Systems, devices and methods for non-invasive image-based plaque analysis and risk assessment

Abstract

Various embodiments described herein relate to systems, devices, and methods for non-invasive image-based plaque analysis and risk assessment. In particular, in some embodiments, the systems, devices, and methods described herein relate to the analysis of one or more regions of plaque, such as coronary artery plaque, using non-invasively acquired images that can be analyzed using computer vision or machine learning to identify, diagnose, characterize, treat, and / or track coronary artery disease.

Description

[Technical Field]

[0001] Priority and Related Applications This application is a continuation-in-part of U.S. Application No. 18 / 179,921, filed March 7, 2023, and is also a continuation-in-part of U.S. Provisional Application No. 63 / 385,179, filed November 28, 2022, U.S. Provisional Application No. 63 / 476,255, filed December 20, 2022, U.S. Provisional Application No. 63 / 477,640, filed December 29, 2022, U.S. Provisional Application No. 63 / 477,638 filed December 29, 2022, U.S. Provisional Application No. 63 / 477,985 filed December 30, 2022, U.S. Provisional Application No. 63 / 477,961 filed December 30, 2022, U.S. Provisional Application No. 63 / 478,076 filed December 30, 2022, U.S. Provisional Application No. 63 / 478,084, U.S. Provisional Application No. 63 / 383,632 filed November 14, 2022, U.S. Provisional Application No. 63 / 383,904 filed November 15, 2022, U.S. Provisional Application No. 63 / 386,297 filed December 6, 2022, U.S. Provisional Application No. 63 / 476,251 filed December 20, 2022 This application claims the benefit of priority to U.S. Provisional Application No. 63 / 476,245, filed December 20, 2022, U.S. Provisional Application No. 63 / 477,656, filed December 29, 2022, U.S. Provisional Application No. 63 / 385,472, filed November 30, 2022, and U.S. Provisional Application No. 63 / 386,376, filed December 7, 2022.

[0002] This application is also related to U.S. Patent No. 10,813,612, filed January 23, 2020, U.S. Patent No. 11,501,436, filed January 5, 2021, and U.S. Patent No. 11,302,001, filed August 4, 2021, as well as U.S. Application No. 17 / 820,439, filed August 17, 2022, and U.S. Application No. 18 / 179,921, filed March 7, 2023, each of which is incorporated herein by reference in its entirety.

[0003] This application relates to non-invasive image-based plaque analysis and risk assessment. Summary of the Invention

[0004] Various embodiments described herein relate to systems, devices, and methods for non-invasive image-based plaque analysis and risk determination. In particular, in some embodiments, the systems, devices, and methods described herein relate to the analysis of one or more regions of plaque, such as coronary artery plaque, using non-invasively acquired images that can be analyzed using computer vision or machine learning to identify, diagnose, characterize, treat, and / or track coronary artery disease.

[0005] A better understanding of the apparatus and methods described herein may be obtained by reference to the following description taken in conjunction with the accompanying drawings. [Brief explanation of the drawings]

[0006] [Figure 1] 1 shows a schematic diagram of an example embodiment of a system 100 including a processing system configured to characterize coronary artery plaque. [Figure 2] 1 is a schematic diagram showing an example of the myocardium and its coronary arteries. [Figure 3] FIG. 1 shows an example series of images generated from a scan along a coronary artery, including selected images of a portion of the coronary artery, and how the image data correspond to values ​​on the Hounsfield scale. [Figure 4A] FIG. 1 is a block diagram illustrating a computer system in which various embodiments may be implemented. [Figure 4B] 4 is a block diagram illustrating computer modules of a computer system 400 in which various embodiments may be implemented. [Figure 5A] 1 shows an example of a flowchart of a process for analyzing coronary artery plaque. [Figure 5B]5B shows an example of a flowchart that expands a portion of the flowchart of FIG. 5A for determining characteristics of coronary artery plaque. [Figure 6] 1 is a representation of image data showing an example of a portion of a coronary artery (sometimes referred to herein as a "vessel" for ease of reference). [Figure 7] Illustrates the same blood vessel and plaque and fat characteristics as illustrated in FIG. 6, and further illustrates additional examples of areas of arteries and plaque and / or perivascular fat in the vicinity of arteries that may be analyzed to determine characteristics of a patient's arteries. [Figure 8A] FIG. 1 is a block diagram illustrating an example of a process by which features in a medical image are identified using artificial intelligence or machine learning. [Figure 8B] FIG. 1 is a schematic diagram illustrating an example of a neural network for determining patient characteristics based on medical images. [Figure 8C] 1 illustrates a flowchart for training an artificial intelligence or machine learning model, according to some embodiments. [Figure 8D] 1 illustrates an example of training and using an AI / ML model according to some embodiments. [Figure 9] FIG. 1 is a block diagram illustrating an embodiment of a computer hardware system configured to execute software to implement one or more embodiments of the systems, devices, and methods described herein. [Figure 10] 1 is a flow chart illustrating exemplary embodiment(s) of a system, device, and method for non-invasive, image-based ischemic risk assessment. [Figure 11] 1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for determining ischemia based on image-based analysis of a stenosis. [Figure 12] 1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for multivariate image-based analysis of ischemia. [Figure 13]1 is a flowchart illustrating exemplary embodiment(s) of a system, apparatus, and method for image-based analysis to determine cardiac catheterization therapy. [Figure 14] 1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for non-invasive, image-based plaque analysis and risk determination, and patient-specific atherosclerosis treatment based on computational modeling, wherein a second computational model is calculated by converting low-density non-calcified plaque to non-calcified or calcified plaque and non-calcified plaque to calcified plaque. [Figure 15] 1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for transforming a medical image based on a plaque parameter and / or a vascular parameter. [Figure 16] 1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for image-based analysis and tracking of plaque progression. [Figure 17] 1 is a flowchart illustrating exemplary embodiment(s) of a system, apparatus, and method for user training of image-based analysis and / or plaque identification. [Figure 18] 1 is a flowchart illustrating exemplary embodiment(s) of a system, apparatus, and method for automated medical image segmentation and / or analysis for admission. [Figure 19] 1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for image-based myocardial infarction type analysis and risk assessment. [Figure 20] 1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for image-based post-operative myocardial infarction analysis and risk assessment. [Figure 21A] 1 is an exemplary graph of blood flow during different levels of physical exercise in patients with and without coronary artery disease. [Figure 21B]1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for determining prescribed flow reserve for treating cardiovascular disease. [Figure 22] 1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for three-dimensional topological mapping of plaque. [Figure 23] 1 is a flowchart illustrating exemplary embodiment(s) of a system, apparatus, and method for automated retrieval and / or curation of data based on image-derived variables. [Figure 24] 1 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for non-invasive image-based plaque analysis and risk assessment. [Figure 25] 10 is a flowchart illustrating additional exemplary embodiment(s) of a system, device, and method for non-invasive image-based plaque analysis and risk assessment. [Figure 26] 1 is a flow chart illustrating exemplary embodiment(s) of a system, device, and method for multivariate image-based analysis of thin-cap fibroatheroma (TCFA). DETAILED DESCRIPTION OF THE INVENTION

[0007] Although several embodiments, examples, and illustrations are disclosed below, those skilled in the art will understand that the invention described herein extends beyond the specifically disclosed embodiments, examples, and illustrations and includes other uses of the invention, as well as obvious modifications and equivalents thereof. Embodiments of the invention are described with reference to the accompanying figures, in which like numerals refer to like elements throughout. The terminology used in the description presented herein, even when used in conjunction with detailed descriptions of certain specific embodiments of the invention, is not intended to be construed as limiting or restrictive. In addition, embodiments of the invention may be comprised of multiple novel features, and no single feature is solely responsible for its desirable attributes or essential to practicing the invention described herein.

[0008] Disclosed herein are systems, devices, and methods for noninvasive, image-based plaque analysis and risk assessment. In particular, in some embodiments, the systems, devices, and methods described herein relate to the analysis of one or more regions of plaque, such as coronary artery plaque, based on one or more distance, volume, shape, morphology, embedment, and / or axial (or dimensional) measurements. A "plaque" or "area of ​​plaque" or "one or more regions of plaque" may be referred to simply as "plaque" for ease of reference, unless otherwise indicated explicitly or by context. For example, in some embodiments, the systems, devices, and methods described herein relate to plaque analysis based on one or more of the following: the distance between the plaque and the vessel wall, the distance between the plaque and the lumen wall, the length of the plaque along the anterior-posterior axis, the length of the plaque along the lateral axis, the volume of low-density non-calcified plaque, the volume of total plaque, the ratio(s) between the volume of low-density non-calcified plaque and the volume of total plaque, the embedment of low-density non-calcified plaque, and / or the like. In some embodiments, the systems, devices, and methods described herein are configured to determine a risk of coronary artery disease (CAD), e.g., myocardial infarction (MI), based on one or more plaque analyses described herein. In some embodiments, the systems, devices, and methods described herein are configured to generate suggested treatments and / or visual representations based on the determined CAD risk and / or one or more plaque analyses described herein.

[0009] Also disclosed herein are systems, methods, and devices for cardiovascular risk and / or condition assessment using image-based analysis. In particular, in some embodiments, the systems, devices, and methods are the basis for assessing cardiovascular risk and / or disease condition using image-based analysis of vascular surface and / or feature coordinates. In some embodiments, cardiovascular risk and / or disease condition assessments generated using the systems, methods, and devices herein can be utilized to generate patient diagnoses and / or treatment recommendations.

[0010] Also disclosed herein are systems, methods, and devices for cardiovascular risk and / or condition assessment using image-based analysis, and in some embodiments, the systems, devices, and methods relate to cardiovascular risk and / or disease condition assessment using image-based analysis of vascular surface and / or feature coordinates. In some embodiments, cardiovascular risk and / or disease condition assessments generated using the systems, methods, and devices herein may be utilized in patient diagnosis and / or generation of suggested treatments.

[0011] Also disclosed herein are systems, methods, and devices for cardiovascular risk and / or condition assessment using image-based analysis, and in some embodiments, the systems, devices, and methods relate to cardiovascular risk and / or disease and / or condition assessment using corrected and / or normalized image analysis-based plaque parameters. In some embodiments, cardiovascular risk and / or disease and / or condition assessments generated using the systems, methods, and devices herein can be utilized in patient diagnosis and / or generation of suggested treatments.

[0012] Also disclosed herein are systems, methods, and devices for generating patient-specific reports regarding cardiovascular disease risk and / or status assessment, diagnosis, and / or treatment, including, for example, coronary artery disease (CAD). In particular, in some embodiments, the systems, devices, and methods are configured to generate immersive, patient-specific reports regarding a patient's cardiovascular disease risk, status, diagnosis, and / or treatment. In some embodiments, the systems, devices, and methods are configured to generate immersive, patient-specific reports based, at least in part, on image-based analysis of, for example, one or more plaque and / or vascular parameters. In some embodiments, the systems, devices, and methods are configured to view a patient's cardiovascular disease status or risk from a perspective within one or more of the patient's arteries. In some embodiments, the systems, devices, and methods are configured to visually display and / or track the actual or hypothetical progression of a patient's cardiovascular disease status or risk based on actual or proposed treatments from a perspective within one or more of the patient's arteries.

[0013] Also disclosed herein are systems, methods, and devices for cardiovascular risk and / or condition assessment using image-based analysis, and in some embodiments, the systems, devices, and methods relate to cardiovascular risk and / or disease and / or condition assessment using normalized image analysis-based plaque parameters. In some embodiments, cardiovascular risk and / or disease and / or condition assessments generated using the systems, methods, and devices herein can be utilized in generating patient diagnoses and / or treatment recommendations.

[0014] Also disclosed herein are systems, devices, and methods for non-invasive, image-based determination of fractional flow reserve (FFR) and / or ischemia. In particular, in some embodiments, the systems, devices, and methods relate to FFR and / or ischemia analysis of arteries, such as coronary arteries, aorta, and / or carotid arteries, using one or more image analysis techniques. For example, in some embodiments, the systems, methods, and devices can be configured to derive one or more stenosis and / or normal measurements from non-invasively obtainable medical images and use the same to derive an FFR and / or ischemia assessment. In some embodiments, the systems, methods, and devices can be configured to apply one or more allometric scaling laws to one or more stenosis and / or normal measurements to derive and / or generate an FFR and / or ischemia assessment.

[0015] Coronary heart disease afflicts more than 17.6 million Americans. Current trends in the treatment of cardiovascular health problems are generally twofold. First, physicians typically consider a patient's cardiovascular health from a macro level. For example, they analyze a patient's biochemistry or blood constituents or biomarkers to determine whether the patient has high levels of cholesterol components in their bloodstream. In response to high levels of cholesterol, some physicians prescribe one or more medications, such as statins, as part of a treatment plan to reduce the perceived high levels of cholesterol components in the patient's bloodstream.

[0016] Currently, in a second popular trend for treating cardiovascular health issues, physicians use angiograms to evaluate a patient's cardiovascular health and identify large blockages in the patient's various arteries. Upon finding large blockages in various arteries, physicians may perform angioplasty, guiding a balloon catheter to the narrowing point of the blood vessel. Once properly positioned, the balloon is inflated to compress or flatten plaque or fatty material against the vessel wall and / or stretch the artery, increasing blood flow through the vessel and / or to the heart. In some cases, the balloon is used to place and expand a stent within the vessel, compressing plaque and / or keeping the vessel open, allowing more blood to flow. Approximately 500,000 cardiac stent procedures are performed annually in the United States.

[0017] However, a recent $100 million federally funded study questions whether current trends in cardiovascular disease treatment are the most effective treatments for all types of patients. The recent study, which involved more than 5,000 patients with moderate to severe stable heart disease from 320 centers in 37 countries, provided new evidence that stents and bypass surgery are likely less effective than medications combined with lifestyle modifications for patients with stable heart disease. Therefore, patients with stable heart disease may be better served by forgoing invasive surgical procedures like angioplasty and bypass surgery and instead prescribing heart disease medications like statins and engaging in lifestyle modifications like regular exercise. This new treatment could impact thousands of patients worldwide. It is estimated that 500,000 cardiac stent procedures are performed annually in the United States, with one-fifth of these procedures being performed on patients with stable heart disease. Furthermore, it is estimated that 25% of the estimated 100,000 patients with stable heart disease, or approximately 23,000 people, do not experience chest pain. Thus, more than 20,000 patients per year could potentially avoid invasive surgery and its resulting complications.

[0018] A more complete understanding of a patient's cardiovascular disease can be important to determine whether a patient should forgo invasive surgery and instead opt for medical therapy and / or to develop a more effective treatment plan. Specifically, a better understanding of a patient's arterial vascular health is beneficial. For example, it is useful to understand whether a patient's plaque buildup is mostly fatty deposits or mostly calcified. In the former situation, treatment with cardiac medications such as statins may be warranted, whereas in the latter situation, the patient should undergo further regular monitoring without prescribing cardiac medications or implanting a stent. However, if the plaque buildup is significant enough to cause severe narrowing or stenosis of the ductus arteriosus, blocking blood flow to the heart muscle, invasive angioplasty with stent placement may be necessary. Sudden cardiac death is one of the leading causes of natural death in the United States, accounting for approximately 325,000 adult deaths per year and nearly half of all deaths due to cardiovascular disease. SCD is twice as prevalent in men as in women. SCD generally develops between the mid-30s and mid-40s. In over 50% of cases, sudden cardiac arrest occurs without warning.

[0019] For millions of patients suffering from heart disease, there is a need to not only know the blood chemistry or content of the blood flowing through such arteries, but also to better understand the overall health of the arteries within a patient. For example, in some embodiments of the systems, devices, and methods disclosed herein, arteries with "good" or stable plaque, or plaque consisting of hardened calcified content, are considered non-life-threatening to the patient, whereas arteries containing "bad" or unstable plaque, or plaque consisting of fatty material, are considered more life-threatening because such bad plaque may rupture within the artery, thereby releasing such fatty material into the artery. The release of such fatty material into the bloodstream can cause inflammation and result in a blood clot. A blood clot within the artery can prevent blood from being transported to the heart muscle, potentially leading to a heart attack or other cardiac events. Furthermore, in some cases, it is generally more difficult for blood to flow through a fatty plaque buildup than through a calcified plaque buildup. Therefore, there is a need to better understand and analyze a patient's arterial vessel walls.

[0020] Furthermore, while blood tests and drug treatment regimens are useful for mitigating cardiovascular health problems and mitigating cardiovascular events (e.g., heart attacks), such treatment methods are not perfect or flawless in that they may misidentify and / or fail to pinpoint areas of significant cardiovascular risk. For example, a simple analysis of a patient's blood chemistry likely will not identify that the patient has arteries with significant buildup of bad plaque (fatty deposits) along the vessel walls. Similarly, angiograms can help identify stenoses or narrowings, but may not clearly identify areas with significant buildup of bad plaque in the vessel walls. Such areas of bad plaque buildup within arterial vessel walls can be an indicator of a patient at high risk for cardiovascular events, such as a heart attack. In certain circumstances, areas with bad plaque may rupture, releasing fatty material into the arterial bloodstream, potentially resulting in a blood clot in the artery. A blood clot in the artery can block blood flow to heart tissue and potentially cause a heart attack. Therefore, there is a need for new techniques that can analyze arterial vessel walls and / or identify areas within arterial vessel walls that constitute plaque buildup, whether malignant or not.

[0021] In some embodiments, the systems, devices, and methods described herein are configured to utilize non-invasive medical imaging techniques, such as CT images or CCTA, which may be input into a computer system configured to automatically and / or dynamically analyze the medical images to identify one or more coronary arteries and / or plaques therein. For example, in some embodiments, the system may be configured to automatically and / or dynamically analyze the medical images using one or more machine learning and / or artificial intelligence algorithms to identify, quantify, and / or classify one or more coronary arteries and / or plaques. In some embodiments, the system may be further configured to utilize the identified, quantified, and / or classified one or more coronary arteries and / or plaques to generate a treatment plan, track disease progression, and / or generate a patient-specific medical report, e.g., using one or more artificial intelligence and / or machine learning algorithms. In some embodiments, the system may be further configured to dynamically and / or automatically generate a visualization of the identified, quantified, and / or classified one or more coronary arteries and / or plaques, e.g., in the form of a graphical user interface. Additionally, in some embodiments, to calibrate medical images obtained from different medical image scanners and / or different scanning parameters or environments, the system can be configured to utilize a normalization device consisting of one or more sections of one or more materials.

[0022] As described in further detail, the systems, devices, and methods described herein enable automated and / or dynamic quantitative analysis of various parameters related to plaque, cardiovascular arteries, and / or other structures. More specifically, in some embodiments described herein, medical images of a patient, such as coronary CT images or CCTA, can be taken at a medical facility. Rather than a physician visually inspecting or generally assessing the patient, the medical images are transmitted to a back-end main server, which in some embodiments is configured to perform one or more analyses of the medical images in a reproducible manner. Thus, in some embodiments, the systems, methods, and devices described herein can provide quantified measurements of one or more features of the coronary CT images using an automated and / or dynamic process. For example, in some embodiments, the main server system can be configured to identify one or more blood vessels, plaque, fat, and / or one or more measurements thereof from the medical images. Based on the identified features, in some embodiments, the system can be configured to generate one or more quantified measurements from the raw medical images, such as the radiodensity of one or more regions of plaque, identification of stable and / or unstable plaque, its volume, its surface area, geometric shape, its heterogeneity, and / or the like. In some embodiments, the system may also generate one or more quantified measurements of blood vessels from the raw medical images, such as, for example, diameter, volume, morphology, and / or the like. Based on the identified features and / or quantified measurements, in some embodiments, the system may be configured to generate a risk and / or disease state assessment and / or track the progression of a plaque-based disease or condition, such as, for example, atherosclerosis, stenosis, and / or ischemia, using the raw medical images. Additionally, in some embodiments, the system may be configured to generate a GUI visualization of the identified one or more features and / or quantified measurements, such as a quantized color mapping of different features.In some embodiments, the systems, devices, and methods described herein are configured to utilize medical image-based processing to assess a subject's risk of a cardiovascular event, a major adverse cardiovascular event (MACE), rapid plaque progression, and / or non-response to drug therapy. In particular, in some embodiments, the systems may be configured to automatically and / or dynamically assess such health risks of a subject by analyzing only non-invasively obtained medical images. In some embodiments, one or more of the processes may be automated using artificial intelligence (AI) and / or machine learning (ML) algorithms. In some embodiments, one or more of the processes described herein may be performed within minutes in a reproducible manner. This is in contrast to existing means today that do not produce reproducible prognoses or assessments, are time-consuming, and / or require invasive procedures. In some embodiments, the systems, methods, and devices described herein are configured and / or configured to utilize any one or more of such techniques described in U.S. Patent Application Publication No. US2021 / 0319558, the entire contents of which are incorporated herein by reference.

[0023] Thus, in some embodiments, the systems, devices, and methods described herein can provide physicians and / or patients with specific quantification and / or measurement data related to a patient's plaque and / or ischemia that currently does not exist. In some embodiments, this level of detail in quantified plaque parameters from image processing and downstream analysis can provide more accurate and useful tools for assessing patient health and / or risk in entirely novel ways.

[0024] A method for identifying high-risk plaques is disclosed that utilizes volumetric characterization of coronary plaques and perivascular adipose tissue data from computed tomography (CT) scans. Volumetric characterization of coronary plaques and perivascular adipose tissue allows for the determination of the inflammatory state of the plaques from CT scans, which is useful for the diagnosis, prognosis, and treatment of coronary artery disease. While certain exemplary embodiments are illustratively shown in the drawings and described in detail herein, these embodiments are susceptible to various modifications and alternative forms. It is not intended that the exemplary embodiments be limited to the particular forms disclosed; on the contrary, the exemplary embodiments are intended to cover all modifications, equivalents, and alternatives falling within the scope of the exemplary embodiments.

[0025] Although terms such as "first," "second," etc. may be used herein to describe various elements, it should be understood that these elements are not limited by these terms. These terms are used merely to distinguish one element from another. For example, a first element could be termed a second element, and similarly, a second element could be termed a first element, without departing from the scope of the exemplary embodiments. As used herein, the term "and / or" includes all combinations of one or more of the associated listed items.

[0026] The terms used herein are for the purpose of describing embodiments only and are not intended to limit example embodiments. As used herein, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly dictates otherwise. It will be further understood that as used herein, the terms "comprises," "comprising," "includes," and / or "including" specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. As used herein, the term "and / or" refers not only to the individual items but also to all combinations thereof.

[0027] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the exemplary embodiments belong. Furthermore, it will be understood that terms as defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense unless expressly so defined herein. It should also be noted that in some alternative implementations, the functions / acts described may occur out of the order depicted in the figures. For example, two figures shown in succession may actually be performed simultaneously or may be performed in the reverse order depending on the functions / acts involved.

[0028] In the drawings, the dimensions of layers and regions may be exaggerated for clarity of illustration. Also, when a layer (or structure) is referred to as "on" another layer or structure, it will be understood that it can be directly on the other layer or substrate, or that intervening layers may be present. Furthermore, when a layer is referred to as "underlying" another layer, it will be understood that it can be directly under, or that one or more intervening layers may be present. Additionally, when a layer is referred to as being "between" two layers, it will be understood that it can be the only layer between the two layers, or that one or more intervening layers may also be present. Like reference numerals refer to like elements throughout.

[0029] Overview of an Example Processing System for Assessing Coronary Artery Plaque The present disclosure includes methods and systems that use data generated from images collected by scanning a patient's arteries to identify coronary plaques that are at high risk of causing a future heart attack or acute coronary syndrome. In particular, the relationship between perivascular coronary fat, coronary plaque, and / or coronary lumen characteristics, as well as the relationship between perivascular coronary fat, coronary plaque, and / or coronary lumen characteristics, is discussed to determine how to identify coronary plaques that are likely to contribute to future ACS, heart attack, and death. Images used to generate image data may be CT images, CCTA images, or images generated using any applicable technology capable of depicting the relative densities of coronary plaque, perivascular fat, and the coronary lumen. For example, CCTA images can be used to generate two-dimensional (2D) or volumetric (three-dimensional (3D)) image data, which can be analyzed to determine specific characteristics related to the radiodensity of coronary plaque, perivascular fat, and / or the coronary lumen. In some embodiments, the Hounsfield scale is used to provide a measure of the radiodensity of these features. Hounsfield units, as is well known, represent arbitrary units of X-ray attenuation used in CT scans. Each pixel (2D) or voxel (3D) of a feature in the image data can be assigned a radiodensity value on the Hounsfield scale, and these values ​​can be analyzed to characterize the feature.

[0030] In various embodiments, processing of the image information can include: (1) determining scan parameters (e.g., mA (milliamperes), kvP (peak kilovoltage)); (2) determining scan image quality (e.g., noise, signal-to-noise ratio, contrast-to-noise ratio); (3) measuring scan-specific coronary artery lumen density (e.g., from a distal point on the coronary artery wall to a proximal point on the coronary artery wall to a distal point on the coronary artery, and from a central location on the coronary artery to an outer location (e.g., outer relative to radial distance from the coronary artery); (4) measuring scan-specific plaque density (e.g., from center to outer, abruptness of the change from high to low or low to high within the plaque) as a function of three-dimensional shape; and (5) measuring scan-specific perivascular coronary fat density (close to far from the artery) as a function of three-dimensional shape.

