to a remote computer system, which analyzes the data and the results can be sent back (or elsewhere) for use.

[0057] In this regard, the present invention may include a database component that collects and stores data from one or more patients and may use this information to modify the effectiveness thresholds employed by the present invention and / or provide predictions or suggestions regarding historical results that may be relevant to a particular query subject.

[0058] As used herein, the term organ generally refers to any group of tissue adapted to perform a specific function, including but not limited to liver, lungs, lymph nodes, intestines, and the like.

[0059] Certain terms are used herein for reference purposes only and, therefore, are not intended to be limiting. For example, terms such as "upper," "lower," "above," and "below" refer to the direction of the drawings to which reference is made. Terms such as "front," "rear," "back," "bottom," and "side" describe the orientation of portions of a component within a consistent but arbitrary reference frame, which becomes clear by reference to the text and associated drawings describing the component in question. These terms may include the words specifically mentioned above, derivatives thereof, and words of similar meaning. Similarly, the terms "first," "second," and other such numerical terms referring to structures do not imply a sequence or order unless the context clearly indicates otherwise.

[0060] When introducing the elements or features of the present disclosure and the exemplary embodiments, the articles "a", "an", "the" and "said" are intended to mean that there are one or more such elements or features. The terms "comprise", "comprising" and "having" are intended to be inclusive and mean that there may be other elements or features in addition to the elements or features specifically pointed out. It should also be understood that, unless specifically identified as an execution order, the method steps, processes and operations described herein should not be interpreted as necessarily requiring them to be performed in the specific order discussed or shown. It should also be understood that additional or alternative steps may be adopted.

[0061] References to "microprocessor" and "processor" or "the microprocessor" and "the processor" may be understood to include one or more microprocessors that may communicate in one or more independent and / or distributed environments, and thus may be configured to communicate with other processors via wired or wireless communications, wherein such one or more processors may be configured to operate on one or more processor-controlled devices, which may be similar or different devices. Furthermore, unless otherwise specified, references to memory may include one or more processor-readable and accessible memory elements and / or components, which may be internal to a processor-controlled device, external to a processor-controlled device, and accessible via a wired or wireless network.

[0062] In particular, the present invention is not intended to be limited to the embodiments and descriptions contained herein, and the claims should be understood to include modifications of these embodiments within the scope of the appended claims, including portions of the embodiments and combinations of elements of different embodiments. All publications described herein, including patent and non-patent publications, are hereby incorporated by reference in their entirety.

[0063] To assist the Patent Office and any reader of any patent issuing based on this application in interpreting the appended claims, Applicant wishes to point out that unless the words "means for" or "step for" are expressly used in a particular claim, Applicant does not intend that any appended claim or claim element invoke 35 U.S.C. §112(f).

Claims

1. An apparatus for evaluating the organ health of at least one non-involved patient organ during treatment of a patient having a diseased cancerous lesion in another organ, comprising: An electronic computer executing a stored program to: (a) receiving a structural image of at least one patient organ; (b) receiving a physiological image of the at least one organ indicative of organ function; (c) processing the structural image according to organ function information to create a mask corresponding to the at least one organ; (d) selecting a portion of the physiological image associated with the at least one organ using the mask; (e) applying toxicity rules specific to the component to provide an assessment of organ health; as well as (f) outputting an organ health indication of the at least one organ based on the evaluating.

2. The device according to claim 1, wherein The electronic computer executing the stored program also: A portion of the physiological image associated with the at least one organ is processed to identify disease lesions in the at least one organ, and the mask is refined to remove the lesions before the toxicity rule is applied.

3. The device according to claim 2, wherein Processing the structural image to create a mask labels the mask to an organ type, and wherein identification of lesions in the at least one organ is based on lesion identification rules associated with the organ type.

4. The device according to claim 1, wherein The electronic computer executing the stored program further processes the structural image to identify a lymph node region mask; and wherein the lymph node region mask is used to select a portion of the physiological image that is related to the lymph nodes; and wherein a lymph node activity rule is applied to a portion of the physiological image related to the lymph nodes to assess the activity of the lymph nodes; and Wherein the output further provides a lymph node activity indication indicating activation of the lymph node system.

5. The device according to claim 1, wherein The stored program provides a set of toxicity rules associated with different organs, and wherein processing the structural image creates a set of masks for different organs, each mask being associated with a specific organ type; and wherein different masks are used to select different portions of the physiological image associated with the different organs, and the different portions are associated with specific organ types according to the masks used; and wherein different toxicity rules are applied to different parts according to the organs associated with the toxicity rules to provide an assessment of organ health of a plurality of different organs; and Wherein the output indicates organ health of a plurality of organs.

6. The device according to claim 1, wherein The output provides a composite measure of organ health for the plurality of organs.

7. The device according to claim 1, wherein The output provides an image based on the structural image and augmented with organ health data.

8. The device according to claim 1, wherein The stored program retains previous output indicative of organ health associated with previous structural images and previous physiological images to provide a display of organ health trends for the plurality of organs.

9. The device according to claim 1, wherein The toxicity rules provide thresholds to predict organ health for a plurality of different organs; and Wherein the output indicates predicted organ health of the plurality of organs.

10. The device according to claim 1, wherein The electronic computer also analyzes the physiological images to assess changes in disease; and Wherein the output provides an indication of a change in a cancerous lesion of the disease.

11. The device according to claim 1, wherein The physiological image is a PET scan or a functional CT or a functional MRI / MRS scan.

12. The device according to claim 1, wherein The structural images are selected from the group consisting of CT and MRI scans.

13. The device according to claim 1, wherein The electronic computer processes the structural image using machine learning trained on different organ types to create a mask describing at least one organ.

14. The device according to claim 1, wherein The at least one organ is an intestine, and the physiological image indicates 18 F-FDG uptake.

15. The apparatus of claim 1 further comprising means for obtaining structural and physiological images in communication with said electronic computer.

16. A method for assessing organ health comprising analyzing images of a patient's information using the apparatus according to any one of claims 1 to 15.

17. The method of claim 16, further comprising the step of altering the patient's treatment regimen based on the analysis.

Images