[0031] From these measurements, which are independent of commonly known features of atherosclerosis causing ischemia, several features can be determined, including but not limited to: 1. The ratio of luminal attenuation to plaque attenuation, where a volumetric model of scan-specific attenuation density gradients within the lumen adjusts for reduced luminal density across plaque lesions that are more functionally significant in terms of risk value. 2. Among the subset of plaques considered to be "calcified plaques" based on the ratio of plaque attenuation to fat attenuation, highly radiodense plaques are considered to be of low risk. 3. Ratio of lumen attenuation / plaque attenuation / fat attenuation 4. The ratio of #1-3 as a function of the 3D shape of atherosclerosis, which can include 3D texture analysis of plaque 5. The three-dimensional volumetric shape and path of the lumen along with the attenuation density from the beginning to the end of the lumen. 6.Compare the plaques before and after a particular plaque and the plaque type to learn more about the risk. 7. Determine "high-risk plaque" by "subtracting" calcified (high-density) plaque to obtain a better absolute measure of high-risk plaque (low-density plaque). In other words, this particular embodiment involves identifying calcified plaque and excluding it from further analysis of plaque for the purpose of identifying high-risk plaque. Other characteristics can also be determined.

[0032] Analyzing these features / indicators, along with other indicators, together can assess the risk of plaque contributing to future heart attack, ACS, ischemia, or death. This can be done through traditional risk score development and / or validation or through machine learning methods. Factors for analysis from metrics likely associated with heart attack, ACS, ischemia, or death include: (1) a low [bright lumen:dark plaque] ratio, (2) a low [dark plaque:bright fat] ratio, (3) a low [bright lumen:dark plaque:bright fat] ratio, and (4) a low [dark lumen:dark myocardium in one vascular region] / [lumen:myocardium in another vascular region] ratio. Improvements in the disclosed method and system include: (1) using numerical values ​​from [lumen:plaque], [plaque:fat], and [lumen:plaque:fat] ratios instead of using qualitative definitions of atherosclerotic features; (2) using scan-specific [lumen:plaque attenuation] ratios to characterize plaque; (3) using scan-specific [lumen:plaque attenuation] ratios to characterize plaque; (4) using [lumen:plaque:fat circumference] ratios to characterize plaque; and (5) integrating pre- and post-plaque volume and type as contributors to individual plaque risk.

[0033] Atherosclerotic plaque characteristics can change over time with medical treatment (colchicine and statin drugs), and while some of these drugs may slow plaque progression, they also play a crucial role in promoting plaque changes. While statin drugs slow overall plaque progression, they may actually increase the progression of calcified plaque and decrease non-calcified plaque. This change is associated with reduced heart attacks, acute coronary syndromes, and death, and the disclosed methods can be used to monitor the effect of drug therapy on plaque risk over time. These methods can also be used to identify individuals whose atherosclerotic plaque characteristics or [lumen:plaque] / [plaque:fat] / [lumen:plaque:fat] ratios indicate a greater susceptibility to rapid disease progression or malignant transformation. Additionally, these methods can be applied to a single plaque, or tracking of atherosclerosis throughout the heart can be used to monitor a patient's risk of experiencing a heart attack (rather than attempting to identify specific plaques as the cause of future heart attacks). Tracking can be achieved by automated co-registration of patient-related imaging data over a period of time.

[0034] 1 shows a schematic diagram of an example embodiment of a system 100 including a processing system 120 configured to characterize coronary artery plaque. The processing system 120 includes one or more servers (or computers) 105, each configured with one or more processors. The processing system 120 includes a non-transitory computer memory component for storing data and a non-transitory computer memory component for storing instructions executed by a data communication interface of the one or more processors, the instructions configuring the one or more processors to perform a method for analyzing image information. A more detailed example of a server / computer 105 is described with reference to FIG. 9.

[0035] System 100 also includes a network. Processing system 120 is in communication with network 125. Network 125 may include, at least as part of network 125, the Internet, a wide area network (WAN), a wireless network, etc. In some embodiments, processing system 120 is part of a "cloud" implementation and may be located anywhere in communication with network 125. In some embodiments, processing system 120 is located in the same geographic proximity as an imaging facility that captures and stores patient image data. In other embodiments, processing system 120 is located remotely from where the patient image data is generated or stored.

[0036] FIG. 1 also illustrates various computer systems and devices 130 (e.g., at an imaging facility) in system 100 that are associated with generating patient image data and are also connected to network 125. One or more of devices 130 may be at an imaging facility that generates images of a patient's arteries, at a medical facility (e.g., a hospital, a doctor's office, etc.), or may be a personal computing device of a patient or caregiver. For example, as shown in FIG. 1, an imaging facility server (or computer) 130A may be connected to network 125. In this example, an imaging facility scanner 130B may also be connected to network 125. One or more other computing devices may also be connected to network 125. For example, laptop 130C, personal computer 130D, and / or image information storage system 130E may also be connected to network 125 and communicate with processing system 120 and each other via network 125.

[0037] In some examples, scanner 130B can be a computed tomography (CT) scanner, which uses a rotating X-ray tube and an array of detectors to measure the attenuation of X-rays by different tissues in the body and form corresponding images. In another example, scanner 130B can use a rotating tube ("spiral CT") in which the entire X-ray tube and detectors rotate around a central axis of the scanned area. In another example, scanner 130B can utilize electron beam tomography (EBT). In another example, scanner 130B can be a dual-source CT scanner having two X-ray tube systems. The methods and systems described herein can also use images from other CT scanners. In some examples, scanner 130B is a photon-counting CT scanner, a spectral CT scanner, or a dual-energy CT scanner. Photon-counting CT scanners, spectral CT scanners, or dual-energy CT scanners help provide more detailed, high-resolution images that better show small blood vessels, plaque, and other vascular pathologies, allowing for the determination of absolute material density rather than relative density. Generally, photon-counting CT scanners use X-ray detectors to count photons, quantify their energy, and determine the number of photons counted within several discrete energy bins, resulting in higher contrast-to-noise ratios, improved spatial resolution, and spectral imaging compared to conventional CT scanners. Each registered photon is assigned to a specific bin according to its energy, and each pixel measures a histogram of the incident X-ray spectrum. This spectral information has several advantages. First, it can be used to quantitatively determine the material composition of each pixel in the reconstructed CT image, as opposed to the estimated mean linear attenuation coefficient obtained in conventional CT scans. The spectral / energy information can be used to remove beam-hardening artifacts, which occur due to the higher linear attenuation of many materials, shifting the mean energy of the X-ray spectrum to higher energies. Additionally, the use of two or more energy bins allows for the identification of specific objects (e.g., bone, calcification, contrast agent, tissue).In some embodiments, images generated using a photon-counting CT scanner allow for the evaluation of plaque not only at different monochromatic energies but also at different polychromatic spectra (e.g., 100 kVp, 120 kVp, 140 kVp, etc.), which allows for different definitions of non-calcified and calcified plaque compared to conventional CT scanners. Spectral CT scanners perform CT scans using different X-ray wavelengths (or energies). Dual-energy CT scanners use separate X-ray energies to detect two different energy ranges. As an example, a dual-energy CT scanner (also known as spectral CT) can use an X-ray detector with separate layers to detect two different energy ranges ("dual layer"). In another example, a dual-energy CT scanner can use a single scanner to scan twice using two different energy levels (e.g., electronic kVp switching). The images detected at each different energy level can be combined to form an image, or the images can be used separately to evaluate a patient's medical condition. In addition to providing absolute material density, photon-counting CT scanners also allow for the evaluation of images that are "monochromatic," as opposed to typical CT, which uses a polychromatic spectrum of light. As discussed above, features (e.g., low-density noncalcified plaque, calcified plaque, noncalcified plaque) depicted in images generated using a photon-counting CT scanner, a spectral CT scanner, or a dual-energy CT scanner may have different radio densities than those depicted in images generated from a conventional CT scanner, i.e., such images may affect or alter the definition of calcified and noncalcified plaque. However, the radio densities or other features of calcified and noncalcified plaque depicted in images generated from a photon-counting CT scanner, a spectral CT scanner, or a dual-energy CT scanner can be normalized to correspond to the densities of a conventional CT scanner and the densities disclosed herein.Thus, the radiation densities disclosed herein can be directly correlated to the radiation densities of images produced by a photon-counting CT scanner, a spectral CT scanner, or a dual-energy CT scanner, such that the systems and methods, analyses, plaque densities, etc. disclosed herein are directly applicable to images produced from a photon-counting CT scanner, a spectral CT scanner, or a dual-energy CT scanner, normalized to the radiation density of an equivalent conventional CT scanner.

[0038] Information communicated from device 130 to processing system 120 via network 125 may include image information 135. In various embodiments, image information 135 may include 2D or 3D image data of the patient, scan information related to the image data, patient information, and other image or image-related information related to the patient. For example, image information may include patient information including one or more characteristics of the patient, such as age, sex, body mass index (BMI), medications, blood pressure, heart rate, height, weight, race, whether the patient is a smoker or non-smoker, body type (e.g., "body size" or "body type," which can be based on a wide range of factors), medical images, diabetes, hypertension, history of coronary artery disease (CAD), dietary habits, medication history, family history of disease, information related to other previously collected image information, exercise habits, drinking habits, lifestyle information, test results, etc. In some embodiments, image information includes patient identification information, such as the patient's name, the patient's address, driver's license number, social security number, or another indicator of the patient's identity. Once the processing system 120 analyzes the image information 135, patient-related information 140 may be communicated from the processing system 120 over the network 125 to the devices 130. The patient information 140 may include, for example, a patient report. The patient information 140 may also include various patient information available from a patient portal that may be accessed by one of the devices 130.

[0039] In some embodiments, image information, consisting of multiple images of a patient's coronary arteries and patient information / characteristics, may be provided from one or more devices 130 to one or more servers 105 of processing system 120 via network 125. Processing system 120 is configured to generate coronary artery information using the multiple images of the patient's coronary arteries to generate two-dimensional and / or three-dimensional data representations of the patient's coronary arteries. Processing system 120 then analyzes the data representations to generate a patient report documenting the patient's health status and risks associated with coronary plaque. The patient report may include images of the patient's arteries and visual depictions of types of coronary plaque within or near the coronary arteries. Machine learning or other artificial intelligence techniques may be used to compare the data representations of the patient's coronary arteries with data representations of other patients (e.g., stored in a database) to determine additional information regarding the patient's health. For example, the state of specific plaque in the patient's coronary arteries may determine the patient's likelihood of suffering a heart attack or other adverse coronary events. Additional information, for example, regarding the patient's CAD risk may also be determined.

[0040] FIG. 2 is a schematic diagram illustrating an example of myocardium 225 and its coronary arteries. The coronary arteries include a complex network of blood vessels ranging from large arteries to arterioles, capillaries, veins, and venules. FIG. 1 depicts a model 220 of a portion of the coronary artery system that circulates blood to and within the heart, including multiple coronary arteries, e.g., the left anterior descending (LAD) artery 215, the left circumflex (LCX) artery 220, and the aorta 240, which supplies blood to the right coronary artery (RCA) 230, described further below. The coronary arteries supply blood to the myocardium 225. Like all other tissues in the body, the myocardium 225 requires oxygen-rich blood to function. It must also transport away oxygen-depleted blood. The coronary arteries wrap around the outside of the myocardium 225. Smaller branches penetrate the myocardium 225 to transport blood. The example methods and systems described herein can be used to determine information related to blood flowing through the coronary arteries in any vessels extending therefrom. In particular, the example methods and systems described can be used to determine various information related to one or more portions of a coronary artery in which plaque has formed, which can be used to determine the risks associated with such plaque, such as whether plaque formation poses a risk of causing an adverse event to the patient.

[0041] The right side 230 of the heart 225 is depicted on the left side (relative to the page) of Figure 2, and the left side 235 of the heart is depicted on the right side of Figure 2. The coronary arteries include the right coronary artery (RCA) 205, which extends downward from the aorta 240 along the right side 230 of the heart 225, and the left coronary artery (LMCA) 210, which extends downward from the aorta 240 on the left side 235 of the heart 225. The RCA 205 supplies blood to the right ventricle, right atrium, and the SA (sinoatrial) node and AV (atrioventricular) node, which regulate the rhythm of the heart. The RCA 205 branches into smaller branches, such as the right posterior descending artery and the acute marginal artery. Together with the left anterior descending artery 215, the RCA 205 serves to supply blood to the middle section, or septum, of the heart.

[0042] The LMCA 210 branches into two arteries: the left anterior interventricular coronary artery (also known as the left anterior descending (LAD) artery 215) and the left circumflex coronary artery 220. The LAD artery 215 supplies blood to the anterior left side of the heart. Blockage of the LAD artery 215 is often referred to as a widow's infarction. The circumflex branch of the left coronary artery 220 surrounds the myocardium. The circumflex branch of the left coronary artery 220 courses first to the left and then to the right along the left portion of the coronary groove, nearly to the posterior longitudinal groove, supplying blood to the outside and back of the heart.

[0043] 3 shows an example series of images generated from a scan along a coronary artery, including selected images of a portion of the coronary artery, and how the image data may correspond to values ​​on the Hounsfield scale. In addition to obtaining image data, as described with reference to FIG. 1, scan information, including metrics related to the image data, and patient information, including patient characteristics, may also be collected.

[0044] A portion of a heart 225, the LMCA 210, and the LAD artery 215 are shown in the example of FIG. 3. A set of images 305 can be acquired along a portion of the LMCA 210 and the LAD artery 215, in this example from a first point 301 on the LMCA 210 to a second point 302 on the LAD artery 215. In some examples, the image data can be obtained using non-invasive imaging techniques. For example, CCTA image data can be generated using a scanner to create images of the heart, including the coronary arteries and other blood vessels extending therefrom. The acquired CCTA image data can then be used to generate a three-dimensional image model of features included in the CCTA image data (e.g., the right coronary artery 205, the left coronary artery 210, the left anterior descending artery 215, the circumflex branch of the left coronary artery 220, the aorta 240, and other blood vessels associated with the heart that appear in the image data).

[0045] In various embodiments, different imaging methods can be used to collect image data. For example, ultrasound or magnetic resonance imaging (MRI) can be used. In some embodiments, the imaging method includes the use of a contrast agent to help identify the structure of the coronary arteries, which is injected into the patient before the imaging procedure. Various imaging methods may have advantages and disadvantages, including resolution and suitability for imaging the coronary arteries. Imaging methods that can be used to collect coronary artery image data are constantly improving as improvements in hardware (e.g., sensors and emitters) and software are made. The disclosed systems and methods contemplate the use of CCTA image data and / or any other type of image data that can provide or convert a representative 3D depiction of the coronary arteries, plaque contained therein, and perivascular fat located proximate to the coronary arteries, including plaque, to obtain attenuation or radiodensity values ​​for the coronary arteries, plaque, and / or perivascular fat.

[0046] With further reference to FIG. 3 , a particular image 310 of image data 305 is shown, representing an image of a portion of the left anterior descending artery 215. Image 310 contains image information, and the smallest point of information manipulated by the system, generally referred to herein as a pixel, is, for example, pixel 315 of image 310. The resolution of the imaging system used to capture the image data affects the size of the smallest feature discernible in the image. Furthermore, subsequent manipulation of the image may affect the dimensions of the pixel. As an example, digital image 310 may include 4,000 pixels in each row and 3,000 pixels in each column. Pixel 315, and each pixel in image data 310 and image data 305, may be associated with a radiodensity value corresponding to the density of the pixel in the image. An exemplary mapping of pixel 315 to a point on the Hounsfield scale 320 is shown in FIG. 3 . Hounsfield scale 320 is a quantitative scale for describing radiodensity. The Hounsfield unit scale linearly transforms the original linear attenuation coefficient measurements to one in which the radiometric density of distilled water at standard pressure and temperature is defined as zero Hounsfield units (HU) and the radiometric density of air at standard pressure and temperature is defined as -1000 HU. While Figure 3 shows an example of mapping pixels 315 of image 310 to points on the Hounsfield scale 320, such association of pixels with radiometric density values ​​can also be done with 3D data, such as after the image data 305 has been used to generate a three-dimensional representation of the coronary arteries.

[0047] Once the data has been acquired and rendered into a three-dimensional representation, various processes can be performed on the data to identify an analysis region. For example, a three-dimensional representation of a coronary artery may be segmented to define multiple portions of the artery and identified as such in the data. In some embodiments, the data may be filtered (e.g., smoothed) by various methods to remove anomalies that are the result of scanning or various other errors. Various known methods for segmenting and smoothing D data may be used, and therefore, for the sake of brevity of this disclosure, will not be described in further detail herein.

[0048] 4A is a block diagram illustrating a computer system 400 in which various embodiments may be implemented. The computer system 400 includes a bus 402 or other communication mechanism for communicating information, and a hardware processor or processors 404 coupled with the bus 402 for processing information. The hardware processor 404 may be, for example, one or more general-purpose microprocessors.

[0049] Computer system 400 also includes a main memory 406, such as a random access memory (RAM), cache, and / or other dynamic storage device, coupled to bus 402 for storing information and instructions executed by processor 404. Main memory 406 may also be used to store temporary variables or other intermediate information during execution of instructions to be executed by processor 404. Such instructions, when stored in a storage medium accessed by processor 404, make computer system 400 a special-purpose machine customized to perform the operations specified in the instructions. Main memory 406 may include, for example, instructions for analyzing image information to determine characteristics of coronary arteries (e.g., plaque, perivascular fat, and coronary arteries) and generating a patient report including information characterizing aspects of the patient's health related to the coronary arteries. For example, one or more metrics may be determined, including one or more of the slope / gradient of a feature, maximum density, minimum density, the ratio of the slope of one feature to the slope of another feature, the ratio of the maximum density of one feature to the maximum density of another feature, the ratio of the minimum density of a feature to the minimum density of the same feature, or the ratio of the minimum density of a feature to the maximum density of another feature.

[0050] Computer system 400 further includes a read-only memory (ROM) 408 or other static storage device coupled to bus 402 for storing static information and instructions for processor 404. A storage device 410, such as a magnetic disk, optical disk, or USB thumb drive (flash drive), is provided and coupled to bus 402 for storing information and instructions.

[0051] The computer system 400 can be coupled via bus 402 to a display 412, such as a cathode ray tube (CRT) or LCD display (or touch screen), for displaying information to a computer user. An input device 414, including alphanumeric and other keys, is coupled to bus 402 for communicating information and command selections to the processor 404. Another type of user input device is a cursor control 416, such as a mouse, trackball, or cursor direction keys, for communicating directional information and command selections to the processor 404 and for controlling cursor movement on the display 412. This input device typically has two degrees of freedom, a first axis (e.g., x) and a second axis (e.g., y), allowing the device to specify a position in a plane. In some embodiments, the same directional information and command selections as cursor control may be implemented via receiving touches on a cursorless touchscreen.

[0052] Computer system 400 may include a user interface module for implementing a GUI, which may be stored on mass storage as computer-executable program instructions executed by computer device(s). Computer system 400 may further implement the techniques described herein using customized hardwired logic, one or more ASICs or FPGAs, firmware, and / or program logic that, in combination with the computer system, causes or programs computer system 400 to become a special-purpose machine, as described below. According to one embodiment, the techniques herein are performed by computer system 400 in response to processor(s) 404 executing one or more sequences of one or more computer-readable program instructions contained in main memory 406. Such instructions may be read into main memory 406 from another storage medium, such as storage device 410. Execution of the sequences of instructions contained in main memory 406 causes processor(s) 404 to perform the process steps described herein. In alternative embodiments, hard-wired circuitry may be used in place of or in combination with software instructions.

[0053] Various forms of computer-readable storage media may be involved in carrying one or more sequences of one or more computer-readable program instructions to the processor 404 for execution. For example, the instructions may initially be carried on a magnetic disk or solid state drive of a remote computer. The remote computer may load the instructions into its dynamic memory and send the instructions over a telephone line using a modem. A modem local to the computer system 400 may receive the data on the telephone line and use an infrared transmitter to convert the data to an infrared signal. An infrared detector may receive the data carried in the infrared signal and appropriate circuitry may place the data on the bus 402. The bus 402 carries the data to the main memory 406, from which the processor 404 retrieves and executes the instructions. The instructions received by the main memory 406 may optionally be stored on a storage device 410 either before or after execution by the processor 404.

[0054] Computer system 400 also includes a communication interface 418 coupled to bus 402. The communication interface 418 provides a two-way data communication coupling to a network link 420 that is connected to a local network 422. For example, the communication interface 418 may be an Integrated Services Digital Network (ISDN) card, a cable modem, a satellite modem, or a modem to provide a data communication connection to a corresponding type of telephone line. As another example, the communication interface 418 may be a local area network (LAN) card to provide a data communication connection to a corresponding LAN (or a WAN component for communicating with a WAN). Wireless links may also be implemented. In any such implementation, the communication interface 418 sends and receives electrical, electromagnetic or optical signals that carry digital data streams representing various types of information.

[0055] Network link 420 typically provides data communication through one or more networks to other data devices. For example, network link 420 may provide a connection through local network 422 to a host computer 424 or to data equipment operated by an Internet Service Provider (ISP) 426. ISP 426 provides data communication services through the worldwide packet data communication network now commonly referred to as the "Internet" 428. Local network 422 and Internet 428 both use electrical, electromagnetic, or optical signals that carry digital data streams. The signals through the various networks, the signals on network link 420, and the signals through communication interface 418, which carry the digital data to and from computer system 400, are examples of transmission media.

[0056] Computer system 400 can send messages and receive data, including program code, through the network(s), network link 420 and communication interface 418. In the Internet example, a server 430 might transmit a requested code for an application program through Internet 428, ISP 426, local network 422 and communication interface 418.

[0057] The received code may be executed by processor 404 as it is received, and / or stored in storage device 410, or other non-volatile storage for later execution.

[0058] Thus, in an embodiment, the computer system 105 includes a non-transitory computer storage medium storage device 410 configured to store at least patient image information. The computer system 105 may also include a non-transitory computer storage medium storage that stores instructions for the one or more processors 404 to execute a process (e.g., a method) for characterizing coronary plaque tissue data and perivascular tissue data using image data collected from a computed tomography (CT) scan along a blood vessel, the image information including radiodensity values ​​of coronary plaque and perivascular tissue located adjacent to the coronary plaque. By executing the instructions, the one or more processors 404 can quantify radio density in regions of coronary plaque in the image data; quantify radio density in at least one region of corresponding perivascular tissue adjacent to the coronary plaque in the image data; determine a gradient of the quantified radio density values ​​in the coronary plaque and the corresponding perivascular tissue; determine a ratio of the quantified radio density values ​​in the coronary plaque and the corresponding perivascular tissue; and characterize the coronary plaque by analyzing one or more of the gradient of the quantified radio density values ​​in the coronary plaque and the corresponding perivascular tissue, or the ratio of the radio density values ​​of the coronary plaque and the corresponding perivascular tissue.

[0059] Various embodiments of the present disclosure may be systems, methods, and / or computer program products at any possible level of technical detail. A computer program product may include a computer-readable storage medium (or medium) having computer-readable program instructions thereon for causing a computer processor to execute aspects of the present disclosure. For example, the functions described herein may be performed by software instructions executed by one or more hardware processors and / or any other suitable computing devices, and / or in response to the software instructions being executed by one or more hardware processors and / or any other suitable computing devices. The software instructions and / or other executable code may be read from a computer-readable storage medium (or medium).

[0060] A computer-readable storage medium may be a tangible device capable of holding and storing data and / or instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but is not limited to, an electronic storage medium (including any volatile and / or non-volatile electronic storage device), a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. A non-exhaustive list of more specific examples of computer-readable storage media includes the following: portable computer diskettes, hard disks, solid-state drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital versatile disks (DVD), memory sticks, floppy disks, mechanically encoded devices such as punch cards or ridge structures in grooves having instructions recorded thereon, and any suitable combination of the foregoing. As used herein, computer-readable storage media is not to be construed as a transitory signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through fiber optic cables), or electrical signals transmitted over electrical wires.

[0061] The computer-readable program instructions described herein can be downloaded to each computing / processing device from a computer-readable storage medium or to an external computer or storage device over a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network may be comprised of copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage on a computer-readable storage medium in each computing / processing device.

[0062] Computer-readable program instructions (also referred to herein as, for example, “code,” “instructions,” “modules,” “applications,” “software applications,” etc.) for performing operations of the present disclosure may be either assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, configuration data for integrated circuits, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., firmware instructions, state setting data, configuration data for integrated circuits, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and procedural programming languages ​​such as the “C” programming language. The computer-readable program instructions may be callable from other instructions or from themselves and / or may be called in response to a detected event or interrupt. Computer-readable program instructions configured to execute on a computing device may be provided on a computer-readable storage medium and / or as a digital download (and may originally be stored in a compressed or installable format that requires installation, decompression, or decryption before execution) and then stored on a computer-readable storage medium. Such computer-readable program instructions may be stored, partially or entirely, in a memory device (e.g., a computer-readable storage medium) of the executing computing device for execution by the computing device. The computer-readable program instructions may execute entirely on the user's computer (e.g., the executing computing device), partially on the user's computer as a stand-alone software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer or server.In the latter scenario, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, including, for example, a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may utilize the state information of the computer-readable program instructions to execute the computer-readable program instructions and personalize the electronic circuitry to carry out aspects of the present disclosure.

[0063] Aspects of the present disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0064] These computer-readable program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus to manufacture a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for performing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer-readable program instructions can also be stored on a computer-readable storage medium that can direct a computer, programmable data processing apparatus, and / or other device to function in a particular manner, and in such a way, the computer-readable storage medium having instructions stored therein constitutes an article of manufacture containing instructions that implement an aspect of the function / act specified in the flowchart(s) and / or block diagram(s) block or blocks.

[0065] Computer-readable program instructions may be loaded into a computer, other programmable data processing device, or other device to cause the computer, other programmable device, or other device to perform a series of operational steps to generate a computer-implemented process, such that the instructions executing on the computer, other programmable device, or other device perform the function / acts specified in the flowchart and / or block diagram block or blocks. For example, the instructions may initially reside on a magnetic disk or solid-state drive of a remote computer. The remote computer may load the instructions and / or modules into its dynamic memory and transmit the instructions over a telephone, cable, or optical line using a modem. A modem local to the server computing system may receive the data over the telephone / cable / optical line and, using a converter device including appropriate circuitry, place the data on a bus. The bus carries the data to memory, from which a processor retrieves and executes the instructions. The instructions received by memory are optionally stored on a storage device (e.g., a solid-state drive) either before or after execution by a computer processor.

[0066] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams may represent a module, segment, or portion of instructions, which constitute one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may in fact be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending on the functionality involved. Moreover, in some implementations, certain blocks may be omitted. Also, the methods and processes described herein are not limited to a particular order, and the associated blocks or states may be executed in any other order as appropriate.

[0067] 4B is a block diagram illustrating an example of representative instructions that may be executed by one or more computer hardware processors in one or more computer modules in a representative processing system (computer system) 120 that may implement various embodiments described herein. As shown in FIG. 1, processing system 120 may be implemented in one computer (e.g., a server) or two or more computers (two or more servers). While instructions are represented in FIG. 4B as being in seven modules 450, 455, 460, 465, 470, 475, and 480, in various implementations, executable instructions may be in fewer modules, including one module, or in more modules.

[0068] Processing system 120 includes image information stored in storage device 410, which may come from network 125 shown in FIG. 1 . The image information may include image data, scan information, and / or patient data. In this example, storage device 410 also includes stored plaque information of other patients. For example, the stored plaque information of other patients may be stored in a database on storage device 410. In another example, the stored plaque information of other patients is stored in a storage device in communication with processing system 120. The stored plaque information of other patients may be a collection of information from one, tens, hundreds, thousands, tens of thousands, hundreds of thousands, millions, or more patients.

[0069] The information for each patient may include characteristics of the patient's plaque, such as the density or density gradient of the plaque, and the location of the plaque relative to nearby or adjacent perivascular tissue. The information for each patient may include patient information. For example, the information may include one or more of: gender, age, body mass index (BMI), medications, blood pressure, heart rate, weight, height, race, build, smoking history, medical history or diagnosed history of diabetes, medical history or diagnosed history of hypertension, medical history of coronary artery disease, family history of coronary artery disease and / or other diseases, or one or more test results (e.g., blood test results). The information for each patient may include scan information. For example, the information may include one or more of the following: contrast-to-noise ratio, signal-to-noise ratio, tube current, tube voltage, contrast type, contrast volume, flow rate, flow duration, slice thickness, slice spacing, pitch, vasodilators, beta-blockers, recon option (repeated projection or filtered backprojection), recon type (standard or high-resolution), field of view, rotation speed, gating (fluoroscopic triggering or retrospective gating), stent, heart rate, or blood pressure. Cardiac information may also be included for each patient. For example, the information may include plaque characterization, including one or more of density, volume, geometry (shape), location, remodeling, baseline anatomy (diameter, length), compartment (medial, lateral, medial), stenosis (diameter, area), myocardial mass, plaque volume, and / or plaque composition, texture, or uniformity.

[0070] The processing system 120 also includes memory 406, 408, which may be the processing system's main memory or read-only memory (ROM). The memory 406, 408 stores instructions executable by one or more computer hardware processors 404 (collectively referred to herein as "modules") for characterizing coronary artery plaque. With reference to this figure, the memory 406, 408 will be collectively referred to as memory 406 for simplicity. Examples of functions performed by the executable instructions are described below.

[0071] The memory 406 includes a module 450 that generates, from the image data stored in the storage device 410, a 2-D or 3-D representation of the coronary arteries, including plaque, and perivascular tissue adjacent to or proximal to the plaque-containing coronary arteries. The generation of the 2-D or 3-D representation of the coronary arteries can be performed from a series of images 305 (e.g., CCTA images), as described above with reference to FIG. 3. Once the representation of the coronary arteries is generated, different portions or segments of the coronary arteries can be identified for evaluation. For example, portions of interest in the right coronary artery 205, the left anterior descending artery 215, or the circumflex branch of the left coronary artery 220 can be identified as analysis regions (regions of interest) based on input from a user or based on features determined from the representation of the coronary arteries (plaque).

[0072] In module 460, one or more computer hardware processors quantify the radiodensity of the region of the coronary plaque. For example, the radiodensity of the region of the coronary plaque is set to a value on the Hounsfield scale. In module 465, one or more computer hardware processors quantify the radiodensity of perivascular tissue adjacent to the coronary plaque and quantify the radiodensity value of the lumen of the vessel of interest. In module 470, one or more computer hardware processors determine the gradient of the radiodensity values ​​of the plaque, perivascular tissue, and / or lumen. In module 475, one or more computer hardware processors determine one or more ratios of the radiodensity values ​​of the plaque, perivascular tissue, and / or lumen. Next, in module 480, one or more computer hardware processors characterize the coronary plaque using the gradient of the plaque, perivascular tissue, and / or lumen and / or characterize the ratio of the radiodensity value of the coronary plaque to the perivascular tissue and / or lumen, including comparing the gradient and / or ratio with a database containing information on the gradient and ratio of plaques of other patients. For example, the gradients and / or ratios are compared to patient data stored in memory device 410. Determining the gradients and ratios of plaque, perivascular tissue, and lumen is described in more detail with reference to Figures 6-12.

[0073] FIG. 5A shows an example flowchart of a process 500 for analyzing coronary artery plaque. In block 505, process 500 generates image information including image data related to the coronary arteries. In various embodiments, this may be performed by scanner 130B (FIG. 1). In block 510, a processing system may receive the image information via network 125 (FIG. 1), where the image information includes the image data. In block 515, process 500 generates a 3D representation of the coronary arteries, including perivascular fat and plaque, on the processing system. The functions of blocks 505, 510, and 515 may be performed using, for example, various imaging techniques (e.g., CCTA) to generate the image data, communication techniques to transfer the data over a network, and processing techniques to generate the 3D representation of the coronary arteries from the image data.

[0074] In block 520, the processing system performs a portion of step 500 to analyze coronary artery plaque, which is described in further detail with reference to step 550 of Figure 5B. Additional details of this process of analyzing coronary artery plaque are described with reference to Figures 6-12.

[0075] FIG. 5B shows an example of a flowchart that expands on a portion of the flowchart of FIG. 5A for determining characteristics of coronary plaque. Referring now to FIG. 5B, in block 555, step 550 can utilize one or more processors 404 to quantify radiodensity in regions of coronary plaque. In block 560, step 550 can utilize one or more processors 404 to quantify radiointensity in at least one region of corresponding perivascular tissue (meaning perivascular tissue adjacent to the coronary plaque) in the image data. In block 565, step 550 determines gradients of the quantified radiodensity values ​​in the coronary plaque and the corresponding perivascular tissue. One or more processors 404 can be instrumental in determining these gradients. In block 570, step 550 can determine a ratio between the quantified radiodensity values ​​in the coronary plaque and the corresponding perivascular tissue, e.g., perivascular tissue adjacent to the coronary plaque. One or more processors 404 can determine these ratios. At block 575, process 550 may utilize one or more processors 404 to characterize the coronary plaque by analyzing one or more of the gradient of the quantified radiodensity values ​​in the coronary plaque and corresponding perivascular tissue, or the ratio of the radiodensity values ​​of the coronary plaque to the radiodensity values ​​of the corresponding perivascular tissue, after which process 550 may return to process 500, as indicated by circle A.

[0076] Referring again to FIG. 5A , at block 525, process 500 may compare the determined coronary plaque information for a particular patient with stored patient data, such as, for example, patient data stored in storage device 410. Figure 5 shows an example of coronary plaque information for a particular patient that may be compared with stored patient data. One or more pieces of scan information may be used to better understand and / or help determine the patient's coronary plaque information. Additionally, when comparing the coronary plaque information for a particular patient with previously stored coronary plaque information, for example, one or more characteristics of the patient may be compared. In some examples, the coronary plaque information for the patient being examined may be compared or analyzed with reference to patients having one or more of the same or similar patient characteristics. For example, the patient being examined may be compared with patients having the same or similar characteristics, such as gender, age, BMI, medications, blood pressure, heart rate, weight, height, race, build, smoking, diabetes, hypertension, history of coronary artery disease, family history, test results, etc. Such comparisons may be performed by various means, such as machine learning and / or artificial intelligence techniques. In some examples, a neural network is used to compare the patient's coronary artery information with that of a large number (e.g., 10,000 or more) of other patients, and for those patients with similar patient information and similar cardiac information, a plaque risk assessment for the examined patient can be determined.

[0077] Figure 6 shows an example of a region indicated by box 605 in which the contrast attenuation pattern in a proximal portion of a coronary artery lumen can be analyzed, with box 605 extending from a central region of blood vessel 665 toward vessel wall 661. Figure 6 shows another example of a region indicated by box 652 in which the contrast attenuation pattern in a portion of a coronary artery lumen of blood vessel 665 can be analyzed, with box 652 extending longitudinally relative to blood vessel 665 from a central region of blood vessel 665 toward vessel wall 661. Figure 6 further shows an example of a region indicated by box 662 in which the contrast attenuation pattern of a portion of the lumen, a portion of fibrous plaque 610, and a portion of plaque 620 can be analyzed, with box 662 thus covering a portion of blood vessel 665, a portion of fibrous plaque 610, and a portion of plaque 620. 6 further illustrates an example of a region indicated by box 642, which extends across the portion of plaque 635 and the density of fat 640 disposed adjacent to plaque 635, in which the contrast attenuation pattern of the portion of plaque 635 and the portion of fat 640 disposed adjacent to plaque 635 may be analyzed. Information determined by analyzing various aspects of the density of coronary artery characteristics (e.g., the lumen, plaque, and / or perivascular fat) can be combined with other information to determine characteristics of the patient's arteries. In some examples, the determined information may include, for either the lumen, plaque, or perivascular fat, the slope / gradient of the feature, the maximum density, the minimum density, the ratio of the slope of the density of one feature to the slope of the density of another feature, the ratio of the maximum density of one feature to the maximum density of another feature, the ratio of the minimum density of one feature to the minimum density of the same feature, the directionality of the density ratio, which may include one or more of the density ratios, for example, the density ratio of features facing one direction to features facing the opposite direction (e.g., the radial density ratio of features facing the inside of the myocardium to features facing the outside of the pericardium), or the ratio of the minimum density of one feature to the maximum density of another feature. Such determined information may indicate a clear difference in plaque risk in a patient. In some examples, the determined information (e.g., as described above) may be used in conjunction with the percentage stenosis diameter to determine the characteristics of the patient's arteries.

[0078] Still referring to FIG. 6 , in one example of the directionality of the radio density ratio, the density of a portion of the necrotic core plaque 615 relative to the density of a portion of the blood vessel 665 (e.g., plaque:vessel inward ratio) can be determined, which can indicate a certain risk of plaque. In another example of the directionality of the radio density ratio, the density of a portion of the blood vessel 665 relative to the density of the necrotic core plaque 615 (e.g., vessel:plaque outward ratio) can be determined, which can indicate a certain risk of plaque. In another example, the density ratio of the necrotic core plaque 615 relative to the density of a portion of the blood vessel 665 (e.g., plaque:vessel inward ratio) can be compared to the density ratio of the necrotic core plaque 615 relative to the fibrous plaque 620 (e.g., plaque:plaque outward ratio) which can indicate a certain risk of plaque. In other examples, adjacently located features can be used to determine radio density values ​​in the inward and / or outward directions, which can be used to indicate the risk associated with the plaque. Such ratios can provide a clear distinction in plaque risk. Various embodiments of the directional radiodensity values ​​and / or directional radiodensity ratios can be included with any of the other information described herein to indicate plaque risk.

[0079] Compartment size can also be used to indicate plaque-associated risk. For example, the assessment of plaque-associated risk can be based at least in part on compartment size, with radiodensity ratios influencing the risk assessment and functions of compartment size also influencing the risk assessment. The presence of plaque in a patient with a high plaque:fat ratio may indicate high-risk plaque, but a small amount of plaque (e.g., a small compartment of plaque) would be at higher risk than a large compartment of the same plaque with the same radiodensity ratio of plaque to fat. In one embodiment, the size (e.g., volume) of compartment features (e.g., of the lumen, plaque, perivascular tissue (fat), and myocardium) can be determined, and the radiodensity ratio can be determined, which can then be weighted based on compartment size. For example, a larger compartment can have a higher ratio weight, making it more indicative of plaque-associated risk. Similarly, a smaller compartment can have a lower ratio weight, making it less indicative of plaque-associated risk. In some embodiments, only plaque compartment size is used to weight (or adjust) the ratio. In some embodiments, the compartment sizes of both features used in the radiodensity ratio can be used to weight the ratio to determine the resulting risk. In some embodiments, the compartment sizes of any of the plaque, lumen, perivascular tissue, and myocardium are used to weight (or adjust) the risk associated with the radiodensity ratio. In embodiments, the compartment sizes of one or more of the plaque, lumen, perivascular tissue, and myocardium are used to weight the risk associated with the radiodensity ratio. Various embodiments that use compartment size to determine plaque risk can be included along with any of the other information described herein to indicate plaque risk.

[0080] FIG. 7 illustrates the same blood vessel 665 and plaque and fat characteristics as illustrated in FIG. 6 , and further illustrates additional examples of regions of an artery, as well as plaque and / or perivascular fat near the artery, that may be analyzed to determine characteristics of a patient's artery. Such regions are depicted in FIG. 7 by rectangular boxes, similar to the diagram in FIG. 6 . While specific locations of the rectangular boxes are depicted in FIGS. 6 and 7 , these are merely examples of regions that may be analyzed. In one example, FIG. 7 illustrates a box 660 that includes a portion of a blood vessel 665, a portion of a necrotic core plaque 615, a portion of a fibrous plaque 610, a portion of a plaque 620, and a portion of a fat 625. In another example, FIG. 7 illustrates a box 655 that includes a portion of a blood vessel 665, a portion of a fibrous plaque 610, a portion of a plaque 620, a portion of a necrotic core plaque 615, and a portion of a fat 625. Box 655 may, in some cases, illustrate a general region for analysis due to the presence of three different types of plaque 610, 615, and 620 and adjacently disposed fat 625. Specific portions of the general region for analysis may be analyzed to better understand the signature formed by adjacent features. For example, FIG. 7 illustrates a general region 665 that includes box 660 (discussed above), box 673 extending across a portion of fibrous plaque 610 and plaque 620, and box 674 extending across a portion of plaque 620 and perivascular fat 625. As another example, FIG. 7 also illustrates another box 672 extending across a blood vessel 655 and a portion of necrotic core plaque 615. As a further example, FIG. 7 illustrates box 671 extending across a portion of blood vessel 665 and fat 640 juxtaposed to blood vessel 665. As a further example, FIG. 7 illustrates box 670 extending across a portion of blood vessel 665 and plaque 635. As indicated above, the characteristics of a patient's arteries that can be analyzed based on these features include, but are not limited to: 1. The ratio of lumen attenuation to plaque attenuation, where a volumetric model of scan-specific attenuation density gradients within the lumen adjusts for reduced lumen density throughout plaque lesions that are more functionally significant in terms of risk value. 2. Among the subset of plaques considered to be "calcified plaques" based on the ratio of plaque attenuation to fat attenuation, highly radiodense plaques are considered to be of low risk. 3. Ratio of lumen attenuation / plaque attenuation / fat attenuation. 4. The ratio of #1-3 as a function of the 3D shape of the atherosclerosis, which can include 3D texture analysis of the plaque. 5. The three-dimensional volumetric shape and path of the lumen along with the attenuation density from the beginning to the end of the lumen. 6. By looking at the plaques before and after a certain plaque and the type of plaque, you can learn more about the risk. 7. Determine "high-risk plaque" by "subtracting" calcified (high-density) plaque to obtain a better absolute measure of high-risk plaque (low-density plaque). In other words, this particular embodiment involves identifying calcified plaque and excluding it from further analysis of plaque for the purpose of identifying high-risk plaque.

[0081] In some embodiments, the systems, devices, and methods described herein can automatically and / or dynamically perform quantitative analysis of various parameters related to plaque, cardiovascular arteries, and / or other structures. For example, rather than a physician visually inspecting or generally assessing a patient, in some embodiments, medical images can be sent to a back-end main server configured to perform such analysis, advantageously in a consistent, objective, and / or reproducible manner. In some embodiments, the systems, methods, and devices described herein can provide quantified measurements of one or more features of coronary CT images using an automated and / or dynamic process. For example, in some embodiments, the main server system can be configured to identify one or more vessels, plaque, and / or fat from the medical images. Based on the identified features, in some embodiments, the system can be configured to generate one or more quantified measurements from the raw medical images, such as, for example, the density and / or radiodensity of one or more regions of plaque, identification of stable and / or unstable plaque, perivascular fat, pericoronary adipose tissue (PCAT), fat attenuation index (FAI), its volume, its surface area, its geometric shape, its heterogeneity, and / or the like. In some embodiments, the system can also generate one or more quantified measurements of the blood vessels from the raw medical images, such as diameter, volume, morphology, and / or the like.

[0082] Based on the identified features and / or quantified measurements, in some embodiments, the system can be configured to use the raw medical images to generate a risk assessment and / or track the progression of a plaque-based disease or condition, such as atherosclerosis, stenosis, ischemia, myocardial infarction, and / or major adverse cardiovascular events (MACE). As described further herein, in some embodiments, the system can perform risk assessment and / or tracking of plaque-based disease progression based on other patient information. For example, tracking of plaque-based disease progression can be performed by comparing or evaluating a patient's medical image features and patient information (e.g., age, sex, BMI, medications, blood pressure, heart rate, height, weight, race, whether the patient is a smoker or non-smoker, medical history, family history of disease, etc.) with other patients' medical image features and related patient information, including outcomes over a period of time.

[0083] Additionally, in some embodiments, the system may be configured to generate a GUI visualization of one or more identified features and / or quantified measurements, such as a quantized color mapping of different features. In some embodiments, the systems, devices, and methods described herein are configured to utilize medical image-based processing to assess a subject's risk of a cardiovascular event, a major adverse cardiovascular event (MACE), rapid plaque progression, and / or non-response to drug therapy and / or lifestyle changes and / or other therapies and / or non-invasive procedures. In particular, in some embodiments, the system may be configured to automatically and / or dynamically assess such health risks of a subject by analyzing only non-invasively obtained medical images. In some embodiments, one or more of the processes can be automated using artificial intelligence (AI) and / or machine learning (ML) algorithms. In some embodiments, one or more of the processes described herein can be performed within minutes in a reproducible manner. This is in contrast to existing means today that do not produce reproducible prognoses or assessments and require time-consuming and / or invasive procedures.

[0084] In some embodiments, image information, including multiple images of a patient's coronary arteries and patient information / characteristics, may be provided from one or more devices over a network to one or more servers of a processing system. The processing system is configured to generate coronary artery information using the multiple images of the patient's coronary arteries to generate two-dimensional and / or three-dimensional data representations of the patient's coronary arteries. The processing system then analyzes the data representations to generate a patient report documenting the patient's health status and risks associated with coronary plaque. The patient report may include images of the patient's arteries and visual depictions of types of coronary plaque within or near the coronary arteries. Machine learning or other artificial intelligence techniques may be used to compare the data representations of the patient's coronary arteries with data representations of other patients (e.g., stored in a database) to determine additional information regarding the patient's health. In some embodiments, the artificial intelligence may be trained using a dataset of data representations of other patients to identify data correlations. For example, based on the status of specific plaque in the patient's coronary arteries, a patient's likelihood of suffering a heart attack or other adverse coronary events may be determined. Additional information, for example, regarding the patient's CAD risk may also be determined.

[0085] In some embodiments, the coronary plaque information of the patient being examined may be compared with or analyzed with reference to patients having one or more of the same or similar patient characteristics. For example, the patient being examined may be compared with patients having the same or similar characteristics of gender, age, BMI, medication, blood pressure, heart rate, weight, height, race, build, smoking, diabetes, hypertension, history of coronary artery disease, family history, and test results. Such comparisons may be performed by various means, such as machine learning and / or artificial intelligence techniques. In some examples, a neural network is used to compare the patient's coronary artery information with that of a large number (e.g., 10,000 or more) of other patients. For such patients with similar patient information and similar cardiac information, a plaque risk assessment for the patient being examined may be determined.

[0086] In some embodiments, deep learning (DL), machine learning (ML), and artificial intelligence (AI) methods can be used to analyze image information. In one example, this analysis can consist of image segmentation, feature extraction, and classification. In some embodiments, ML methods can include image feature extraction from raw data and image-based learning. In some embodiments, ML methods can receive input from a large training set and learn to ignore variations that could otherwise skew the results of the method. In some embodiments, DL can construct neural networks (NNs) with three or more layers, which can improve the accuracy of the determination. Advantageously, in some embodiments, DL can eliminate the need for data preprocessing and instead process raw data. For example, a human can input a hierarchy of important features of coronary artery image information for the ML algorithm to make a determination, while the DL algorithm can determine which features are important and use these features to make the determination. Advantageously, in some embodiments, the DL algorithm can tune itself for accuracy and precision. In some embodiments, ML and DL algorithms can perform supervised learning, unsupervised learning, and reinforcement learning.

[0087] In some embodiments, NN approaches, including convolutional neural networks (CNNs) and recurrent convolutional neural networks (RCNNs), among others, can be used to analyze information in a manner similar to the high-level cognitive functions of the human mind. In some embodiments, the NN approach involves training an object recognition system with a large number of medical images to teach it patterns in the images that correlate with specific labels. In some embodiments, the CNN can be configured as an NN in which nodes in each layer are clustered, the clusters overlap, and each cluster feeds multiple nodes in the next layer. In some embodiments, the RCNN can be configured as a CNN in which recurrent connections are incorporated into each convolutional layer. Advantageously, in some embodiments, the recurrent connections can make object recognition a dynamic process, even though the input is static.

[0088] In some embodiments, a vessel identification algorithm, a coronary artery identification algorithm, and / or a plaque identification algorithm can be trained on a plurality of medical images in which regions of one or more vessels, coronary arteries, and / or plaque have been pre-identified. Based on such training, such as using CNNs in some embodiments, the system can be configured to automatically and / or dynamically identify the presence and / or parameters of vessels, coronary arteries, and / or plaque from raw medical images. In some embodiments, the system can be configured to utilize one or more AI and / or ML algorithms to identify and / or analyze vessels or plaque, derive one or more quantification indices and / or classifications, and / or generate a treatment plan. In some embodiments, the system can be configured to utilize AI and / or ML algorithms to identify regions within an artery exhibiting plaque accumulation within, along, within, and / or outside the artery. In some embodiments, inputs to the AI ​​and / or ML algorithms may include patient images and patient information (or characteristics), such as one or more of age, sex, body mass index (BMI), medications, blood pressure, heart rate, height, weight, race, whether the patient is a smoker or non-smoker, body habitus (e.g., "body size" or "body type"), although "body size" or "body type" can refer to a wide range of factors, medical history, diabetes, hypertension, history of coronary artery disease (CAD), dietary habits, medication history, family history of disease, information related to other previously collected image information, exercise habits, drinking habits, lifestyle information, or test results. In examples where a NN is used, the NN can be trained using information from multiple patients, where each patient's information may include a medical image and one or more patient characteristics.

[0089] In some embodiments, the system can be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically identify one or more regions of plaque using image processing. For example, in some embodiments, one or more AI and / or ML algorithms can be trained using CNNs on a set of medical images in which regions of plaque have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify regions of plaque directly from the medical images. In some embodiments, the system can be configured to identify the vessel wall and lumen wall for each identified coronary artery in the medical images. In some embodiments, the system is then configured to determine the volume between the vessel wall and the lumen wall as plaque. In some embodiments, the system can be configured to identify regions of plaque based on radiodensity values ​​typically associated with plaque, for example, by setting a predetermined threshold or range of radiodensity values ​​typically associated with plaque, with or without normalization using a normalizer.

[0090] In some embodiments, one or more vascular morphological parameters and / or plaque parameters may constitute quantified parameters derived from medical images. For example, in some embodiments, the system may be configured to determine one or more vascular morphological parameters and / or plaque parameters using AI and / or ML algorithms or other algorithms. As another example, in some embodiments, the system may be configured to determine one or more vascular morphological parameters, such as a classification of arterial remodeling due to plaque, which may further include positive arterial remodeling, negative arterial remodeling, and / or intermediate arterial remodeling. In some embodiments, the classification of arterial remodeling is determined based on the ratio of the maximum vessel diameter in the region of the plaque to a normal reference vessel diameter in the same region, which can be retrieved from a normal database. In some embodiments, the system may be configured to classify arterial remodeling as positive if the ratio of the maximum vessel diameter in the region of the plaque to the normal reference vessel diameter in the same region is 1.1 or greater. In some embodiments, the system may be configured to classify arterial remodeling as negative if the ratio of the maximum vessel diameter in the region of the plaque to the normal reference vessel diameter in the same region is less than 0.95. In some embodiments, the system may be configured to classify arterial remodeling as intermediate if the ratio of maximum vessel diameter to normal reference vessel diameter in the region of plaque is between 0.95 and 1.1.

[0091] In some embodiments, the system is configured to classify the subject's atherosclerosis into one or more of high risk, medium risk, or low risk based on the quantified atherosclerosis. In some embodiments, the system is configured to classify the subject's atherosclerosis based on the quantified atherosclerosis using AI, ML, and / or other algorithms. In some embodiments, the system is configured to classify the subject's atherosclerosis by combining and / or weighting one or more of surface area, volume, heterogeneity index, and radiodensity ratio of one or more regions of plaque.

[0092] In some embodiments, the system may be configured to automatically and / or dynamically identify one or more regions of fat, such as epicardial fat, in medical images, e.g., using one or more AI and / or ML algorithms. In some embodiments, the one or more AI and / or ML algorithms may be trained using CNNs on a set of medical images in which regions of fat have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify regions of fat directly from the medical images. In some embodiments, the system may be configured to identify regions of fat based on radiodensity values ​​typically associated with fat, e.g., by setting a predetermined threshold or range of radiodensity values ​​typically associated with fat, with or without normalization using a normalizer.

[0093] In some embodiments, the system is configured to utilize AI, ML, and / or other algorithms to characterize changes in calcium scores based on one or more plaque parameters derived from medical images. For example, in some embodiments, the system may be configured to utilize AI and / or ML algorithms trained using CNNs and / or datasets of known medical images that combine identified plaque parameters with calcium scores. In some embodiments, the system may be configured to characterize changes in calcium scores by accessing a known dataset of the same stored in a database. For example, the known dataset may include a dataset of changes in calcium scores and / or medical images of other subjects and / or plaque parameters derived therefrom. In some embodiments, the system may be configured to characterize changes in calcium scores and / or determine their causes on a vessel-by-vessel, segment-by-segment, plaque-by-plaque, and / or subject-by-subject basis.

[0094] In some embodiments, the systems disclosed herein can be used to dynamically and automatically determine the type, length, diameter, gauge, strength, and / or any other stent parameters required for a particular patient based on processed medical image data, e.g., using AI, ML, and / or other algorithms.

[0095] In some embodiments, the system may be configured to utilize AI and / or ML algorithms to generate a patient-specific report, which in some embodiments may include documents, AR experiences, VR experiences, videos, and / or audio components.

[0096] FIG. 8A is a block diagram illustrating an example of a system and / or process 800 (both referred to herein as the “system” for ease of reference) for identifying patient characteristics and / or risk information using AI / ML based on non-invasively acquired patient medical images and / or patient information. Current patient medical data, including images and / or patient information, is first acquired and electronically stored in medical data storage 816 (e.g., cloud storage, a hard disk, etc.). System 800 retrieves medical images and / or patient information 818 from medical data storage 816 and, if necessary, preprocesses them, e.g., reformatting them as needed for further processing. System 800 can also obtain a training set 822 of medical images and / or patient information from a stored dataset 820 of medical images and / or information of other patients (e.g., hundreds, thousands, tens of thousands, or hundreds of thousands or more other patients). The medical images and information of the other patients can be used to train an AI / ML algorithm 824 prior to processing the current patient's medical images and / or patient information 818, as described in further detail with reference to FIGS. 8C and 8D. In some embodiments, the AI / ML algorithm 824 may include one or more NNs, for example, as described with reference to the exemplary NN illustrated in Figure 8B. The ML / AI 824 processes the medical images and / or patient information 818 of the current patient and generates an output of identified characteristics and / or risk information 826 of the current patient.

[0097] 8B is a schematic diagram illustrating an example of a NN 812 that makes a decision 814 regarding a characteristic of a (current) patient based on inputs including a medical image 802. In some embodiments, the NN 812 can be configured to receive other inputs 804. In some embodiments, the other inputs 804 can be medical images of other patients. In some embodiments, the other inputs 804 can be medical histories of other patients. In some embodiments, the other inputs 804 can be medical histories of the (current) patient. The NN 812 can include an input layer 806. In some embodiments, the NN 812 can be configured to present training patterns to the input layer 806. In some embodiments, the NN 812 can include one or more hidden layers 808. In some embodiments, the input layer 806 can provide signals to the hidden layer 808, and the hidden layer 808 can receive signals from the input layer 806. In some embodiments, the hidden layer 808 can pass signals to the output layer 810. In some embodiments, one or more hidden layers 808 may be configured as convolutional layers (comprising neurons / nodes connected by weights, where the weights correspond to the strength of the connections between the neurons), pooling layers, fully connected layers, and / or normalization layers. In some embodiments, the NN 812 may be configured with pooling layers that combine the outputs of neuron clusters in one layer into a single neuron in the next layer. In some embodiments, max pooling and / or average pooling may be utilized. In some embodiments, max pooling may utilize the maximum value from each of the neuron clusters in the previous layer. In some embodiments, backpropagation may be utilized, and the corresponding neural network weights may be adjusted to minimize or reduce the error. In some embodiments, the loss function may comprise a binary cross-entropy loss function.

[0098] In some embodiments, the NN 812 can include an output layer 810. In some embodiments, the output layer 810 can receive signals from the hidden layer 808. In some embodiments, the output layer can generate a decision 814. In some embodiments, the NN 812 can make a decision 814 regarding a characteristic of the patient. In some embodiments, the decision 814 can include a set of plaque features. In some embodiments, the decision 814 can include the patient's risk of CAD.

[0099] 8C shows an example of a process in a flow diagram for training an artificial intelligence or machine learning model. Process 828 may be executed on a computer system. Various embodiments of such a process for training an AI or ML model may include additional features and / or exclude certain illustrated features (e.g., if a transformed dataset is accessed such that "apply transformation" in block 832 need not be performed).

[0100] 8C, in block 830, the system receives a dataset including patient health information, which may include medical images, a user survey, past test results, genetic information, and / or other patient information (e.g., height, weight, age, etc.) The dataset may also include non-health information, such as, for example, employment information, income information, transportation information, housing information, distance to a pharmacy, and / or distance to a healthcare provider.

[0101] In block 832, one or more transformations may be performed on the data. In one example, data may require transformation to conform to an expected input format, such as date format, units (e.g., pounds vs. kilograms, Celsius vs. Fahrenheit, inches vs. centimeters, etc.), address conventions, consistent formatting, etc. In some embodiments, addresses may be converted or altered to be in a consistent format and / or conform to standards issued by the United States Postal Service or similar postal authority. In some embodiments, data may undergo transformation to prepare it for use in training AI or ML algorithms; for example, categorical data may be encoded in a particular manner. In some embodiments, nominal data may be encoded using one-hot encoding, binary encoding, feature hashing, or other suitable encoding method. In some embodiments, ordinal data may be encoded using ordinal encoding, polynomial encoding, Helmert encoding, etc. In some embodiments, numeric data may be normalized, for example, by scaling the data to a maximum value of 1 and a minimum value of 0 or -1. These are merely examples, and one of ordinary skill in the art will readily recognize that other transformations are possible.

[0102] At block 834, the system may create a training dataset, a tuning dataset, and a test / validation dataset from the received dataset. In some embodiments, the training dataset 836 may be used during training to determine features for forming a predictive model. In some embodiments, the tuning dataset 838 may be used to select a final model and to prevent or correct overfitting that may occur during training with the training dataset 836, as the trained model should be generally applicable to a wide range of patients. In some embodiments, the test dataset 840 may be used after training and tuning to evaluate the model. For example, in some embodiments, the test dataset 840 may be used to check whether the model is overfitting to the training dataset. In some embodiments, the system may train a model using the training dataset 836 at block 842 in a training loop 856. In some embodiments, training may be performed in a supervised, unsupervised, or partially supervised manner. In some embodiments, at 844, the system may evaluate the model according to one or more evaluation criteria. For example, in some embodiments, evaluation may include determining how often the model determines a reasonable score for a patient's risk of CAD. At 846, in some embodiments, the system may determine whether the model meets one or more evaluation criteria. In some embodiments, if the model fails evaluation, the system may tune the model at 848 using the tuning dataset 838 and repeat training 842 and evaluation 844 until the model passes evaluation at 846. In some embodiments, once the model passes evaluation at 846, the system may exit the model training loop 856. In some embodiments, the test dataset 836 is run through the trained model 842, and at block 844, the system may evaluate the results.In some embodiments, if the evaluation fails, at block 846, the system may re-enter the training loop 856 for additional training and tuning. If the model passes, the system stops the learning process, resulting in a trained model 850. In some embodiments, the training process may be modified. For example, in some embodiments, the system does not use the tuning data set 838. In some embodiments, the model does not use the test data set 840.

[0103] Although described above with respect to determining a CAD risk score, the model can be trained for use in a variety of problems.

[0104] FIG. 8D shows an example of a process for training and using an AI / ML model. In some embodiments, the process of FIG. 8D can be used for various purposes, such as determining a patient's CAD risk score or characterizing plaque. In some embodiments, a training data store 858 can store data for training the model. For example, in some embodiments, the training data store 858 can store patient medical images, as well as information about the patient's health status, age, socioeconomic status, employment status, living arrangements, transportation, etc. In some embodiments, the training data can be annotated to include information about user outcomes. For example, in some embodiments, a user's outcome can indicate whether the user had to take time off work due to illness, whether they were hospitalized, whether they visited an emergency room, whether they visited an urgent care facility, etc. In some embodiments, the training data can indicate whether the user received medication to treat the illness at home, whether they received treatment at a hospital or other medical facility, whether they did not receive treatment, etc. At block 860, in some embodiments, the system can be configured to prepare training data for use in training the model, if the training data has not been previously prepared. In some embodiments, as briefly described above, preparing the training data may include performing one or more normalization procedures, standardization procedures, etc., such as converting units (e.g., between Fahrenheit and Celsius, between inches and centimeters, between pounds and kilograms), converting dates to a standard format, converting time to a standard format, etc. In some embodiments, similar treatments or symptoms may be described or coded differently by different healthcare providers. In some embodiments, different healthcare providers may use different coding schemes. In some embodiments, even within a particular coding scheme, healthcare providers may select different codes to indicate similar information. In some embodiments, a large number of similar codes may result in coding variability. Thus, in some embodiments, one code may be changed to another related code.In some embodiments, certain codes may be excluded if they are not relevant to the problem the model is intended to address. In some embodiments, excluding certain data may be desirable because adding data consumes additional computational resources and increases the time it takes to train the model. However, in some embodiments, excluding may be undesirable because there may be a risk of excluding factors that are actually associated with patient risk. In some embodiments, data preparation at block 860 may include modifying or removing coding data, treatment data, etc. At block 862, the system may extract features from the training data, and at block 864, the system may train a model using the training data to generate a model 866. At block 868, in some embodiments, the system may evaluate the model to determine whether it passes one or more criteria. In some embodiments, at decision point 870, if the model fails, the system may perform additional training. In some embodiments, if the model passes at decision point 870, the system may make available the model 872 after completion of training.

[0105] In some embodiments, the trained model 872 can be used to evaluate a particular user. User data 874 may be associated with a particular user for whom the output of the trained model 872 is desired. At block 876, the system may prepare data, for example, as described above in connection with stored training data. In some embodiments, at block 878, the system may extract features from the prepared user data. In some embodiments, the system may be configured to provide the extracted features to the trained model 872 to generate results 880. The results 880 can be used, for example, to determine a risk level associated with the user and / or to determine one or more risk subscores for the user.

[0106] In some embodiments, the user data 874, the results 880, and other information about the user (e.g., information about the user's outcomes after receiving or not receiving treatment for a plaque-based disease) can be used to train the model. At block 882, in some embodiments, the system may have the user prepare the user data 874 and the results 880 for use in training. In some embodiments, preparing the data may include, for example, anonymizing the data. For example, in some embodiments, any information regarding the patient's name, social security number, or other information that may be personally identifiable may be removed. In some embodiments, the system may partially anonymize the user data 874 by changing the user's date of birth, for example, retaining only the year the user was born (as age is often an important factor in evaluating a medical history) or the year and month the user was born. In some embodiments, the system may store the prepared data in the training data store 858. In some embodiments, the prepared data may additionally or alternatively be stored in another database or data store. In some embodiments, the system may retrain the model periodically, continuously, or whenever an operator indicates to the system that the model needs to be retrained. Thus, in some embodiments, the trained model 872 can evolve over time, such that risk assessment can improve over time, for example, as the model is trained with additional data.

[0107] In some embodiments, the systems, processes, and methods described herein are implemented using a computer system such as that illustrated in Figure 9. Exemplary computer system 928 is in communication with one or more computing systems 946 and / or one or more data sources 948 via one or more networks 944. While Figure 9 illustrates one embodiment of computer system 928, it will be recognized that the functionality provided in the components and modules of computer system 928 may be combined into fewer components and modules or further separated into additional components and modules.

[0108] The computer system 928 may include a plaque analysis and / or risk assessment module 940 that performs the functions, methods, operations, and / or processes described herein. The plaque analysis and / or risk assessment module 940 is executed on the computer system 928 by a central processing unit 932, which is discussed further below.

[0109] Generally, as used herein, the term "module" refers to logic embodied in hardware or firmware, or a collection of software instructions with entry and exit points. Modules are written in programming languages ​​such as JAVA, C, or C++. Software modules can be compiled or linked into executable programs, installed in dynamic link libraries, or written in interpreted languages ​​such as BASIC, PERL, LAU, PHP, or Python. Software modules can be called by other modules, by themselves, and / or in response to detected events or interrupts. Hardware-implemented modules may contain connected logic units such as gates and flip-flops and / or may include programmable units such as programmable gate arrays or processors.

[0110] Generally, modules described herein refer to logical modules that can be combined with other modules or divided into sub-modules, regardless of physical organization or storage. Modules are executed by one or more computer systems and can be stored on or in any suitable computer-readable medium, or can be implemented in whole or in part in specially designed hardware or firmware. While not all calculations, analyses, and / or optimizations require the use of a computer system, any of the methods, calculations, processes, or analyses described above can be facilitated by the use of a computer. Furthermore, in some embodiments, process blocks described herein can be modified, rearranged, combined, and / or omitted.

[0111] The computer system 928 includes one or more processing units (CPUs) 932, which may be comprised of a microprocessor. The computer system 928 further includes physical memory 936, such as random access memory (RAM) for temporarily storing information and read-only memory (ROM) for persistently storing information, and mass storage device 930, such as a backing store, hard drive, rotating magnetic disk, solid-state disk (SSD), flash memory, phase-change memory (PCM), 3D XPoint memory, diskette, or optical media storage device. Alternatively, the mass storage device may be implemented in an array of servers. Typically, the components of the computer system 928 are connected to the computer using a standards-based bus system. The bus system may be implemented using various protocols, such as Peripheral Component Interconnect (PCI), MicroChannel, SCSI, Industrial Standard Architecture (ISA), and Extended ISA (EISA) architectures.

[0112] The computer system 928 includes one or more input / output (I / O) devices and interfaces 938, such as a keyboard, mouse, touchpad, printer, etc. The I / O devices and interfaces 938 may include one or more display devices, such as a monitor, that enable the visual presentation of data to a user. More particularly, the display devices provide, for example, the presentation of GUIs, such as application software data, and multimedia presentations. The I / O devices and interfaces 938 may also provide communication interfaces to various external devices. The computer system 928 may comprise one or more multimedia devices 934, such as, for example, speakers, a video card, a graphics accelerator, and a microphone.

[0113] Computer system 928 may run on a variety of computing devices, such as a server, a Windows server, a Structure Query Language server, a Unix server, a personal computer, a laptop computer, etc. In other embodiments, computer system 928 may run on a cluster computer system, a mainframe computer system, and / or other computing systems suitable for controlling and / or communicating with large databases, performing high-volume transaction processing, and generating reports from large databases. Computer system 928 is generally controlled and coordinated by operating system software, such as z / OS, Windows, Linux, UNIX, BSD, PHP, SunOS, Solaris, MacOS, iCloud services, or other compatible operating systems, including proprietary operating systems. The operating system controls and schedules the execution of computer processes, performs memory management, provides file system, networking, I / O services, and provides a user interface, such as a graphical user interface (GUI), among other things.

[0114] The computer system 928 shown in FIG. 9 is coupled to a network 944, such as a LAN, a WAN, or the Internet, via a communications link 942 (wired, wireless, or a combination thereof). The network 944 communicates with various computing devices and / or other electronic devices. The network 944 is in communication with one or more computer systems 946 and one or more data sources 948. The plaque analysis and / or risk assessment module 940 can access or be accessed by the computing systems 946 and / or data sources 948 via a web-enabled user access point. The connection can be a direct physical connection, a virtual connection, or other connection types. The web-enabled user access point can consist of a browser module that presents data textually, visually, using audio, video, and other media, and allows interaction with the data over the network 944.

[0115] The output module can be implemented as an all-points-addressable display such as a cathode ray tube (CRT), a liquid crystal display (LCD), a plasma display, or other types and / or combinations of displays. The output module can be implemented to communicate with input devices 938, which also include software with an appropriate interface that allows a user to access data through the use of stylized screen elements such as menus, windows, dialog boxes, toolbars, and controls (e.g., radio buttons, check boxes, sliding scales, etc.). Additionally, the output module can communicate with a series of input / output devices to receive signals from a user.

[0116] Computer system 928 may include one or more internal and / or external data sources (e.g., data source 948). In some embodiments, one or more of the data repositories and data sources described above may be implemented using relational databases, such as DB2, Sybase, Oracle, CodeBase, and Microsoft® SQL Server, as well as other types of databases, such as flat file databases, entity-relationship databases, and object-oriented databases, and / or record-based databases.

[0117] The computer system 928 may also have access to one or more databases 948. The databases 948 may be stored in a database or data repository. The computer system 928 may access the one or more databases 948 over a network 944 or may access the databases or data repositories directly via I / O devices and interfaces 938. The data repository that stores the one or more databases 948 may reside within the computer system 928.

[0118] In some embodiments, including any of the embodiments disclosed herein (above or below), one or more features of the systems, methods, and devices described herein may utilize URLs and / or cookies, for example, to store and / or transmit data or user information. A uniform resource locator (URL) may include a web address and / or a reference to a web resource stored in a database and / or server. A URL may specify the location of a resource on a computer and / or computer network. A URL may include a mechanism for retrieving a network resource. A source of a network resource may receive the URL, locate the web resource, and return the web resource to the requester. A URL may be translated into an IP address, and the Domain Name System (DNS) may look up the URL and its corresponding IP address. A URL may be a reference to a web page, file transfer, email, database access, or other application. A URL may include a path, domain name, file extension, hostname, query, fragment, scheme, protocol identifier, port number, username, password, flags, object, resource name, and / or a sequence of characters that identify these. The systems disclosed herein can generate, receive, send, apply, parse, serialize, render, and / or perform actions on URLs.

[0119] Cookies, also known as HTTP cookies, web cookies, Internet cookies, or browser cookies, can contain data sent from a website and / or stored on a user's computer. This data may be stored by a user's web browser while the user browses. Cookies can contain useful information that allows a website to remember previous browsing information, such as an online store's shopping cart, button clicks, login information, and / or a record of previously visited web pages or network resources. Cookies can also contain information entered by a user, such as name, address, password, or credit card information. Cookies can also perform computer functions. For example, an authentication cookie can be used by an application (e.g., a web browser) to identify whether a user is already logged in (e.g., to a website). Cookie data can be encrypted to provide consumer security. Tracking cookies can be used to compile an individual's past browsing history. The system disclosed herein can generate and use cookies to access personal data. The system can also generate and use JSON Web Tokens to store authenticity information, HTTP authentication as an authentication protocol, IP addresses for tracking session or identity information, URLs, etc.

[0120] Image-based ischemic risk assessment Various embodiments described herein relate to systems, devices, and methods for assessing risk of ischemia based on non-invasive images. In particular, in some embodiments, the systems, devices, and methods described herein relate to facilitating risk assessment of ischemic lesions based at least in part on non-invasive medical image analysis of myocardium subtracted by the ischemic lesion. In some embodiments, the systems, devices, and methods described herein utilize the amount of myocardium subtracted and its location on the arterial tree to determine the risk level. In some embodiments, the systems, devices, and methods described herein determine that the higher the ischemic lesion is located on the arterial tree, the more myocardium the ischemic lesion subtracts. In some embodiments, the systems, devices, and methods described herein determine that the lower the ischemic lesion is located on the arterial tree, the less myocardium the ischemic lesion subtracts. In some embodiments, the systems, devices, and methods described herein determine that the greater the amount of myocardium subtracted, the higher the risk. In some embodiments, the systems, devices, and methods described herein determine that the less myocardium the ischemic lesion is subtracted, the lower the risk. In some embodiments, the systems, devices, and methods described herein generate a graphical representation of the risk level. In some embodiments, the graphical representation is a caricature of the heart. In some embodiments, the graphical representation is a display of the volume of myocardium that is or is not subtracted by a particular ischemic lesion. In some embodiments, the systems, devices, and methods described herein can be repeated for different ischemic lesions and displayed together in a single graphical representation. In some embodiments, a user can select an ischemic lesion, and the portion of the myocardium that is subtracted by that ischemic lesion is highlighted or displayed in a different color. In some embodiments, the systems, devices, and methods described herein can generate a myocardial perfusion map that represents the perfusion of blood through the myocardium that is subtracted by the ischemic lesion.In some embodiments, the systems, devices, and methods described herein can be used to determine the risk of ischemic lesions.

[0121] Generally, ischemia can refer to a condition in which blood flow and / or oxygen is restricted or reduced in a part of the body. For example, myocardial ischemia can refer to a restriction or reduction in blood flow from a coronary artery to the myocardium. Within the coronary arteries, ischemia can exist in one or more coronary arteries and / or in one or more lesions within one or more coronary arteries. While the presence of ischemia anywhere is considered problematic, the location or lesion where ischemia exists can affect the severity or significance of the disease. For example, an ischemic lesion can appear anywhere along the coronary artery tree. However, for two equally ischemic lesions, the ischemic lesion that appears higher in the coronary artery tree can be considered more problematic and / or riskier than the ischemic lesion that appears lower in the coronary artery tree. This may be because the ischemic lesion that appears higher in the coronary artery tree may affect more downstream coronary arteries and / or myocardium than the equally ischemic lesion that appears lower in the coronary artery tree. In other words, an ischemic lesion that drains more branches within the coronary artery tree is likely to affect more myocardial mass than an equally ischemic lesion that drains fewer branches within the coronary artery tree. Therefore, the location of an ischemic lesion within an arterial tree, such as the coronary artery tree, can be important in assessing the risk and / or likelihood that an ischemic lesion will lead to a major adverse event, such as a major adverse cardiovascular event (MACE). For this reason, the location of an ischemic lesion within an arterial tree can have a significant impact in addition to how ischemic the lesion is. However, existing technologies do not determine and / or provide the severity of an ischemic lesion based on its location within the arterial tree and / or the downstream tissues affected by that ischemic lesion. Some embodiments of the systems, methods, and devices herein address this technical shortcoming. For example, in some embodiments, the systems, methods, and devices are configured to determine how much tissue is affected by an ischemic lesion and / or generate a visualization of the ischemic lesion to aid in determining the severity, significance, and / or potential risk of a major adverse event occurring due to the ischemic lesion.In particular, in some embodiments, systems, devices, and methods are configured to analyze one or more coronary arteries to determine the presence of ischemia and / or to determine the myocardium subtracted by a particular ischemic lesion, which can be used to determine a subject's risk of MACE based on the ischemic lesion. In some embodiments, systems, methods, and devices can be configured to generate a visual and / or graphical representation of the myocardium subtracted by the ischemic lesion to provide a clinician and / or subject with a visual overview of the risk associated with the ischemic lesion.

[0122] FIG. 10 is a flowchart illustrating exemplary embodiment(s) of a system, device, or method for non-invasive image-based ischemia risk assessment. As shown in FIG. 10 , in some embodiments, the system may be configured to access and / or modify one or more medical images at block 1002. In some embodiments, the medical images may include one or more arteries, such as the subject's coronary arteries, carotid arteries, and / or other arteries. In some embodiments, the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterior lateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterior lateral branch (R-PLB). In some embodiments, the medical images may be stored in a medical image database 1004. In some embodiments, the medical image database 1004 may be accessed locally by the system and / or may be located remotely and accessed via a network connection. The medical images may include images acquired using one or more modalities, such as CT, dual-energy computed tomography (DECT), spectral CT, photon-counting CT, X-ray, ultrasound, echocardiography, intravascular ultrasound (IVUS), magnetic resonance (MR) imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single-photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS). In some embodiments, the medical images include one or more of contrast-enhanced CT images, non-contrast-enhanced CT images, MR images, and / or images obtained using any of the aforementioned modalities.

[0123] In some embodiments, the system may be configured to automatically and / or dynamically perform one or more analyses of the medical images as discussed herein. For example, in some embodiments, at block 1006, the system may be configured to identify one or more blood vessels, such as one or more arteries. The one or more arteries may include, among others, coronary arteries, carotid arteries, aorta, renal arteries, arteries of the lower limbs, arteries of the upper limbs, and / or cerebral arteries. In some embodiments, the system may be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically identify one or more arteries or coronary arteries using image processing. For example, in some embodiments, one or more AI and / or ML algorithms may be trained using a convolutional neural network (CNN) on a set of medical images in which arteries or coronary arteries have been identified, thereby enabling the AI ​​and / or ML algorithms to automatically identify the arteries or coronary arteries directly from the medical images. In some embodiments, the arteries or coronary arteries are identified by size and / or location.

[0124] In some embodiments, at block 1008, the system may be configured to identify one or more regions of plaque within the medical image. In some embodiments, the system may be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically identify one or more regions of plaque using image processing. For example, in some embodiments, one or more AI and / or ML algorithms may be trained using a convolutional neural network (CNN) on a set of medical images in which regions of plaque have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify regions of plaque directly from the medical images. In some embodiments, the system is configured to identify the vessel wall and the lumen wall and classify everything between the vessel wall and the lumen wall as plaque.

[0125] In some embodiments, at block 1010, the system may be configured to analyze and / or characterize one or more regions of plaque based on density. For example, in some embodiments, the system may be configured to analyze and / or characterize one or more regions of plaque based on absolute density and / or relative density and / or radiometric density. In some embodiments, the volume of one or more regions of plaque is determined based at least in part on analyzing the density of one or more pixels corresponding to one or more regions of plaque within the medical image. In some embodiments, low-density non-calcified plaque corresponds to one or more pixels having a radiometric density value between about -189 and about 30 Hounsfield units. In some embodiments, non-calcified plaque corresponds to one or more pixels having a radiometric density value between about 190 and about 350 Hounsfield units. Calcified plaque corresponds to one or more pixels having a radiometric density value between about 351 and 2500 Hounsfield units. In some embodiments, the system may be configured to classify a region of plaque as one of low-density non-calcified plaque, non-calcified plaque, and calcified plaque using any one or more of the processes and / or features described herein.

[0126] In some embodiments, at block 1012, the system may be configured to analyze and / or characterize one or more regions of plaque based on one or more distances. For example, as described herein, in some embodiments, the system may be configured to determine the distance between low-density, non-calcified plaque and the lumen wall and / or vessel wall. In some embodiments, the proximity of low-density, non-calcified plaque to the lumen wall may indicate high-risk plaque and / or CAD. Conversely, in some embodiments, the location of low-density, non-calcified plaque farther from the lumen wall may indicate a lower risk. In some embodiments, the system may be configured to utilize one or more predetermined thresholds in determining risk factors associated with the proximity of low-density, non-calcified plaque to the vessel wall and / or lumen wall. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine one or more distances to and / or from one or more regions of plaque.

[0127] In some embodiments, at block 1012, the system may be configured to analyze and / or characterize one or more regions of the low-density non-calcified plaque based on the morphology or shape and / or one or more axial measurements of the plaque. As described herein, in some embodiments, the system may be configured to determine the length of one or more axes of the low-density non-calcified plaque, such as, for example, the long axis of a longitudinal section and / or the long and / or short axes of a transverse section of the low-density non-calcified plaque. In some embodiments, the system may be configured to determine the morphology and / or shape of the low-density non-calcified plaque using the one or more axial measurements. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine one or more axial measurements of one or more regions of the plaque.

[0128] In some embodiments, the system can be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically classify the shape of one or more regions of plaque using image processing. For example, in some embodiments, one or more AI and / or ML algorithms can be trained using a convolutional neural network (CNN) on a set of medical images in which the shapes of regions of plaque have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify the shape or morphology of regions of plaque directly from the medical images. In some embodiments, the system can be configured to classify the shape or morphology of regions of plaque as one or more of crescent, lobe, round, or bean-shaped. In some embodiments, round and / or bean-shaped plaques can be associated with a high risk, while crescent and / or lobe-shaped plaques can be associated with a low risk of CAD.

[0129] In some embodiments, at block 1012, the system may be configured to analyze and / or characterize one or more regions of plaque based on one or more sizes and / or volumes. For example, in some embodiments, the system may be configured to determine the size and / or volume of plaque based at least in part on measurements of one or more axes described herein. In some embodiments, the system may be configured to determine the size and / or volume of a region of plaque directly from analysis of a three-dimensional image scan. In some embodiments, the system may be configured to determine the size and / or volume of total plaque, low-density non-calcified plaque, non-calcified plaque, calcified plaque, and / or the ratio between two of the aforementioned volumes or sizes. In some embodiments, a high total plaque volume and / or a high low-density non-calcified plaque and / or non-calcified plaque volume may be associated with a high risk of CAD. In some embodiments, a high ratio of low-density non-calcified plaque volume to total plaque volume and / or a high ratio of non-calcified plaque volume to total plaque volume may be associated with a high risk of CAD. In some embodiments, a high calcified plaque volume and / or a high ratio of calcified plaque volume to total plaque volume may be associated with a low risk of CAD. In some embodiments, the system may be configured to utilize one or more predetermined thresholds to determine the risk of CAD based on plaque volume, size, or one or more ratios thereof. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine the size and / or volume of one or more regions of plaque.

[0130] In some embodiments, at block 1012, the system may be configured to analyze and / or characterize plaque based on its degree of embedment. For example, in some embodiments, the system may be configured to determine the extent to which low-density non-calcified plaque is embedded or surrounded by non-calcified plaque or calcified plaque. In some embodiments, the system may be configured to analyze the embedment of low-density non-calcified plaque based on the degree to which the low-density non-calcified plaque is surrounded by other types of plaque. In some embodiments, a higher degree of embedment of low-density non-calcified plaque may indicate a higher risk of CAD. For example, in some embodiments, low-density non-calcified plaque surrounded by 270 degrees or more of non-calcified plaque may be associated with a higher risk of CAD. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine the embedment degree of one or more regions of plaque.

[0131] In some embodiments, at block 1014, the system may be configured to determine ischemic lesions in the plurality of vessels based at least in part on a plurality of plaque parameters. In some embodiments, the plurality of plaque parameters include one or more of lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, post-stenosis baseline diameter, pre-stenosis baseline diameter, vessel length, lumen volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or low-density plaque volume. In some embodiments, the ischemic lesions in the plurality of vessels are determined by a machine learning algorithm. In some embodiments, the machine learning algorithm is trained at least in part on a dataset including the plurality of plaque parameters and the presence of ischemia derived using invasive fractional flow reserve. In some embodiments, the machine learning algorithm is trained based at least in part on a dataset including a plurality of plaque parameters and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0132] In some embodiments, at block 1016, the system may be configured to determine, based at least in part on the mapping of the plurality of vessels, myocardium that is subtracted by the ischemic lesion and myocardium that is not subtracted by the ischemic lesion. In some embodiments, if the ischemic lesion appears higher in the arterial tree, the system may determine that the vessel containing the ischemic lesion subtracts more myocardium and therefore is associated with a higher risk. In some embodiments, if the ischemic lesion appears lower in the arterial tree, the system may determine that the vessel containing the ischemic lesion subtracts less myocardium and therefore is associated with a lower risk. In some embodiments, an ischemic lesion higher in the arterial tree means that the ischemic lesion drains more branches of the arterial tree. In some embodiments, an ischemic lesion lower in the arterial tree means that the ischemic lesion drains fewer branches of the arterial tree. In some embodiments, the system may be configured to generate an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular event (MACE) based at least in part on the myocardium that is subtracted by the ischemic lesion. In some embodiments, the system may be configured to generate a graphical representation of the assessment of risk of CAD or MACE. In some embodiments, the system may be configured to generate a recommended treatment for the subject based at least in part on the assessment of risk of CAD or MACE.

[0133] In some embodiments, at block 1018, the system may be configured to generate a graphical visualization of myocardium subtracted by ischemic lesions and / or myocardium not subtracted by ischemic lesions. In some embodiments, the graphical visualization includes a graphical representation of myocardium. In some embodiments, the graphical visualization includes a representation of a volume of myocardium subtracted by ischemic lesions. In some embodiments, the graphical visualization includes a representation of a volume of myocardium subtracted by ischemic lesions and a representation of a volume of myocardium not subtracted by ischemic lesions. In some embodiments, the graphical visualization includes a schematic image of myocardium subtracted by ischemic lesions. In some embodiments, the graphical visualization includes a schematic image of myocardium subtracted by ischemic lesions and myocardium not subtracted by ischemic lesions. In some embodiments, the graphical visualization includes a schematic image of the heart. In some embodiments, the graphical visualization includes a display of the volume of myocardium subtracted or not subtracted by a particular ischemic lesion. In some embodiments, the system of the present invention can be repeated for different ischemic lesions and displayed together in a single graphical representation. In some embodiments, the user can select one ischemic lesion and the portion of the myocardium subtracted by that ischemic lesion will be highlighted or displayed in a different color.

[0134] In some embodiments, at block 1020, the system may be configured to generate a myocardial perfusion map representing the perfusion of blood through the myocardium subtracted by the ischemic lesion. In some embodiments, the system may be configured to generate an overlay of a graphical visualization of the myocardium subtracted by the ischemic lesion and the myocardial perfusion map. In some embodiments, the myocardial perfusion map is configured to be used to determine the presence of a perfusion defect in the myocardium subtracted by the ischemic lesion. In some embodiments, a perfusion defect that appears smaller than the myocardium subtracted by the ischemic lesion indicates collateral vessels supplying blood to the myocardium subtracted by the ischemic lesion. In some embodiments, a perfusion defect that appears larger than the myocardium subtracted by the ischemic lesion indicates other diseases.

[0135] In some embodiments, at block 1022, the system may be configured to determine the subject's risk of an ischemic lesion. In some embodiments, a graphical visualization of myocardium subtracted by ischemic lesions compared to myocardium not subtracted by ischemic lesions is configured to be utilized to determine the subject's risk of an ischemic lesion. In some embodiments, a greater amount of myocardium subtracted by ischemic lesions indicates a higher risk compared to a smaller amount of myocardium subtracted by ischemic lesions. In some embodiments, a greater amount of myocardium subtracted in vessels higher in the vascular tree poses a higher risk. In some embodiments, the risk determination is presented as a displayed number. In some embodiments, the graphical visualization includes a schematic image of the heart. In some embodiments, the graphical visualization includes a display of the volume of myocardium subtracted or not subtracted by a particular ischemic lesion. In some embodiments, the system can be cycled for different ischemic lesions and displayed together in a single graphical representation. In some embodiments, a user can select one ischemic lesion, and the portion of myocardium subtracted by that ischemic lesion is highlighted or displayed in a different color.

[0136] The computer system 902 of FIG. 9, and possibly the analysis and / or risk assessment module 940, may be configured to perform the functions, methods, operations, and / or processes for image-based ischemic risk assessment described herein, such as those described above with reference to FIG. 10.

[0137] The following are non-limiting examples of specific embodiments of systems and methods for image-based ischemic risk assessment. Other embodiments may include one or more other or different features discussed herein.

[0138] Embodiment 1: A computer-implemented method for facilitating risk assessment of ischemic lesions based at least in part on non-invasive medical image analysis of myocardium subtracted by ischemic lesions, comprising: accessing, by a computer system, medical images of a subject, wherein the medical images of the subject are non-invasively acquired; analyzing, by the computer system, the medical images of the subject to map a plurality of blood vessels comprising one or more regions of plaque; identifying, by the computer system, one or more regions of plaque in the plurality of blood vessels; analyzing, by the computer system, the one or more regions of plaque to generate a plurality of plaque parameters comprising a density and a volume of the one or more regions of plaque; and, based at least in part on the plurality of plaque parameters, determining, by a computer system, myocardium subtracted by the ischemic lesions and myocardium not subtracted by the ischemic lesions based at least in part on the mapping of the plurality of blood vessels; and generating, by the computer system, a graphical visualization of the myocardium subtracted by the ischemic lesions and myocardium not subtracted by the ischemic lesions, wherein the graphical visualization of the myocardium subtracted by the ischemic lesions compared to myocardium not subtracted by the ischemic lesions is configured to be utilized to determine the subject's risk of ischemic lesions, wherein a greater amount of myocardium subtracted by the ischemic lesions indicates a higher risk compared to a lesser amount of myocardium subtracted by the ischemic lesions; and wherein the computer system comprises a computer processor and an electronic storage medium.

[0139] Embodiment 2: The computer-implemented method of embodiment 1, wherein the graphical visualization comprises a graphical representation of the myocardium.

[0140] Embodiment 3: The computer-implemented method of embodiment 1, wherein the graphical visualization includes a representation of the volume of myocardium subtracted by the ischemic lesion.

[0141] Embodiment 4: The computer-implemented method of embodiment 1, wherein the graphical visualization includes a representation of the volume of myocardium that is subtracted by the ischemic lesion and a representation of the volume of myocardium that is not subtracted by the ischemic lesion.

[0142] Embodiment 5: The computer-implemented method of embodiment 1, wherein the graphical visualization includes a schematic image of the myocardium subtracted by the ischemic lesion.

[0143] Embodiment 6: The computer-implemented method of embodiment 1, wherein the graphical visualization includes a schematic representation of myocardium subtracted by ischemic lesions and myocardium not subtracted by ischemic lesions.

[0144] Embodiment 7: The computer-implemented method of embodiment 1, wherein the plurality of plaque parameters includes stenosis.

[0145] Embodiment 8: The computer-implemented method of embodiment 1, wherein the volume of the one or more regions of plaque comprises one or more of the volume of total plaque, the volume of low-density non-calcified plaque, the volume of non-calcified plaque, or the volume of calcified plaque.

[0146] Embodiment 9: The computer-implemented method of embodiment 8, wherein the volume of one or more regions of plaque is determined based at least in part on analyzing the density of one or more pixels corresponding to the one or more regions of plaque in the medical image.

[0147] Embodiment 10: The computer-implemented method of embodiment 9, wherein the density comprises a material density.

[0148] Embodiment 11: The computer-implemented method of embodiment 9, wherein the density includes radial density.

[0149] Embodiment 12: The computer-implemented method of embodiment 11, wherein low-density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about -189 and about 30 Hounsfield units.

[0150] Embodiment 13: The computer-implemented method of embodiment 11, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 190 and about 350 Hounsfield units.

[0151] Embodiment 14: The computer-implemented method of embodiment 11, wherein the calcified plaque corresponds to one or more pixels having a radiodensity value between approximately 351 and 2500 Hounsfield units.

[0152] Embodiment 15: The computer-implemented method of embodiment 1, wherein the medical image comprises a computed tomography (CT) image.

[0153] Embodiment 16: The computer-implemented method of embodiment 1, wherein the medical image is obtained using imaging techniques including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0154] Embodiment 17: The computer-implemented method of embodiment 1, wherein the plurality of plaque parameters include one or more of: lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, post-stenosis baseline diameter, pre-stenosis baseline diameter, vessel length, lumen volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or volume of low-density plaque.

[0155] Embodiment 18: The computer-implemented method of embodiment 1, wherein the plurality of blood vessels includes one or more coronary arteries.

[0156] Embodiment 19: The computer-implemented method of embodiment 18, wherein the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0157] Embodiment 20: The computer-implemented method of embodiment 1, wherein the ischemic lesions in the plurality of vessels are determined by a machine learning algorithm.

[0158] Embodiment 21: The computer-implemented method of embodiment 20, wherein the machine learning algorithm is trained based at least in part on a dataset comprising a plurality of plaque parameters and the presence of ischemia derived using invasive fractional flow reserve.

[0159] Embodiment 22: The computer-implemented method of embodiment 20, wherein the machine learning algorithm is trained based at least in part on a dataset comprising a plurality of plaque parameters and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0160] Embodiment 23: The computer-implemented method of embodiment 1, further comprising generating, by the computer system, an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular events (MACE) based at least in part on the myocardium subtracted by ischemic lesions.

[0161] Embodiment 24: The computer-implemented method of embodiment 23, further comprising generating, by the computer system, a graphical representation of the assessment of the risk of CAD or MACE.

[0162] Embodiment 25: The computer-implemented method of embodiment 23, further comprising generating, by the computer system, a recommended treatment for the subject based at least in part on the assessment of the risk of CAD or MACE.

[0163] Embodiment 26: The computer-implemented method of embodiment 1, further comprising generating, by the computer system, a myocardial perfusion map representing the perfusion of blood through the myocardium subtracted by ischemic lesions.

[0164] Embodiment 27: The computer-implemented method of embodiment 26, further comprising generating, by the computer system, an overlay of a graphical visualization of the myocardium subtracted by the ischemic lesion and a myocardial perfusion map.

[0165] Embodiment 28: The computer-implemented method of embodiment 26, wherein the myocardial perfusion map is configured to be used to determine the presence of perfusion defects in the myocardium that are subtracted by ischemic lesions.

[0166] Embodiment 29: The computer-implemented method of embodiment 28, wherein a perfusion defect that appears smaller than the myocardium subtracted by the ischemic lesion indicates a collateral vessel that supplies blood to the myocardium subtracted by the ischemic lesion.

[0167] Embodiment 30: The computer-implemented method of embodiment 28, wherein perfusion defects that appear larger than the myocardium subtracted by ischemic lesions indicate other diseases.

[0168] Embodiment 31: A non-transitory computer-readable medium configured for facilitating risk assessment of ischemic lesions based at least in part on non-invasive medical image analysis of myocardium subtracted by ischemic lesions, the computer-readable medium having program instructions for causing a hardware processor to execute a method, the method comprising: accessing, by a computer system, medical images of a subject, wherein the medical images of the subject are non-invasively acquired; analyzing, by the computer system, the medical images of the subject to map a plurality of blood vessels containing one or more regions of plaque; identifying, by the computer system, one or more regions of plaque in the plurality of blood vessels; analyzing, by the computer system, the one or more regions of plaque to generate a plurality of plaque parameters including a density and a volume of the one or more regions of plaque; and determining, by a computer system, myocardium subtracted by the ischemic lesions and myocardium not subtracted by the ischemic lesions based at least in part on the mapping of the plurality of vessels; and generating, by the computer system, a graphical visualization of the myocardium subtracted by the ischemic lesions and myocardium not subtracted by the ischemic lesions, wherein the graphical visualization of the myocardium subtracted by the ischemic lesions compared to the myocardium not subtracted by the ischemic lesions is configured to be utilized to determine the subject's risk of ischemic lesions, wherein a greater amount of myocardium subtracted by the ischemic lesions indicates a higher risk compared to a lesser amount of myocardium subtracted by the ischemic lesions, and the computer system comprises a computer processor and an electronic storage medium.

[0169] Embodiment 32: A non-transitory computer-readable medium configured as in embodiment 31, wherein the graphical visualization includes a graphical representation of the myocardium.

[0170] Embodiment 33: A non-transitory computer-readable medium configured as in embodiment 31, wherein the graphical visualization includes a representation of the volume of myocardium subtracted by the ischemic lesion.

[0171] Embodiment 34: A non-transitory computer-readable medium configured as in embodiment 31, wherein the graphical visualization includes a representation of a volume of myocardium subtracted by ischemic lesions and a representation of a volume of myocardium not subtracted by ischemic lesions.

[0172] Embodiment 35: A non-transitory computer-readable medium configured as in embodiment 31, wherein the graphical visualization includes a schematic image of the myocardium subtracted by ischemic lesions.

[0173] Embodiment 36: A non-transitory computer-readable medium configured as in embodiment 31, wherein the graphical visualization includes a schematic representation of myocardium subtracted by ischemic lesions and myocardium not subtracted by ischemic lesions.

[0174] Embodiment 37: A non-transitory computer-readable medium configured as in embodiment 31, wherein the plurality of plaque parameters includes stenosis.

[0175] Embodiment 38: A non-transitory computer-readable medium configured as in embodiment 31, wherein the volume of one or more regions of plaque comprises one or more of a volume of total plaque, a volume of low-density non-calcified plaque, a volume of non-calcified plaque, or a volume of calcified plaque.

[0176] Embodiment 39: A non-transitory computer-readable medium configured as in embodiment 38, wherein the volume of one or more regions of plaque is determined based at least in part on analyzing the density of one or more pixels corresponding to the one or more regions of plaque in a medical image.

[0177] Embodiment 40: A non-transitory computer-readable medium configured as in embodiment 39, wherein the density comprises a material density.

[0178] Embodiment 41: A non-transitory computer-readable medium configured as in embodiment 39, wherein the density includes a radial density.

[0179] Embodiment 42: A non-transitory computer-readable medium configured as in embodiment 41, wherein the low-density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about -189 and about 30 Hounsfield units.

[0180] Embodiment 43: A non-transitory computer-readable medium configured as in embodiment 41, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 190 and about 350 Hounsfield units.

[0181] Embodiment 44: A non-transitory computer-readable medium configured as in embodiment 41, wherein the calcified plaque corresponds to one or more pixels having a radiodensity value between approximately 351 and 2500 Hounsfield units.

[0182] Embodiment 45: A non-transitory computer-readable medium configured as in embodiment 31, wherein the medical images include computed tomography (CT) images.

[0183] Embodiment 46: A non-transitory computer-readable medium configured as in embodiment 31, wherein the medical image is obtained using imaging techniques including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0184] Embodiment 47: A non-transitory computer-readable medium configured as in embodiment 31, wherein the plurality of plaque parameters include one or more of lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, post-stenosis baseline diameter, pre-stenosis baseline diameter, vessel length, lumen volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or volume of low-density plaque.

[0185] Embodiment 48: A non-transitory computer-readable medium configured as in embodiment 31, wherein the plurality of blood vessels includes one or more coronary arteries.

[0186] Embodiment 49: A non-transitory computer-readable medium configured as in embodiment 48, wherein the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0187] Embodiment 50: A non-transient computer-readable medium configured as in embodiment 31, wherein ischemic lesions in the plurality of vessels are determined by a machine learning algorithm.

[0188] Embodiment 51: A non-transitory computer-readable medium configured as in embodiment 50, wherein the machine learning algorithm is trained based at least in part on a dataset including a plurality of plaque parameters and the presence of ischemia derived using invasive fractional flow reserve.

[0189] Embodiment 52: A non-transient computer-readable medium configured as in embodiment 50, wherein the machine learning algorithm is trained based at least in part on a dataset including a plurality of plaque parameters and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0190] Embodiment 53: A non-transitory computer-readable medium configured as in embodiment 31, further comprising generating, by the computer system, an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular events (MACE) based at least in part on myocardium subtracted by ischemic lesions.

[0191] Embodiment 54: A non-transitory computer-readable medium configured as in embodiment 53, further comprising generating, by a computer system, a graphical representation of the assessment of the risk of CAD or MACE.

[0192] Embodiment 55: A non-transitory computer-readable medium configured as in embodiment 53, further comprising generating, by the computer system, a recommended treatment for the subject based at least in part on the assessment of the risk of CAD or MACE.

[0193] Embodiment 56: A non-transitory computer-readable medium configured as in embodiment 31, further comprising generating, by the computer system, a myocardial perfusion map representing the perfusion of blood through the myocardium subtracted by ischemic lesions.

[0194] Embodiment 57: A non-transitory computer-readable medium configured as in embodiment 56, further comprising generating, by the computer system, an overlay of a graphical visualization of the myocardium subtracted by ischemic lesions and a myocardial perfusion map.

[0195] Embodiment 58: A non-transitory computer-readable medium configured as in embodiment 56, wherein the myocardial perfusion map is configured to be used to determine the presence of perfusion defects in the myocardium that are subtracted by ischemic lesions.

[0196] Embodiment 59: A non-transient computer-readable medium configured as in embodiment 58, wherein a perfusion defect that appears smaller than the myocardium subtracted by the ischemic lesion indicates collateral vessels that supply blood to the myocardium subtracted by the ischemic lesion.

[0197] Embodiment 60: A non-transient computer-readable medium configured as in embodiment 58, wherein perfusion defects appearing larger than the myocardium subtracted by ischemic lesions indicate other diseases.

[0198] Embodiment 61: A system comprising: accessing, by a computer system, medical images of a subject, wherein the medical images of the subject are non-invasively acquired; analyzing, by the computer system, the medical images of the subject to map a plurality of blood vessels including one or more regions of plaque; identifying, by the computer system, one or more regions of plaque in the plurality of blood vessels; analyzing, by the computer system, the one or more regions of plaque to generate a plurality of plaque parameters including a density and a volume of the one or more regions of plaque; determining, by the computer system, ischemic lesions in the plurality of blood vessels based at least in part on the plurality of plaque parameters; and determining, based at least in part on mapping of a plurality of blood vessels, myocardium that is subtracted by ischemic lesions and myocardium that is not subtracted by ischemic lesions; and generating, by a computer system, a graphical visualization of the myocardium that is subtracted by ischemic lesions and myocardium that is not subtracted by ischemic lesions, wherein the graphical visualization of the myocardium that is subtracted by ischemic lesions compared to myocardium that is not subtracted by ischemic lesions is configured to be utilized to determine the subject's risk of ischemic lesions, wherein a greater amount of myocardium that is subtracted by ischemic lesions indicates a higher risk compared to a smaller amount of myocardium that is subtracted by ischemic lesions, and the computer system comprises a computer processor and an electronic storage medium.

[0199] Embodiment 62: The system described in embodiment 61, wherein the graphical visualization includes a graphical representation of the myocardium.

[0200] Embodiment 63: The system described in embodiment 61, wherein the graphical visualization includes a representation of the volume of myocardium subtracted by the ischemic lesion.

[0201] Embodiment 64: The system described in embodiment 61, wherein the graphical visualization includes a representation of the volume of myocardium subtracted by the ischemic lesion and a representation of the volume of myocardium not subtracted by the ischemic lesion.

[0202] Embodiment 65: The system described in embodiment 61, wherein the graphical visualization includes a schematic image of the myocardium subtracted by ischemic lesions.

[0203] Embodiment 66: The system described in embodiment 61, wherein the graphical visualization includes a schematic image of myocardium subtracted by ischemic lesions and myocardium not subtracted by ischemic lesions.

[0204] Embodiment 67: The system described in embodiment 61, wherein the plurality of plaque parameters includes stenosis.

[0205] Embodiment 68: The system of embodiment 61, wherein the volume of one or more regions of plaque comprises one or more of the volume of total plaque, the volume of low-density non-calcified plaque, the volume of non-calcified plaque, or the volume of calcified plaque.

[0206] Embodiment 69: The system described in embodiment 68, wherein the volume of one or more regions of plaque is determined at least in part based on analyzing the density of one or more pixels corresponding to one or more regions of plaque in a medical image.

[0207] Embodiment 70: The system of embodiment 69, wherein the density comprises a material density.

[0208] Embodiment 71: The system described in embodiment 69, wherein the density includes a radial density.

[0209] Embodiment 72: The system of embodiment 71, wherein the low density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about -189 and about 30 Hounsfield units.

[0210] Embodiment 73: The system of embodiment 71, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 190 and about 350 Hounsfield units.

[0211] Embodiment 74: The system described in embodiment 71, wherein the calcified plaque corresponds to one or more pixels having a radiodensity value between approximately 351 and 2500 Hounsfield units.

[0212] Embodiment 75: The system described in embodiment 61, wherein the medical image includes a computed tomography (CT) image.

[0213] Embodiment 76: The system described in embodiment 61, wherein the medical image is obtained using imaging techniques including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0214] Embodiment 77: The system of embodiment 61, wherein the plurality of plaque parameters include one or more of lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, baseline diameter after stenosis, baseline diameter before stenosis, vessel length, lumen volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or volume of low-density plaque.

[0215] Embodiment 78: The system described in embodiment 61, wherein the plurality of blood vessels includes one or more coronary arteries.

[0216] Embodiment 79: The system of embodiment 78, wherein the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0217] Embodiment 80: The system described in embodiment 61, wherein the ischemic lesions in multiple blood vessels are determined by a machine learning algorithm.

[0218] Embodiment 81: The system described in embodiment 80, wherein the machine learning algorithm is trained based at least in part on a dataset including multiple plaque parameters and the presence of ischemia derived using invasive fractional flow reserve.

[0219] Embodiment 82: The system described in embodiment 80, wherein the machine learning algorithm is trained based at least in part on a dataset including multiple plaque parameters and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0220] Embodiment 83: The system of embodiment 61, further comprising generating, by the computer system, an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular events (MACE) based at least in part on the myocardium subtracted by ischemic lesions.

[0221] Embodiment 84: The system of embodiment 83, further comprising generating, by the computer system, a graphical representation of the assessment of the risk of CAD or MACE.

[0222] Embodiment 85: The system of embodiment 83, further comprising generating, by the computer system, a recommended treatment for the subject based at least in part on the assessment of the risk of CAD or MACE.

[0223] Embodiment 86: The system of embodiment 61, further comprising generating, by the computer system, a myocardial perfusion map representing the perfusion of blood through the myocardium subtracted by ischemic lesions.

[0224] Embodiment 87: The system described in embodiment 86, further comprising generating, by the computer system, an overlay of a graphical visualization of the myocardium subtracted by the ischemic lesions and a myocardial perfusion map.

[0225] Embodiment 88: The system described in embodiment 86, wherein the myocardial perfusion map is configured to be used to determine the presence of perfusion defects in the myocardium subtracted by ischemic lesions.

[0226] Embodiment 89: The system of embodiment 88, wherein a perfusion defect that appears smaller than the myocardium subtracted by the ischemic lesion indicates collateral vessels that supply blood to the myocardium subtracted by the ischemic lesion.

[0227] Embodiment 90: The system of embodiment 88, wherein perfusion defects that appear larger than the myocardium subtracted by the ischemic lesion indicate other diseases.

[0228] Determination of ischemia based on image-based stenosis analysis Disclosed herein are systems, devices, and methods for determining ischemia based on image-based stenosis analysis. In particular, in some embodiments, the systems, devices, and methods described herein involve determining the severity of a stenosis and using that data to determine the likelihood of ischemia. The percentage of stenosis, or the percentage to which a blood vessel is narrowed, can correlate to the likelihood of ischemia or blood flow restriction. In some embodiments, the percentage of stenosis is determined by measuring plaque within the blood vessel. In some embodiments, the systems, devices, and methods described herein determine the percentage of multiple stenoses (i.e., two or more stenoses) within a blood vessel to determine the likelihood of ischemia. In some embodiments, the likelihood of ischemia is used to determine whether to measure a patient's fractional flow reserve (FFR).

[0229] 11 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for determining ischemia based on image-based analysis of stenosis. As shown in FIG. 11 , in some embodiments, the system may be configured to access and / or modify one or more first medical images at block 1102. In some embodiments, the first medical images may include one or more arteries, such as coronary arteries, carotid arteries, and / or other arteries of the subject. In some embodiments, the first medical images may be stored in a medical image database 1104. In some embodiments, the medical image database 1104 may be accessed locally by the system and / or may be located remotely and accessed via a network connection. The first medical image can include images acquired using one or more modalities, such as CT, dual-energy computed tomography (DECT), spectral CT, photon-counting CT, X-ray, ultrasound, echocardiography, intravascular ultrasound (IVUS), magnetic resonance (MR) imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single-photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS). In some embodiments, the first medical image includes one or more of a contrast-enhanced CT image, a non-contrast-enhanced CT image, an MR image, and / or an image obtained using any of the aforementioned modalities. In some embodiments, the first medical image(s) can be modified using one or more image processing techniques, for example, to enhance image quality, alter contrast, and / or identify regions of interest.

[0230] In some embodiments, the system may be configured to automatically and / or dynamically perform one or more analyses of the first medical image as discussed herein. For example, in some embodiments, at block 1106, the system may be configured to identify one or more blood vessels, such as one or more arteries. The one or more arteries may include, among others, coronary arteries, carotid arteries, aorta, renal arteries, arteries of the lower limbs, arteries of the upper limbs, and / or cerebral arteries. In some embodiments, the system may be configured to utilize image segmentation to identify coronary artery regions. In some embodiments, the system may be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically identify one or more arteries or coronary arteries using image processing. For example, in some embodiments, one or more AI and / or ML algorithms may be trained using a convolutional neural network (CNN) on a set of medical images in which arteries or coronary arteries have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify arteries or coronary arteries directly from the medical images. In some embodiments, the arteries or coronary arteries are identified by size and / or location.

[0231] In some embodiments, at block 1108, the system may be configured to identify one or more regions of plaque in the first medical image. In some embodiments, the system may be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically identify one or more regions of plaque using image processing. For example, in some embodiments, the one or more AI and / or ML algorithms may be trained using a convolutional neural network (CNN) on a set of medical images in which regions of plaque have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify regions of plaque directly from the medical images. In some embodiments, the system is configured to identify the vessel wall and the lumen wall and classify everything between the vessel wall and the lumen wall as plaque.

[0232] In some embodiments, at block 1108, the system may be configured to analyze and / or characterize one or more regions of plaque based on density. For example, in some embodiments, the system may be configured to analyze and / or characterize one or more regions of plaque based on absolute density and / or relative density and / or radiometric density and / or material density. In some embodiments, the system may be configured to classify regions of plaque as one of low-density non-calcified plaque, non-calcified plaque, and calcified plaque using any one or more processes and / or features described herein.

[0233] In some embodiments, at block 1108, the system may be configured to analyze and / or characterize one or more regions of plaque based on one or more distances. For example, as described herein, in some embodiments, the system may be configured to determine the distance between low-density non-calcified plaque and the lumen wall and / or vessel wall. In some embodiments, the proximity of low-density non-calcified plaque to the lumen wall may indicate a high-risk plaque and / or CAD. Conversely, in some embodiments, the location of low-density non-calcified plaque farther from the lumen wall may indicate a lower risk. In some embodiments, the system may be configured to utilize one or more predetermined thresholds in determining risk factors associated with the proximity of low-density non-calcified plaque to the vessel wall and / or lumen wall. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine one or more distances to and / or from one or more regions of plaque. In some embodiments, the region of plaque is identified based at least in part on the density of one or more pixels in the first medical image.

[0234] In some embodiments, at block 1108, the system may be configured to analyze and / or characterize one or more regions of the low-density non-calcified plaque based on the morphology or shape and / or one or more axial measurements of the plaque. As described herein, in some embodiments, the system may be configured to determine the length of one or more axes of the low-density non-calcified plaque, such as, for example, the long axis of a longitudinal section and / or the long and / or short axes of a transverse section of the low-density non-calcified plaque. In some embodiments, the system may be configured to determine the morphology and / or shape of the low-density non-calcified plaque using the one or more axial measurements. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine the one or more axial measurements of one or more regions of the plaque.

[0235] In some embodiments, the system can be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically classify the shape of one or more regions of plaque using image processing. For example, in some embodiments, one or more AI and / or ML algorithms can be trained using a convolutional neural network (CNN) on a set of medical images in which the shapes of regions of plaque have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify the shape or morphology of regions of plaque directly from the medical images. In some embodiments, the system can be configured to classify the shape or morphology of regions of plaque as one or more of crescent, lobe, round, or bean-shaped. In some embodiments, round and / or bean-shaped plaques can be associated with a high risk, while crescent and / or lobe-shaped plaques can be associated with a low risk of CAD.

[0236] In some embodiments, at block 1108, the system may be configured to analyze and / or characterize one or more regions of plaque based on one or more sizes and / or volumes. For example, in some embodiments, the system may be configured to determine plaque size and / or volume based at least in part on one or more axial measurements described herein. In some embodiments, the system may be configured to determine the size and / or volume of a region of plaque directly from analysis of a three-dimensional image scan. In some embodiments, the system may be configured to determine the size and / or volume of total plaque, low-density non-calcified plaque, non-calcified plaque, calcified plaque, and / or the ratio between two of the aforementioned volumes or sizes. In some embodiments, a high total plaque volume and / or a high low-density non-calcified plaque and / or non-calcified plaque volume may be associated with a high risk of CAD. In some embodiments, a high ratio of low-density non-calcified plaque volume to total plaque volume and / or a high ratio of non-calcified plaque volume to total plaque volume may be associated with a high risk of CAD. In some embodiments, a high calcified plaque volume and / or a high ratio of calcified plaque volume to total plaque volume may be associated with a low risk of CAD. In some embodiments, the system may be configured to utilize one or more predetermined thresholds to determine the risk of CAD based on plaque volume, size, or one or more ratios thereof. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine the size and / or volume of one or more regions of plaque.

[0237] In some embodiments, at block 1108, the system may be configured to analyze and / or characterize plaque based on its degree of embedment. For example, in some embodiments, the system may be configured to determine the extent to which low-density non-calcified plaque is embedded or surrounded by non-calcified plaque or calcified plaque. In some embodiments, the system may be configured to analyze the embedment of low-density non-calcified plaque based on the degree to which the low-density non-calcified plaque is surrounded by other types of plaque. In some embodiments, a higher degree of embedment of low-density non-calcified plaque may indicate a higher risk of CAD. For example, in some embodiments, low-density non-calcified plaque surrounded by 270 degrees or more of non-calcified plaque may be associated with a higher risk of CAD. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine the embedment degree of one or more regions of plaque.

[0238] In some embodiments, at block 1110, the system may be configured to determine a percentage of one or more stenoses present in a region of the coronary artery resulting from one or more regions of plaque. In some embodiments, the percentage of the one or more stenoses is determined based at least in part on interpolating the luminal volume or diameter of the region of the coronary artery in the absence of the one or more regions of plaque. In some embodiments, the percentage of the one or more stenoses is determined by determining dimensions of the coronary artery in a first medical image. In some embodiments, the percentage of the one or more stenoses is determined by comparing dimensions of the coronary artery in the first medical image with dimensions of a reference coronary artery. In some embodiments, the reference coronary artery is taken from the patient at a different time. In some embodiments, the reference coronary artery is taken from a reference database.

[0239] In some embodiments, at block 1112, the system may be configured to determine a probability threshold of ischemia for the identified region of the coronary artery. In some embodiments, the probability threshold of ischemia for the region of the coronary artery is based at least in part on one or more percentages of stenosis present in the region of the coronary artery. In some embodiments, the probability threshold of ischemia for the region of the coronary artery is based at least in part on multiple reference values ​​of percentages of stenosis with known presence or absence of ischemia derived from multiple other subjects in the reference value database 1114. In some embodiments, the probability threshold of ischemia comprises a statistical likelihood that the region of the coronary artery is ischemic. In some embodiments, a probability threshold of ischemia for the region of the coronary artery higher than a predetermined threshold indicates the need for further evaluation of the subject for ischemia. In some embodiments, the probability threshold of ischemia for the region of the coronary artery is determined using a machine learning algorithm trained on multiple reference values ​​of percentages of stenosis with known presence or absence of ischemia derived from multiple other subjects. In some embodiments, the probability threshold of ischemia for a region of a coronary artery is determined based at least in part on a percentage of all stenoses identified within the region of the coronary artery. In some embodiments, the probability threshold of ischemia for a region of a coronary artery is determined based at least in part on a weighted scale of the percentage of all stenoses identified within the region of the coronary artery. In some embodiments, the probability threshold of ischemia for a region of a coronary artery is determined by calculating the volume of the region of the coronary artery where a stenosis is located. In some embodiments, the probability threshold of ischemia for a region of a coronary artery is determined by adding together the sizes of each stenosis. In some embodiments, the probability threshold of ischemia for a region of a coronary artery is determined by determining the distance between stenoses. In some embodiments, the probability threshold of ischemia for a region of a coronary artery is determined by calculating the blood flow within the region of the coronary artery where a stenosis is located. In some embodiments, the probability threshold of ischemia for a region of a coronary artery comprises a binary output. In some embodiments, the probability threshold of ischemia for a region of a coronary artery comprises a continuous scale output.

[0240] In some embodiments, the system may be configured to determine that further analysis of ischemia and / or fractional flow reserve (FFR) is warranted and / or required based on the determined ischemia probability threshold. For example, in some embodiments, if the ischemia probability threshold is determined to be high or higher than a predetermined threshold based on plaque and / or stenosis, the system may be configured to determine that further analysis of ischemia and / or FFR is warranted and / or required for the patient. As part of performing the further analysis, for example, in some embodiments, at block 1116, the system may be configured to access and / or modify one or more second medical images. In some embodiments, the second medical images may include one or more arteries, such as the subject's coronary arteries, carotid arteries, and / or other arteries. In some embodiments, the second medical images are configured to be used to determine the fractional flow reserve of a region of the coronary arteries. In some embodiments, the second medical images may be stored in a medical image database 1104. In some embodiments, the medical image database 1104 may be accessed locally by the system and / or may be remotely located and accessed via a network connection. The second medical images can include images acquired using one or more modalities, such as CT, dual-energy computed tomography (DECT), spectral CT, photon-counting CT, X-ray, ultrasound, echocardiography, intravascular ultrasound (IVUS), magnetic resonance (MR) imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single-photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS). In some embodiments, the second medical images include one or more of contrast-enhanced CT images, non-contrast-enhanced CT images, MR images, and / or images obtained using any of the aforementioned modalities. In some embodiments, the second medical images can be the same as the first medical images. In other words, in some embodiments, further analysis regarding ischemia, such as FFR, can be performed on the same images analyzed for plaque, stenosis, and / or probability thresholds for ischemia.In some embodiments, the second medical image differs from the first medical image. For example, in some embodiments, the second medical image for further analysis of ischemia may have different resolution, quality, contrast, region of interest, and / or the like compared to the first medical image. As such, in some embodiments, the first medical image and / or its analysis can be used as a gatekeeper to determine whether further analysis of ischemia and / or acquisition of a different medical image is necessary. In some embodiments, the second medical image is modified for analysis using one or more image processing techniques, for example, to enhance image quality, alter contrast, and / or identify regions of interest.

[0241] In some embodiments, at block 1118, the system may be configured to determine the fractional flow reserve of a region of a coronary artery using one or more second and / or first medical images. In some embodiments, the systems, devices, and methods determine the fractional flow reserve of a region of a coronary artery if a probability threshold of ischemia for the region of the coronary artery is higher than a predetermined threshold. In some embodiments, determining the fractional flow reserve includes determining, by a computer system, the fractional flow reserve of the region of the coronary artery using computational fluid dynamics. In some embodiments, the computational fluid dynamics analysis is performed on the one or more second and / or first medical images. In some embodiments, the fractional flow reserve is determined based on one or more of invasive fractional flow reserve, computed tomography (CT) fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio. In some embodiments, further assessment of the subject's ischemia includes invasive fractional flow reserve. In some embodiments, the further assessment of the subject's ischemia includes computed tomography (CT) fractional flow reserve. In some embodiments, the further assessment of the subject's ischemia includes one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0242] In some embodiments, the system may be configured to repeat one or more processes described in connection with blocks 1102-118, for example, for one or more other vessels, segments, regions of plaque, different subjects, and / or the same subject at different times. Thus, in some embodiments, the system may provide longitudinal disease tracking and / or personalized treatment for a subject.

[0243] The computer system 902 of FIG. 9, and possibly the analysis and / or risk assessment module 940, may be configured to perform functions, methods, operations, and / or processes for determining ischemia based on the image-based analysis of stenosis described herein, such as those described above with reference to FIG. 11.

[0244] The following are non-limiting examples of specific embodiments of systems and methods for determining ischemia based on image-based analysis of stenosis. Other embodiments may include one or more other or different features discussed herein.

[0245] Embodiment 1: A computer-implemented method for determining a probability threshold for coronary ischemia based at least in part on a percentage of stenosis generated from an image-based analysis, the method comprising: accessing, by a computer system, a first medical image of a subject, the first medical image including a region of a coronary artery of the subject; analyzing, by the computer system, the first medical image using image segmentation to identify the region of the coronary artery; and identifying, by the computer system, one or more regions of plaque within the region of the coronary artery, the one or more regions of plaque being identified based at least in part on a density of one or more pixels in the first medical image corresponding to the one or more regions of plaque; and identifying, by the computer system, one or more regions of plaque present in the region of the coronary artery resulting from the one or more regions of plaque. determining a percentage of stenosis above one or more of the stenosis present in the region of the coronary artery, wherein the one or more percentages of stenosis are determined based at least in part on interpolating the luminal volume or diameter of the region of the coronary artery in the absence of one or more regions of plaque; and determining, by the computer system, a probability threshold of ischemia for the region of the coronary artery based at least in part on the percentage of stenosis present in the region of the coronary artery and a plurality of reference values ​​of percentage of stenosis derived from a plurality of other subjects, where the presence or absence of ischemia is known, wherein the probability threshold of ischemia comprises a statistical likelihood that the region of the coronary artery is ischemic, and wherein a probability threshold of ischemia for the region of the coronary artery higher than the predetermined threshold indicates a need for further evaluation of ischemia in the subject, the computer system comprising a computer processor and an electronic storage medium.

[0246] Embodiment 2: The computer-implemented method of embodiment 1, further comprising determining the fractional flow reserve of the region of the coronary artery if the probability threshold of ischemia for the region of the coronary artery is higher than a predetermined threshold.

[0247] Embodiment 3: The computer-implemented method of embodiment 2, wherein determining the fractional flow reserve includes accessing, by the computer system, a second medical image, the second medical image including a region of a coronary artery of the subject, and determining, by the computer system, the fractional flow reserve of the region of the coronary artery using computational fluid dynamics.

[0248] Embodiment 4: The computer-implemented method of embodiment 2, wherein the fractional flow reserve is determined based on one or more of invasive fractional flow reserve, computed tomography (CT) fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0249] Embodiment 5: The computer-implemented method of embodiment 1, wherein the further assessment of the subject's ischemia comprises invasive fractional flow reserve.

[0250] Embodiment 6: The computer-implemented method of embodiment 1, wherein the further assessment of the subject's ischemia includes computed tomography (CT) fractional flow reserve.

[0251] Embodiment 7: The computer-implemented method of embodiment 1, wherein the further assessment of the subject's ischemia includes one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0252] Embodiment 8: The computer-implemented method of embodiment 1, wherein the probability threshold of ischemia for a region of a coronary artery is determined using a machine learning algorithm trained on multiple reference values ​​of percentage stenosis where the presence or absence of ischemia is known, derived from multiple other subjects.

[0253] Embodiment 9: The computer-implemented method of embodiment 8, wherein the presence or absence of ischemia is derived from a plurality of other subjects using one or more of invasive fractional flow reserve, CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0254] Embodiment 10: The computer-implemented method of embodiment 1, wherein the probability threshold of ischemia for a region of a coronary artery is determined based at least in part on a percentage of all stenoses identified within the region of the coronary artery.

[0255] Embodiment 11: The computer-implemented method of embodiment 1, wherein the probability threshold of ischemia for a region of a coronary artery is determined based at least in part on a weighted measure of the percentage of all stenoses identified within the region of the coronary artery.

[0256] Embodiment 12: The computer-implemented method of embodiment 1, wherein the probability threshold of ischemia for a region of a coronary artery constitutes a binary output.

[0257] Embodiment 13: The computer-implemented method of embodiment 1, wherein the probability threshold of ischemia for the coronary artery territory constitutes a continuous scale output.

[0258] Embodiment 14: The computer-implemented method of embodiment 1, wherein the density includes a radial density.

[0259] Embodiment 15: The computer-implemented method of embodiment 1, wherein the density comprises a material density.

[0260] Embodiment 16: The computer-implemented method of embodiment 1, wherein the first medical image is acquired using CT.

[0261] Embodiment 17: The computer-implemented method of embodiment 1, wherein the first medical image is acquired using an imaging modality including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0262] Embodiment 18: A non-transitory computer-readable medium configured for determining a probability threshold for coronary ischemia based at least in part on a percentage of stenosis generated from an image-based analysis, the computer-readable medium having program instructions for causing a hardware processor to execute a method, the method comprising: accessing a first medical image of a subject, the first medical image including a region of a coronary artery of the subject; analyzing the first medical image using image segmentation to identify the region of the coronary artery; and identifying one or more regions of plaque within the region of the coronary artery, the one or more regions of plaque being identified based at least in part on a density of one or more pixels in the first medical image corresponding to the one or more regions of plaque. determining a percentage of one or more stenosis present in a region of the coronary artery arising from one or more regions of plaque, wherein the percentage of the one or more stenosis is determined based at least in part on interpolating the luminal volume or diameter of the region of the coronary artery in the absence of the one or more regions of plaque; and determining a probability threshold of ischemia for the region of the coronary artery based at least in part on the percentage of one or more stenosis present in the region of the coronary artery and a plurality of reference values ​​of percentage of stenosis derived from a plurality of other subjects, the reference values ​​being known to have or not have ischemia, wherein the probability threshold of ischemia comprises a statistical likelihood that the region of the coronary artery is ischemic, and wherein a probability threshold of ischemia for the region of the coronary artery higher than the predetermined threshold indicates the need for further evaluation of ischemia in the subject.

[0263] Embodiment 19: A non-transitory computer-readable medium configured as in embodiment 18, further comprising determining fractional flow reserve of a region of a coronary artery if the probability threshold of ischemia for the region of a coronary artery is higher than a predetermined threshold.

[0264] Embodiment 20: A non-transitory computer-readable medium configured as in embodiment 19, wherein determining fractional flow reserve comprises accessing a second medical image, the second medical image including a region of a coronary artery of the subject, and determining fractional flow reserve of the region of the coronary artery using computational fluid dynamics.

[0265] Embodiment 21: A non-transitory computer-readable medium configured as in embodiment 19, wherein the fractional flow reserve is determined based on one or more of invasive fractional flow reserve, computed tomography (CT) fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0266] Embodiment 22: A non-transient computer-readable medium configured as in embodiment 18, wherein the further assessment of the subject's ischemia includes invasive fractional flow reserve.

[0267] Embodiment 23: A non-transitory computer-readable medium configured as in embodiment 18, wherein the further assessment of the subject's ischemia includes computed tomography (CT) fractional flow reserve.

[0268] Embodiment 24: A non-transitory computer-readable medium configured as in embodiment 18, wherein the further assessment of the subject's ischemia includes one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0269] Embodiment 25: A non-transitory computer-readable medium configured as in embodiment 18, wherein the probability threshold of ischemia for a region of a coronary artery is determined using a machine learning algorithm trained on multiple reference values ​​of percentage stenosis where the presence or absence of ischemia is known, derived from multiple other subjects.

[0270] Embodiment 26: A non-transitory computer-readable medium configured as in embodiment 25, wherein the presence or absence of ischemia is derived from a plurality of other subjects using one or more of invasive fractional flow reserve, CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0271] Embodiment 27: A non-transitory computer-readable medium configured as in embodiment 18, wherein the probability threshold of ischemia for a region of a coronary artery is determined based at least in part on a percentage of all stenoses identified within the region of the coronary artery.

[0272] Embodiment 28: A non-transitory computer-readable medium configured as in embodiment 18, wherein the probability threshold of ischemia for a region of a coronary artery is determined based at least in part on a weighted measure of the percentage of all stenoses identified within the region of the coronary artery.

[0273] Embodiment 29: A non-transitory computer-readable medium configured as in embodiment 18, wherein the probability threshold of ischemia for a region of a coronary artery constitutes a binary output.

[0274] Embodiment 30: A non-transitory computer-readable medium configured as in embodiment 18, wherein the ischemia probability threshold for the coronary artery territory constitutes a continuous scale output.

[0275]

[0033] Embodiment 31. The non-transitory computer-readable medium configured as in embodiment 18, wherein the density comprises a radiation density.

[0276] Embodiment 32: The non-transitory computer-readable medium configured as in embodiment 18, wherein the density comprises a material density.

[0277] Embodiment 33: A non-transitory computer-readable medium configured as in embodiment 18, wherein the first medical image is acquired using CT.

[0278] Embodiment 34: A non-transitory computer-readable medium configured as in embodiment 18, wherein the first medical image is acquired using an imaging modality including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0279] Embodiment 35: A system comprising: accessing, by a computer system, a first medical image of a subject, the first medical image including a region of a coronary artery of the subject; analyzing, by the computer system, the first medical image using image segmentation to identify the region of the coronary artery; identifying, by the computer system, one or more regions of plaque within the region of the coronary artery, the one or more regions of plaque being identified based at least in part on a density of one or more pixels in the first medical image corresponding to the one or more regions of plaque; and determining, by the computer system, a percentage of one or more stenoses present in the region of the coronary artery resulting from the one or more regions of plaque, the one or more stenoses being present in the region of the coronary artery. determining, by the computer system, a probability threshold of ischemia for the region of the coronary artery based at least in part on the one or more percentage stenosis present in the region of the coronary artery and a plurality of reference values ​​of percentage stenosis for which the presence or absence of ischemia is known, derived from a plurality of other subjects, wherein the probability threshold of ischemia comprises a statistical likelihood that the region of the coronary artery is ischemic, and wherein a probability threshold of ischemia for the region of the coronary artery higher than the predetermined threshold indicates the need for further evaluation of the subject's ischemia, the computer system comprising a computer processor and an electronic storage medium.

[0280] Embodiment 36: The system of embodiment 35, further comprising determining the fractional flow reserve of the region of the coronary artery if the probability threshold of ischemia for the region of the coronary artery is higher than a predetermined threshold.

[0281] Embodiment 37: The system of embodiment 36, wherein determining the fractional flow reserve comprises accessing, by a computer system, a second medical image, the second medical image including a region of a coronary artery of the subject, and determining, by the computer system, the fractional flow reserve of the region of the coronary artery using computational fluid dynamics.

[0282] Embodiment 38: The system of embodiment 36, wherein the fractional flow reserve is determined based on one or more of invasive fractional flow reserve, computed tomography (CT) fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0283] Embodiment 39: The system of embodiment 35, wherein the further assessment of the subject's ischemia includes invasive fractional flow reserve.

[0284] Embodiment 40: The system of embodiment 35, wherein the further assessment of the subject's ischemia includes computed tomography (CT) fractional flow reserve.

[0285] Embodiment 41: The system of embodiment 35, wherein the further assessment of the subject's ischemia includes one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0286] Embodiment 42: The system described in embodiment 35, wherein the probability threshold of ischemia for a region of a coronary artery is determined using a machine learning algorithm trained on multiple reference values ​​of percentage stenosis where the presence or absence of ischemia is known, derived from multiple other subjects.

[0287] Embodiment 43: The system of embodiment 35, wherein the presence or absence of ischemia is derived from a plurality of other subjects using one or more of invasive fractional flow reserve, CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0288] Embodiment 44: The system described in embodiment 35, wherein the probability threshold of ischemia for a region of a coronary artery is determined based at least in part on the percentage of all stenoses identified within the region of the coronary artery.

[0289] Embodiment 45: The system of embodiment 35, wherein the probability threshold of ischemia for a region of a coronary artery is determined based at least in part on a weighted measure of the percentage of all stenoses identified within the region of the coronary artery.

[0290] Embodiment 46: The system described in embodiment 35, wherein the probability threshold of ischemia for the coronary artery region constitutes a binary output.

[0291] Embodiment 47: The system described in embodiment 35, wherein the probability threshold of ischemia for the coronary artery region constitutes a continuous scale output.

[0292] Embodiment 48: The system described in embodiment 35, wherein the density includes a radial density.

[0293] Embodiment 49: The system of embodiment 35, wherein the density comprises a substance density.

[0294] Embodiment 50: The system described in embodiment 35, wherein the first medical image is acquired using CT.

[0295] Embodiment 51: The system described in embodiment 35, wherein the first medical image is acquired using an imaging modality including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0296] Multivariate image-based ischemia analysis Disclosed herein are systems, devices, and methods for multivariate image-based ischemia analysis. Various embodiments described herein relate to systems, devices, and methods for multivariate image-based ischemia analysis. In particular, in some embodiments, the systems, devices, and methods described herein relate to determining one or more variables for multiple blood vessels, such as each of a first coronary artery and a second coronary artery. In some embodiments, the variables are determined based on medical images of the patient, such as non-invasively acquired medical images. The variables for the multiple blood vessels can be analyzed, for example, using machine learning algorithms, to determine the presence of ischemia in one of the blood vessels. That is, a determination of the presence of ischemia can be made for one of the blood vessels based on an analysis of variables determined from two or more (e.g., two, three, or more) blood vessels. This can be advantageous because the effects of diseases, such as coronary artery disease, can often spread across several blood vessels. For example, a stenosis in one blood vessel can affect blood flow in another blood vessel. In some embodiments, the analyses described herein account for the diffuse nature of disease effects by determining ischemia in one vessel based on analysis of variables determined for multiple vessels. In some embodiments, the systems, devices, and methods described herein are configured to determine risk of coronary artery disease (CAD), e.g., myocardial infarction (MI), based on one or more plaque analyses described herein. In some embodiments, the systems, devices, and methods described herein are configured to generate suggested treatments and / or visual representations based on the determined risk of CAD and / or one or more plaque analyses described herein.

[0297] In some embodiments, it may be difficult or impossible to determine a measurement of a vessel's fractional flow reserve (FFR) solely by pinpointing the location of a specific stenosis within the vessel. This is because the spread of disease across the length of the vessel or within a vascular network can contribute to the ischemia of a single vessel. For example, the sum of plaque across or along a vessel or across multiple vessels, as well as the vascular morphology across or along a vessel or across multiple vessels, all contribute to the presence of ischemia. As described herein, mixed-effects modeling based on variables determined across or along a vessel and / or across multiple vessels within a vascular network can be used to more accurately determine the presence of ischemia. For example, determining or measuring a single stenosis alone may be insufficient to determine the FFR value of that vessel. Instead, it may be necessary to know all stenoses within a vessel, including location, vessel morphology, lumen volume, and vascular volume, as well as similar parameters of other vessels or arteries within the vascular network.

[0298] Previous approaches to determining the presence of ischemia focused on determining specific instances of stenosis without considering the diffuse effects of disease spread throughout the vascular network. However, as recognized herein, the interplay between all variables across location, geometry, vessel region, distal, proximal, bifurcation, trifurcation, luminal volume, etc., all affect the presence of ischemia.

[0299] For example, a stenosis in one blood vessel can increase blood flow in a less stenotic vessel, and thus, it can be important to consider the effects caused in adjacent vessels when determining the presence of ischemia.

[0300] As a more specific example, in some embodiments, one or more of a plurality of parameters can be determined for one or more regions or one or more lesions of two or more blood vessels. These parameters can include, among others, one or more of the following: lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, post-stenosis baseline diameter, pre-stenosis baseline diameter, vessel length, lumen volume, number of CTOs, vessel volume, number of stenoses, total plaque volume, number of mild stenoses, volume of low-density plaque, etc. These parameters, in some embodiments, can be determined based on one or more medical images (e.g., non-invasively acquired medical images) of the patient, for example, using machine learning, artificial intelligence, or other image-based recognition techniques. The presence of ischemia can then be determined for one of the blood vessels based on the plurality of blood vessels determined for that blood vessel as well as one or more other blood vessels, such as one or more adjacent blood vessels, for example, using a machine learning algorithm. The reason for this is that measurements taken from one vessel or vascular region can be used to inform the probability of ischemia in another vessel.

[0301] In some embodiments, lesion-level data can also be used. In some embodiments, these principles can be applied at even higher granularity, for example, using only a few axial slice-specific data. In some embodiments, stenosis can be thought of as a two-dimensional measurement, for example, the percent stenosis within a two-dimensional slice. However, because stenosis is generally three-dimensional in nature, it may be advantageous to analyze the stenosis three-dimensionally, for example, along the length of a vessel. For example, variables can be determined for each of multiple two-dimensional slices, and the variables determined from each slice can be analyzed together to gain an understanding of the three-dimensional spatial distribution along the length of the vessel. In some embodiments, instead of analyzing segment-level data based on multiple cross-sections merged and analyzed together (like a loaf of bread), it may be advantageous to analyze slices individually (e.g., slice-level data) and apply machine learning algorithms to determine how the three-dimensional spatial distribution affects ischemia.

[0302] As mentioned above, the determination of ischemia is not only dependent on a single stenosis, but also on other parameters, such as other stenoses in the blood vessel and other vascular morphological parameters. Furthermore, blood vessels receive signals from other blood vessels, so it is not just the parameters from one blood vessel that are important.

[0303] FIG. 12 is a flowchart illustrating exemplary embodiment(s) of systems, devices, and methods for multivariate analysis of ischemia. As shown in FIG. 12 , in some embodiments, the system may be configured to access and / or modify one or more medical images at block 1202. In some embodiments, the medical images may include one or more arteries, such as the subject's coronary arteries, carotid arteries, and / or other arteries. In some embodiments, the medical images may be stored in a medical image database 1204. In some embodiments, the medical image database 1204 may be accessed locally by the system and / or may be located remotely and accessed via a network connection. The medical images may include images acquired using, for example, CT, dual-energy computed tomography (DECT), spectral CT, photon-counting CT, x-ray, ultrasound, echocardiography, intravascular ultrasound (IVUS), magnetic resonance (MR) imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single-photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS). In some embodiments, the medical images include one or more of contrast-enhanced CT images, non-contrast-enhanced CT images, MR images, and / or images obtained using any of the above-mentioned modalities.

[0304] In some embodiments, the system may be configured to automatically and / or dynamically perform one or more analyses of the medical image as discussed herein. For example, in some embodiments, in block 1206, the system may be configured to identify one or more blood vessels, such as one or more arteries. As shown in FIG. 12 , in some embodiments, the system may identify at least two blood vessels (e.g., a first blood vessel and a second blood vessel) in the image. In some embodiments, a greater number of blood vessels may be identified, e.g., 3, 4, 5, 6, 7, 8, or more blood vessels. As described herein, two or more of the plurality of blood vessels may be analyzed to determine the presence of ischemia in one of the identified blood vessels. The one or more arteries may include, among others, a coronary artery, a carotid artery, an aorta, a renal artery, a lower limb artery, an upper limb artery, and / or a cerebral artery. In some embodiments, the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0305] In some embodiments, the system may be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically identify one or more arteries or coronary arteries using image processing. For example, in some embodiments, one or more AI and / or ML algorithms may be trained using a convolutional neural network (CNN) on a set of medical images in which arteries or coronary arteries have been identified, thereby enabling the AI ​​and / or ML algorithms to automatically identify arteries or coronary arteries directly from the medical images. In some embodiments, the arteries or coronary arteries are identified by size and / or location.

[0306] In some embodiments, at block 1208, the system may be configured to identify one or more regions or lesions within a plurality of blood vessels, e.g., a first blood vessel and a second blood vessel. In some embodiments, the region may comprise a portion, subportion, or a portion of the length of the blood vessel. For example, the region may be 1%, 5%, 10%, 20%, 25%, or 50% of the length of the blood vessel, or any other percentage of the length of the blood vessel, up to and including the entire length of the blood vessel.

[0307] In some embodiments, at block 1210, the system may be configured to identify one or more regions of plaque within the medical image, such as the identified regions of the first and second blood vessels or regions of plaque associated with a lesion. In some embodiments, the system may be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically identify one or more regions of plaque using image processing. For example, in some embodiments, the one or more AI and / or ML algorithms may be trained using a convolutional neural network (CNN) on a set of medical images in which regions of plaque have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify regions of plaque directly from the medical images. In some embodiments, the system is configured to identify the vessel wall and the lumen wall and classify everything between the vessel wall and the lumen wall as plaque.

[0308] In some embodiments, the system may be configured to analyze and / or characterize one or more regions of plaque based on density. For example, in some embodiments, the system may be configured to analyze and / or characterize one or more regions of plaque based on absolute density and / or relative density and / or radiometric density. In some embodiments, the system may be configured to classify a region of plaque as one of low-density non-calcified plaque, non-calcified plaque, and calcified plaque using any one or more processes and / or features described herein.

[0309] In some embodiments, the system may be configured to analyze and / or characterize one or more regions of plaque based on one or more distances. For example, as described herein, in some embodiments, the system may be configured to determine the distance between low-density non-calcified plaque and the lumen wall and / or vessel wall. In some embodiments, the proximity of low-density non-calcified plaque to the lumen wall may indicate a high-risk plaque and / or CAD. Conversely, in some embodiments, the location of low-density non-calcified plaque farther from the lumen wall may indicate a lower risk. In some embodiments, the system may be configured to utilize one or more predetermined thresholds in determining risk factors associated with the proximity of low-density non-calcified plaque to the vessel wall and / or lumen wall. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine one or more distances to and / or from one or more regions of plaque.

[0310] In some embodiments, the system may be configured to analyze and / or characterize one or more regions of a low-density non-calcified plaque based on the morphology or shape and / or one or more axial measurements of the plaque. In some embodiments, the system may be configured to determine the length of one or more axes of the low-density non-calcified plaque, such as the long axis of a longitudinal section and / or the long and / or short axes of a transverse section of the low-density non-calcified plaque. In some embodiments, the system may be configured to determine the morphology and / or shape of the low-density non-calcified plaque using one or more axial measurements. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine one or more axial measurements of one or more regions of the plaque.

[0311] In some embodiments, the system can be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically classify the shape of one or more regions of plaque using image processing. For example, in some embodiments, one or more AI and / or ML algorithms can be trained using a convolutional neural network (CNN) on a set of medical images in which the shapes of regions of plaque have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify the shape or morphology of regions of plaque directly from the medical images. In some embodiments, the system can be configured to classify the shape or morphology of regions of plaque as one or more of crescent, lobe, round, or bean-shaped. In some embodiments, round and / or bean-shaped plaques can be associated with a high risk, while crescent and / or lobe-shaped plaques can be associated with a low risk of CAD.

[0312] In some embodiments, the system may be configured to analyze and / or characterize one or more regions of plaque based on one or more sizes and / or volumes. For example, in some embodiments, the system may be configured to determine the size and / or volume of plaque based at least in part on measurements of one or more axes described herein. In some embodiments, the system may be configured to determine the size and / or volume of a region of plaque directly from analysis of a three-dimensional image scan. In some embodiments, the system may be configured to determine the size and / or volume of total plaque, low-density non-calcified plaque, non-calcified plaque, calcified plaque, and / or the ratio between two of the aforementioned volumes or sizes. In some embodiments, a high total plaque volume and / or a high low-density non-calcified plaque and / or non-calcified plaque volume may be associated with a high risk of CAD. In some embodiments, a high ratio of low-density non-calcified plaque volume to total plaque volume and / or a high ratio of non-calcified plaque volume to total plaque volume may be associated with a high risk of CAD. In some embodiments, a high calcified plaque volume and / or a high ratio of calcified plaque volume to total plaque volume may be associated with a low risk of CAD. In some embodiments, the system may be configured to utilize one or more predetermined thresholds to determine the risk of CAD based on plaque volume, size, or one or more ratios thereof. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine the size and / or volume of one or more regions of plaque.

[0313] In some embodiments, the system may be configured to analyze and / or characterize plaque based on its degree of embedment. For example, in some embodiments, the system may be configured to determine the extent to which low-density non-calcified plaque is embedded or surrounded by non-calcified plaque or calcified plaque. In some embodiments, the system may be configured to analyze the embedment of low-density non-calcified plaque based on the degree to which the low-density non-calcified plaque is surrounded by other types of plaque. In some embodiments, a higher degree of embedment of low-density non-calcified plaque may indicate a higher risk of CAD. For example, in some embodiments, low-density non-calcified plaque surrounded by 270 degrees or more of non-calcified plaque may be associated with a higher risk of CAD. In some embodiments, the system may be configured to utilize one or more image processing algorithms to automatically and / or dynamically determine the embedment degree of one or more regions of plaque.

[0314] In some embodiments, at block 1212, the system may be configured to analyze a region or lesion of a first vessel, a region or lesion of a second vessel, plaque in the region or lesion of the first vessel, and plaque in the region or lesion of the second vessel to determine a plurality of variables for each of the region or lesion of the first vessel and the second vessel. In some embodiments, the plurality of variables includes stenosis.

[0315] In some embodiments, the plurality of variables includes plaque volume. In some embodiments, the plaque volume includes one or more of total plaque volume, low-density non-calcified plaque volume, non-calcified plaque volume, or calcified plaque volume. In some embodiments, the plaque volume is determined at least in part based on analyzing the density of one or more pixels corresponding to the plaque in the medical image. In some embodiments, the density includes material density. In some embodiments, the density includes radiometric density. For example, in some embodiments, the system may be configured to characterize a particular region of plaque as low-density non-calcified plaque if the radiometric density of image pixels or voxels corresponding to that region of plaque is between about -189 and about 30 Hounsfield Units (HU). In some embodiments, the system may be configured to characterize a particular region of plaque as non-calcified plaque if the radiometric density of image pixels or voxels corresponding to that region of plaque is between about 31 and about 350 HU. In some embodiments, the system can be configured to characterize a particular region of plaque as calcified plaque if the radiometric density of the image pixels or voxels corresponding to that region of plaque is between about 351 and about 2500 HU.In some embodiments, the lower and / or upper Hounsfield unit boundary thresholds for determining whether a plaque corresponds to one or more of low density non-calcified plaque, non-calcified plaque, and / or calcified plaque are about -1000 HU, about -900 HU, about -800 HU, about -700 HU, about -600 HU, about -500 HU, about -400 HU, about -300 HU, about -200 HU, about -190 HU, about -180 HU, about -170 HU, about -160 HU, about -150 HU, about -140 HU, about -130 HU, about -120 HU, about -110 HU, about -15 ... U, approximately -100HU, approximately -90HU, approximately -80HU, approximately -70HU, approximately -60HU, approximately -50HU, approximately -40HU, approximately -30HU, approximately -20HU, approximately -10HU, approximately 0HU, approximately 10HU, approximately 20HU, approximately 30HU, approximately 40HU, approximately 50HU, approximately 60HU, approximately 70HU, approximately 80 HU, about 90HU, about 100HU, about 110HU, about 120HU, about 130HU, about 140HU, about 150HU, about 160HU, about 170HU, about 180HU, about 190HU, about 200HU, about 210HU, about 220HU, about 230HU, about 240HU, about 250HU, about 2 60HU, about 270HU, about 280HU, about 290HU, about 300HU, about 310HU, about 320HU, about 330HU, about 340HU, about 350HU, about 360HU, about 370HU, about 380HU, about 390HU, about 400HU, about 410HU, about 420HU, about 430HU , approx. 440HU, approx. 450HU, approx. 460HU, approx. 470HU, approx. 480HU, approx. 490HU, approx. 500HU, approx. 510HU, approx. 520HU, approx. 530HU, approx. 540HU, approx. The HU may be 0 HU, about 800 HU, about 900 HU, about 1000 HU, about 1100 HU, about 1200 HU, about 1300 HU, about 1400 HU, about 1500 HU, about 1600 HU, about 1700 HU, about 1800 HU, about 1900 HU, about 2000 HU, about 2100 HU, about 2200 HU, about 2300 HU, about 2400 HU, about 2500 HU, about 2600 HU, about 2700 HU, about 2800 HU, about 2900 HU, about 3000 HU, about 3100 HU, about 3200 HU, about 3300 HU, about 3400 HU, about 3500 HU, and / or about 4000 HU.

[0316] In some embodiments, at block 1214, the system may be configured to determine the presence of ischemia for one of the blood vessels, e.g., the first blood vessel, based on the plurality of blood vessels determined for each of the plurality of blood vessels, e.g., the first blood vessel and the second blood vessel. For example, in some embodiments, a machine learning algorithm may be applied to the plurality of blood vessels to determine the presence of ischemia. In some embodiments, the machine learning algorithm is trained based at least in part on a dataset including a plurality of variables and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0317] In some embodiments, at block 1216, the system may be configured to determine the risk of CAD or MI based on one or more plaque analyses described herein, for example, in connection with one or more of blocks 1202-1216. In some embodiments, the system may be configured to utilize some or all of the plaque analysis results. In some embodiments, the system may be configured to generate a weighted scale of some or all of the plaque analyses described herein when determining the risk of CAD. In some embodiments, the system may be configured to reference one or more reference values ​​of the one or more plaque analysis results when determining the risk of CAD. For example, in some embodiments, the one or more reference values ​​may be comprised of one or more values ​​obtained from populations with different CAD risk statuses, and the one or more values ​​may be comprised of one or more distances to and / or from low-density non-calcified plaque, one or more axial measurements, morphological classification, size and / or volume, and / or buried depth of low-density non-calcified plaque. In some embodiments, one or more metrics may be stored in a metrics database 1218, which may be accessed locally by the system and / or may be located remotely and accessed via a network connection.

[0318] In some embodiments, in block 120, the system may be configured to generate a graphical representation of the analysis results, the determined CAD risk, and / or the subject's suggested treatment. In some embodiments, the analysis results may be displayed on a vessel, lesion, and / or subject basis. In some embodiments, the suggested treatment may include, for example, a medical treatment such as a statin, an interventional treatment such as stent implantation, and / or a lifestyle treatment such as exercise or diet. In some embodiments, in determining the risk or status of cardiovascular disease or health and / or treatment, the system may access a plaque risk / treatment database 1222, which may be accessed locally by the system and / or may be remotely located and accessed via a network connection. In some embodiments, the plaque risk / treatment database 1222 may include reference points or data relating one or more treatments to the cardiovascular disease risk or status determined based on one or more baseline plaque analysis values.

[0319] In some embodiments, the system may be configured to repeat one or more of the processes described in connection with blocks 1202-1220, for example, one or more other vessels, segments, regions of plaque, different subjects, and / or the same subject at different times.

[0320] The computer system 902 of FIG. 9, and possibly the analysis and / or risk assessment module 940, may be configured to perform the functions, methods, operations, and / or processes for multivariate image-based analysis of ischemia described herein, such as those described above with reference to FIG. 12.

[0321] The following are non-limiting examples of specific embodiments of systems and methods for multivariate image-based analysis of ischemia. Other embodiments may include one or more other or different features discussed herein.

[0322] Embodiment 1: A computer-implemented method of determining the presence of vessel-specific ischemia based at least in part on a plurality of variables derived from analysis of non-invasive medical images, the method comprising: accessing, by a computer system, medical images of a subject, the medical images of the subject being non-invasively acquired; analyzing, by the computer system, the medical images of the subject to identify a plurality of blood vessels, the plurality of blood vessels including a first blood vessel and a second blood vessel; identifying, by the computer system, one or more regions within the first blood vessel and one or more regions within the second blood vessel; identifying, by the computer system, one or more regions of plaque within the one or more regions within the first blood vessel and the one or more regions within the second blood vessel; and analyzing, by a computer system, one or more regions in a first blood vessel, one or more regions in a second blood vessel, one or more regions of plaque in the one or more regions in the first blood vessel, and one or more regions of plaque in the one or more regions in the second blood vessel to determine a plurality of variables for each of the one or more regions in the first blood vessel and the one or more regions in the second blood vessel; and applying, by the computer system, a machine learning algorithm to determine the presence of ischemia in the first blood vessel based at least in part on the plurality of variables determined for the one or more regions in the first blood vessel and the plurality of variables determined for the one or more regions in the second blood vessel, wherein the computer system comprises a computer processor and an electronic storage medium.

[0323] Embodiment 2: The computer-implemented method of embodiment 1, wherein the plurality of variables includes stenosis.

[0324] Embodiment 3: The computer-implemented method of embodiment 1, wherein the plurality of variables includes plaque volume.

[0325] Embodiment 4: The computer-implemented method of embodiment 3, wherein the plaque volume comprises one or more of total plaque volume, low-density non-calcified plaque volume, non-calcified plaque volume, or calcified plaque volume.

[0326] Embodiment 5: The computer-implemented method of embodiment 4, wherein the volume of the plaque is determined based at least in part on analyzing the density of one or more pixels corresponding to the plaque in the medical image.

[0327] Embodiment 6: The computer-implemented method of embodiment 5, wherein the density comprises a material density.

[0328] Embodiment 7: The computer-implemented method of embodiment 5, wherein the density includes radial density.

[0329] Embodiment 8: The computer-implemented method of embodiment 7, wherein low-density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about -189 and about 30 Hounsfield units.

[0330] Embodiment 9: The computer-implemented method of embodiment 7, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 31 and about 189 Hounsfield units.

[0331] Embodiment 10: The computer-implemented method of embodiment 7, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 190 and about 350 Hounsfield units.

[0332] Embodiment 11: The computer-implemented method of embodiment 7, wherein the calcified plaque corresponds to one or more pixels having a radiodensity value between approximately 351 and 2500 Hounsfield units.

[0333] Embodiment 12: The computer-implemented method of embodiment 1, wherein the medical image comprises a computed tomography (CT) image.

[0334] Embodiment 13: The computer-implemented method of embodiment 1, wherein the medical image is obtained using imaging techniques including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0335] Embodiment 14: The computer-implemented method of embodiment 1, wherein the plurality of variables includes one or more of lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, post-stenosis baseline diameter, pre-stenosis baseline diameter, vessel length, lumen volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or volume of low-density plaque.

[0336] Embodiment 15: The computer-implemented method of embodiment 1, wherein the plurality of blood vessels includes one or more coronary arteries.

[0337] Embodiment 16: The computer-implemented method of embodiment 15, wherein the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0338] Embodiment 17: The computer-implemented method of embodiment 1, wherein the machine learning algorithm is trained based at least in part on a dataset comprising a plurality of variables and the presence of ischemia derived using invasive fractional flow reserve.

[0339] Embodiment 18: The computer-implemented method of embodiment 1, wherein the machine learning algorithm is trained based at least in part on a dataset including a plurality of variables and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0340] Embodiment 19: The computer-implemented method of embodiment 1, further comprising generating, by the computer system, an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular events (MACE) based at least in part on the determination of the presence of ischemia in the first vessel.

[0341] Embodiment 20: The computer-implemented method of embodiment 19, further comprising generating, by the computer system, a graphical representation of the generated assessment of the risk of CAD or MACE.

[0342] Embodiment 21: The computer-implemented method of embodiment 19, further comprising generating, by the computer system, a recommended treatment for the subject based at least in part on the generated assessment of risk of CAD or MACE.

[0343] Embodiment 22: A computer-implemented method for determining the presence of vessel-specific ischemia based at least in part on a plurality of variables derived from analysis of non-invasive medical images, comprising: accessing, by a computer system, medical images of a subject, wherein the medical images of the subject are non-invasively acquired; analyzing, by the computer system, the medical images of the subject to identify a plurality of blood vessels, including a first blood vessel and a second blood vessel; identifying, by the computer system, one or more lesions in the first blood vessel and one or more lesions in the second blood vessel; and identifying, by the computer system, one or more regions of plaque within the one or more lesions in the first blood vessel and one or more regions of plaque within the one or more lesions in the second blood vessel. analyzing, by the computer system, one or more lesions in the first blood vessel, one or more lesions in the second blood vessel, one or more regions of plaque within the one or more lesions in the first blood vessel, and one or more regions of plaque within the one or more lesions in the second blood vessel to determine a plurality of variables for each of the one or more lesions in the first blood vessel and the one or more lesions in the second blood vessel; and applying, by the computer system, a machine learning algorithm to determine the presence of ischemia in the first blood vessel based at least in part on the plurality of variables determined for the one or more lesions in the first blood vessel and the plurality of variables determined for the one or more lesions in the second blood vessel, wherein the computer system comprises a computer processor and an electronic storage medium.

[0344] Embodiment 23: The computer-implemented method of embodiment 22, wherein the plurality of variables includes stenosis.

[0345] Embodiment 24: The computer-implemented method of embodiment 22, wherein the plurality of variables includes plaque volume.

[0346] Embodiment 25: The computer-implemented method of embodiment 24, wherein the plaque volume comprises one or more of total plaque volume, low-density non-calcified plaque volume, non-calcified plaque volume, or calcified plaque volume.

[0347] Embodiment 26: The computer-implemented method of embodiment 25, wherein the volume of the plaque is determined based at least in part on analyzing the density of one or more pixels corresponding to the plaque in the medical image.

[0348] Embodiment 27: The computer-implemented method of embodiment 26, wherein the density comprises a material density.

[0349] Embodiment 28: The computer-implemented method of embodiment 26, wherein the density includes radial density.

[0350] Embodiment 29: The computer-implemented method of embodiment 28, wherein low-density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about -189 and about 30 Hounsfield units.

[0351] Embodiment 30: The computer-implemented method of embodiment 28, wherein the low-density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 31 and about 189 Hounsfield units.

[0352] Embodiment 31: The computer-implemented method of embodiment 28, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 190 and about 350 Hounsfield units.

[0353] Embodiment 32: The computer-implemented method of embodiment 28, wherein the calcified plaque corresponds to one or more pixels having a radiodensity value between approximately 351 and 2500 Hounsfield units.

[0354] Embodiment 33: The computer-implemented method of embodiment 22, wherein the medical image comprises a computed tomography (CT) image.

[0355] Embodiment 34: The computer-implemented method of embodiment 22, wherein the medical image is obtained using imaging techniques including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0356] Embodiment 35: The computer-implemented method of embodiment 22, wherein the plurality of variables includes one or more of lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, post-stenosis baseline diameter, pre-stenosis baseline diameter, vessel length, lumen volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or volume of low-density plaque.

[0357] Embodiment 36: The computer-implemented method of embodiment 22, wherein the plurality of blood vessels includes one or more coronary arteries.

[0358] Embodiment 37: The computer-implemented method of embodiment 36, wherein the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0359] Embodiment 38: The computer-implemented method of embodiment 22, wherein the machine learning algorithm is trained based at least in part on a dataset comprising a plurality of variables and the presence of ischemia derived using invasive fractional flow reserve.

[0360] Embodiment 39: The computer-implemented method of embodiment 22, wherein the machine learning algorithm is trained based at least in part on a dataset including a plurality of variables and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0361] Embodiment 40: The computer-implemented method of embodiment 22, further comprising generating, by the computer system, an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular events (MACE) based at least in part on the determination of the presence of ischemia in the first vessel.

[0362] Embodiment 41: The computer-implemented method of embodiment 40, further comprising generating, by the computer system, a graphical representation of the generated assessment of the risk of CAD or MACE.

[0363] Embodiment 42: The computer-implemented method of embodiment 40, further comprising generating, by the computer system, a recommended treatment for the subject based at least in part on the generated assessment of risk of CAD or MACE.

[0364] Embodiment 43: A system for determining the presence of vessel-specific ischemia based at least in part on a plurality of variables derived from analysis of non-invasive medical images, comprising: a non-transitory computer storage medium configured to store at least computer-executable instructions; and one or more computer hardware processors in communication with the first non-transitory computer storage medium, the one or more computer hardware processors at least accessing medical images of a subject, the medical images of the subject being non-invasively acquired; analyzing the medical images of the subject to identify a plurality of blood vessels, including a first blood vessel and a second blood vessel; identifying one or more regions within the first blood vessel and one or more regions within the second blood vessel; and determining whether the first blood vessel is vascularized. and applying a machine learning algorithm to determine a presence of ischemia in the first blood vessel based at least in part on the plurality of variables determined for the one or more regions in the first blood vessel and the one or more regions in the second blood vessel.

[0365] Embodiment 44: The system of embodiment 43, wherein the plurality of variables includes stenosis.

[0366] Embodiment 45: The system of embodiment 43, wherein the plurality of variables includes plaque volume.

[0367] Embodiment 46: The system of embodiment 45, wherein the plaque volume comprises one or more of total plaque volume, low-density non-calcified plaque volume, non-calcified plaque volume, or calcified plaque volume.

[0368] Embodiment 47: The system described in embodiment 46, wherein the volume of the plaque is determined at least in part based on analyzing the density of one or more pixels corresponding to the plaque in the medical image.

[0369] Embodiment 48: The system of embodiment 47, wherein the density comprises a substance density.

[0370] Embodiment 49: The system described in embodiment 47, wherein the density includes a radial density.

[0371] Embodiment 50: The system of embodiment 49, wherein the low density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about -189 and about 30 Hounsfield units.

[0372] Embodiment 51: The system of embodiment 49, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 31 and about 189 Hounsfield units.

[0373] Embodiment 52: The system of embodiment 49, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 190 and about 350 Hounsfield units.

[0374] Embodiment 53: The system described in embodiment 49, wherein the calcified plaque corresponds to one or more pixels having a radiodensity value between approximately 351 and 2500 Hounsfield units.

[0375] Embodiment 54: The system described in embodiment 43, wherein the medical image includes a computed tomography (CT) image.

[0376] Embodiment 55: The system described in embodiment 43, wherein the medical image is obtained using imaging techniques including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0377] Embodiment 56: The system of embodiment 43, wherein the plurality of variables include one or more of lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, baseline diameter after stenosis, baseline diameter before stenosis, vessel length, luminal volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or volume of low-density plaque.

[0378] Embodiment 57: The system described in embodiment 43, wherein the plurality of blood vessels includes one or more coronary arteries.

[0379] Embodiment 58: The system of embodiment 57, wherein the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0380] Embodiment 59: The system described in embodiment 43, wherein the machine learning algorithm is trained based at least in part on a dataset comprising a plurality of variables and the presence of ischemia derived using invasive fractional flow reserve.

[0381] Embodiment 60: The system described in embodiment 43, wherein the machine learning algorithm is trained based at least in part on a dataset including a plurality of variables and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0382] Embodiment 61: The system described in embodiment 43, wherein the processor is further configured to generate an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular event (MACE) based at least in part on the determination of the presence of ischemia in the first blood vessel.

[0383] Embodiment 62: The system described in embodiment 61, wherein the processor is further configured to generate a graphical representation of the generated assessment of the risk of CAD or MACE.

[0384] Embodiment 63: The system of embodiment 61, wherein the processor is further configured to generate a recommended treatment for the subject based at least in part on the generated assessment of risk of CAD or MACE.

[0385] Embodiment 64: A system for determining the presence of vessel-specific ischemia based at least in part on a plurality of variables derived from analysis of non-invasive medical images, comprising: a non-transitory computer storage medium configured to store at least computer-executable instructions; and one or more computer hardware processors in communication with the first non-transitory computer storage medium, wherein the one or more computer hardware processors are configured to at least: identify one or more lesions in a first vessel and one or more lesions in a second vessel; and identify one or more regions of plaque within the one or more lesions in the first vessel and one or more regions of plaque within the one or more lesions in the second vessel. and applying a machine learning algorithm to determine the presence of ischemia in the first blood vessel based at least in part on the plurality of variables determined for the one or more lesions in the first blood vessel and the one or more lesions in the second blood vessel.

[0386] Embodiment 65: The system of embodiment 64, wherein the plurality of variables includes stenosis.

[0387] Embodiment 66: The system of embodiment 64, wherein the plurality of variables includes plaque volume.

[0388] Embodiment 67: The system of embodiment 66, wherein the plaque volume comprises one or more of total plaque volume, low-density non-calcified plaque volume, non-calcified plaque volume, or calcified plaque volume.

[0389] Embodiment 68: The system described in embodiment 67, wherein the volume of the plaque is determined at least in part based on analyzing the density of one or more pixels corresponding to the plaque in the medical image.

[0390] Embodiment 69: The system of embodiment 67, wherein the density comprises a substance density.

[0391] Embodiment 70: The system described in embodiment 67, wherein the density includes a radial density.

[0392] Embodiment 71: The system of embodiment 70, wherein the low density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about -189 and about 30 Hounsfield units.

[0393] Embodiment 72: The system of embodiment 70, wherein the low-density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 31 and about 189 Hounsfield units.

[0394] Embodiment 73: The system of embodiment 70, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 190 and about 350 Hounsfield units.

[0395] Embodiment 74: The system described in embodiment 70, wherein the calcified plaque corresponds to one or more pixels having a radiodensity value between approximately 351 and 2500 Hounsfield units.

[0396] Embodiment 75: The system described in embodiment 64, wherein the medical image includes a computed tomography (CT) image.

[0397] Embodiment 76: The system described in embodiment 64, wherein the medical image is obtained using imaging techniques including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0398] Embodiment 77: The system of embodiment 64, wherein the plurality of variables include one or more of lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, baseline diameter after stenosis, baseline diameter before stenosis, vessel length, luminal volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or volume of low-density plaque.

[0399] Embodiment 78: The system described in embodiment 64, wherein the plurality of blood vessels includes one or more coronary arteries.

[0400] Embodiment 79: The system of embodiment 78, wherein the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0401] Embodiment 80: The system described in embodiment 64, wherein the machine learning algorithm is trained based at least in part on a dataset comprising a plurality of variables and the presence of ischemia derived using invasive fractional flow reserve.

[0402] Embodiment 81: The system described in embodiment 64, wherein the machine learning algorithm is trained based at least in part on a dataset including a plurality of variables and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0403] Embodiment 82: The system described in embodiment 64, wherein the processor is further configured to generate an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular event (MACE) based at least in part on the determination of the presence of ischemia in the first blood vessel.

[0404] Embodiment 83: The system described in embodiment 82, wherein the processor is further configured to generate a graphical representation of the generated assessment of the risk of CAD or MACE.

[0405] Embodiment 84: The system of embodiment 82, wherein the processor is further configured to generate a recommended treatment for the subject based at least in part on the generated assessment of risk of CAD or MACE.

[0406] Embodiment 85: A non-transitory computer-readable medium configured for determining the presence of vessel-specific ischemia based at least in part on a plurality of variables derived from analysis of non-invasive medical images, the computer-readable medium having program instructions for causing a hardware processor to execute a method, the method comprising: accessing medical images of a subject, the medical images of the subject being non-invasively acquired; analyzing the medical images of the subject to identify a plurality of vessels, the plurality of vessels including a first vessel and a second vessel; identifying one or more regions within the first vessel and one or more regions within the second vessel; and identifying one or more regions within the first vessel and one or more regions within the second vessel. identifying one or more regions of plaque within one or more regions within the first blood vessel; analyzing the one or more regions within the first blood vessel, the one or more regions within the second blood vessel, the one or more regions of plaque within the one or more regions within the first blood vessel, and the one or more regions of plaque within the one or more regions within the second blood vessel to determine a plurality of variables for each of the one or more regions within the first blood vessel and the one or more regions within the second blood vessel; and applying a machine learning algorithm to determine the presence of ischemia in the first blood vessel based at least in part on the plurality of variables determined for the one or more regions within the first blood vessel and the plurality of variables determined for the one or more regions within the second blood vessel.

[0407] Embodiment 86: The non-transitory computer-readable medium of embodiment 85, wherein the plurality of variables includes stenosis.

[0408] Embodiment 87: The non-transitory computer-readable medium of embodiment 85, wherein the plurality of variables comprises plaque volume.

[0409] Embodiment 88: The non-transitory computer-readable medium of embodiment 87, wherein the plaque volume comprises one or more of total plaque volume, low-density non-calcified plaque volume, non-calcified plaque volume, or calcified plaque volume.

[0410] Embodiment 89: A non-transitory computer-readable medium as described in embodiment 88, wherein the volume of the plaque is determined at least in part based on analyzing the density of one or more pixels corresponding to the plaque in the medical image.

[0411] Embodiment 90: The non-transitory computer-readable medium of embodiment 89, wherein the density comprises a material density.

[0412] Embodiment 91: The non-transitory computer-readable medium of embodiment 89, wherein the density comprises a radiometric density.

[0413] Embodiment 92: The non-transitory computer-readable medium of embodiment 91, wherein low-density non-calcified plaque corresponds to one or more pixels with a radiodensity value between about -189 and about 30 Hounsfield units.

[0414] Embodiment 93: A non-transitory computer-readable medium according to embodiment 91, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 31 and about 189 Hounsfield units.

[0415] Embodiment 94: A non-transitory computer-readable medium according to embodiment 91, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 190 and about 350 Hounsfield units.

[0416] Embodiment 95: A non-transitory computer-readable medium according to embodiment 91, wherein the calcified plaque corresponds to one or more pixels having a radiodensity value between approximately 351 and 2500 Hounsfield units.

[0417] Embodiment 96: The non-transitory computer-readable medium of embodiment 85, wherein the medical image comprises a computed tomography (CT) image.

[0418] Embodiment 97: The non-transitory computer-readable medium of embodiment 85, wherein the medical image is obtained using imaging techniques including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0419] Embodiment 98: The non-transitory computer-readable medium of embodiment 85, wherein the plurality of variables comprises one or more of lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, baseline diameter after stenosis, baseline diameter before stenosis, vessel length, luminal volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or volume of low-density plaque.

[0420] Embodiment 99: The non-transitory computer-readable medium of embodiment 85, wherein the plurality of blood vessels includes one or more coronary arteries.

[0421] Embodiment 100: A non-transitory computer-readable medium according to embodiment 99, wherein the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0422] Embodiment 101: A non-transitory computer-readable medium described in embodiment 85, wherein the machine learning algorithm is trained based at least in part on a dataset comprising a plurality of variables and the presence of ischemia derived using invasive fractional flow reserve.

[0423] Embodiment 102: A non-transitory computer-readable medium as described in embodiment 85, wherein the machine learning algorithm is trained based at least in part on a dataset including a plurality of variables and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0424] Embodiment 103: The non-transitory computer-readable medium of embodiment 85, wherein the method further comprises generating, by the computer system, an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular event (MACE) based at least in part on the determination of the presence of ischemia in the first blood vessel.

[0425] Embodiment 104: A non-transitory computer-readable medium described in embodiment 103, wherein the method further comprises generating, by the computer system, a graphical representation of the generated assessment of the risk of CAD or MACE.

[0426] Embodiment 105: The non-transitory computer-readable medium of embodiment 103, wherein the method further comprises generating, by the computer system, a recommended treatment for the subject based at least in part on the generated assessment of risk of CAD or MACE.

[0427] Embodiment 106: A non-transitory computer-readable medium configured for determining the presence of vessel-specific ischemia based at least in part on a plurality of variables derived from analysis of non-invasive medical images, the computer-readable medium having program instructions for causing a hardware processor to execute a method, the method comprising: accessing medical images of a subject, the medical images of the subject being non-invasively acquired; analyzing the medical images of the subject to identify a plurality of vessels, the plurality of vessels including a first vessel and a second vessel; identifying one or more lesions in the first vessel and one or more lesions in the second vessel; and identifying one or more regions of plaque within the one or more lesions in the first vessel. and identifying one or more regions of plaque within the one or more lesions in the second blood vessel; analyzing the one or more lesions in the first blood vessel, the one or more lesions in the second blood vessel, the one or more regions of plaque within the one or more lesions in the first blood vessel, and the one or more regions of plaque within the one or more lesions in the second blood vessel to determine a plurality of variables for each of the one or more lesions in the first blood vessel and the one or more lesions in the second blood vessel; and applying a machine learning algorithm to determine the presence of ischemia in the first blood vessel based at least in part on the plurality of variables determined for the one or more lesions in the first blood vessel and the plurality of variables determined for the one or more lesions in the second blood vessel.

[0428] Embodiment 107: The non-transitory computer-readable medium of embodiment 106, wherein the plurality of variables includes stenosis.

[0429] Embodiment 108: The non-transitory computer-readable medium of embodiment 106, wherein the plurality of variables includes plaque volume.

[0430] Embodiment 109: The non-transitory computer-readable medium of embodiment 108, wherein the plaque volume comprises one or more of total plaque volume, low-density non-calcified plaque volume, non-calcified plaque volume, or calcified plaque volume.

[0431] Embodiment 110: A non-transitory computer-readable medium described in embodiment 109, wherein the volume of the plaque is determined at least in part based on analyzing the density of one or more pixels corresponding to the plaque in the medical image.

[0432] Embodiment 111: The non-transitory computer-readable medium of embodiment 109, wherein the density comprises a material density.

[0433] Embodiment 112: The non-transitory computer-readable medium of embodiment 109, wherein the density includes a radial density.

[0434] Embodiment 113: A non-transitory computer-readable medium according to embodiment 112, wherein the low-density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about -189 and about 30 Hounsfield units.

[0435] Embodiment 114: A non-transitory computer-readable medium according to embodiment 112, wherein the low-density non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 31 and about 189 Hounsfield units.

[0436] Embodiment 115: A non-transitory computer-readable medium according to embodiment 112, wherein the non-calcified plaque corresponds to one or more pixels having a radiodensity value between about 190 and about 350 Hounsfield units.

[0437] Embodiment 116: A non-transitory computer-readable medium described in embodiment 112, wherein the calcified plaque corresponds to one or more pixels having a radiodensity value between approximately 351 and 2500 Hounsfield units.

[0438] Embodiment 117: A non-transitory computer-readable medium as described in embodiment 109, wherein the medical image comprises a computed tomography (CT) image.

[0439] Embodiment 118: A non-transitory computer-readable medium as described in embodiment 109, wherein the medical image is obtained using imaging techniques including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

[0440] Embodiment 119: A non-transitory computer-readable medium as described in embodiment 109, wherein the plurality of variables include one or more of lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, baseline diameter after stenosis, baseline diameter before stenosis, vessel length, luminal volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or volume of low-density plaque.

[0441] Embodiment 120: The non-transitory computer-readable medium of embodiment 109, wherein the plurality of blood vessels includes one or more coronary arteries.

[0442] Embodiment 121: A non-transitory computer-readable medium according to embodiment 120, wherein the one or more coronary arteries include one or more of the left coronary artery (LM), intermediate branch (RI), left anterior descending artery (LAD), diagonal branch 1 (D1), diagonal branch 2 (D2), left circumflex (Cx), obtuse margin 1 (OM1), obtuse margin 2 (OM2), left posterior descending artery (L-PDA), left posterolateral branch (L-PLB), right coronary artery (RCA), right posterior descending artery (R-PDA), or right posterolateral branch (R-PLB).

[0443] Embodiment 122: A non-transitory computer-readable medium described in embodiment 109, wherein the machine learning algorithm is trained based at least in part on a dataset comprising a plurality of variables and the presence of ischemia derived using invasive fractional flow reserve.

[0444] Embodiment 123: A non-transitory computer-readable medium as described in embodiment 109, wherein the machine learning algorithm is trained based at least in part on a dataset including a plurality of variables and the presence of ischemia derived using one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

[0445] Embodiment 124: The non-transitory computer-readable medium of embodiment 109, wherein the method further comprises generating, by the computer system, an assessment of the subject's risk of coronary artery disease (CAD) or major adverse cardiovascular events (MACE) based at least in part on the determination of the presence of ischemia in the first blood vessel.

[0446] Embodiment 125: A non-transitory computer-readable medium described in embodiment 124, wherein the method further comprises generating, by the computer system, a graphical representation of the generated assessment of the risk of CAD or MACE.

[0447] Embodiment 126: The non-transitory computer-readable medium of embodiment 124, wherein the method further comprises generating, by the computer system, a recommended treatment for the subject based at least in part on the generated assessment of risk of CAD or MACE.

[0448] Image-based analysis for cardiac catheterization decisions Disclosed herein are systems, devices, and methods for image-based analysis to determine cardiac catheterization treatment. In particular, in some embodiments, the systems, devices, and methods described herein analyze one or more medical images (e.g., non-invasively acquired medical images) of a patient to identify one or more blood vessels (e.g., coronary arteries) within the images and one or more regions of plaque associated with the blood vessels. The systems, devices, and methods can further analyze the identified one or more blood vessels and one or more regions of plaque to generate multiple image-derived variables associated therewith. The multiple variables can include one or more of: lesion length, remodeling index, percentage of plaque slices, percentage of stenosis area, presence of low-density plaque, percentage of stenosis diameter, presence of positive remodeling, post-stenosis baseline diameter, pre-stenosis baseline diameter, vessel length, lumen volume, number of chronic total occlusions (CTOs), vessel volume, number of stenoses, total plaque volume, number of mild stenoses, or low-density plaque volume.

[0449] The devices, systems, and methods may further include applying a machine learning algorithm to determine the patient's risk of coronary artery disease (CAD) or major adverse cardiovascular event (MACE) based at least in part on the multiple variables. In some embodiments, the machine learning algorithm may be trained at least in part on variables derived from medical images of other subjects at known risk of CAD or MACE. The devices, systems, and methods may then further include determining the patient's need for cardiac catheterization based at least in part on the determined risk of CAD or MACE. In some embodiments, a risk stratification (e.g., low, intermediate, or high) may be determined. In some embodiments, the need for catheterization is determined based on a CAD or MACE risk above a predetermined threshold indicating the need for cardiac catheterization. In some embodiments, the devices, systems, and methods may further generate a graphical representation of the determined need for cardiac catheterization for the patient.

[0450] In some embodiments, the devices, systems, and methods can further determine a type of cardiac catheterization procedure for the patient based at least in part on the determined risk of CAD or MACE and the plurality of variables. The type of cardiac catheterization procedure can include one or more of coronary angiography, right heart catheterization, coronary artery catheterization, pacemaker or defibrillator placement, valve evaluation, pulmonary angiography, shunt evaluation, ventriculography, percutaneous aortic valve replacement, balloon septostomy, stenting, or alcohol septal ablation. In some embodiments, the type of cardiac catheterization procedure for the patient can be determined using a machine learning algorithm trained at least in part on a plurality of image-derived variables derived from medical images of other subjects at known risk of CAD or MACE who have undergone known types of cardiac catheterization procedures.

[0451] In some embodiments, cardiac catheterization can be used for one or more of coronary angiography, right heart catheterization, coronary artery catheterization, pacemaker or defibrillator placement, valve evaluation, pulmonary angiography, shunt evaluation, ventriculography, percutaneous aortic valve replacement, balloon septostomy, stenting, or alcohol septal ablation.

[0452] In some embodiments, systems, devices, and methods for determining cardiac catheterization procedures may be useful in determining whether a patient can be safely discharged from the hospital or whether the patient should be sent to a catheterization lab or emergency room.

[0453] FIG. 13 is a flowchart illustrating exemplary embodiment(s) of a system, device, and method for determining cardiac catheterization therapy. As shown in FIG. 13 , in some embodiments, the system may be configured to access and / or modify one or more medical images at block 1302. In some embodiments, the medical images may include one or more arteries, such as the subject's coronary arteries, carotid arteries, and / or other arteries. In some embodiments, the medical images may be stored in a medical image database 1304. In some embodiments, the medical image database 1304 may be accessed locally by the system and / or may be located remotely and accessed via a network connection. The medical images may include images acquired using one or more modalities, such as CT, dual-energy computed tomography (DECT), spectral CT, photon-counting CT, X-ray, ultrasound, echocardiography, intravascular ultrasound (IVUS), magnetic resonance (MR) imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single-photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS). In some embodiments, the medical images include one or more of contrast-enhanced CT images, non-contrast-enhanced CT images, MR images, and / or images obtained using any of the above-mentioned modalities.

[0454] In some embodiments, the system may be configured to automatically and / or dynamically perform one or more analyses of the medical image as discussed herein. For example, in some embodiments, at block 1306, the system may be configured to identify one or more blood vessels, such as one or more arteries. The one or more arteries may include, among others, a coronary artery, a carotid artery, an aorta, a renal artery, a lower limb artery, an upper limb artery, and / or a cerebral artery. In some embodiments, the one or more coronary arteries include one or more of the left coronary artery (LM), the intermediate branch (RI), the left anterior descending artery (LAD), the diagonal branch 1 (D1), the diagonal branch 2 (D2), the left circumflex (Cx), the obtuse margin 1 (OM1), the obtuse margin 2 (OM2), the left posterior descending artery (L-PDA), the left posterior lateral branch (L-PLB), the right coronary artery (RCA), the right posterior descending artery (R-PDA), or the right posterior lateral branch (R-PLB).

[0455] In some embodiments, the system may be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically identify one or more arteries or coronary arteries using image processing. For example, in some embodiments, one or more AI and / or ML algorithms may be trained using a convolutional neural network (CNN) on a set of medical images in which arteries or coronary arteries have been identified, thereby enabling the AI ​​and / or ML algorithms to automatically identify arteries or coronary arteries directly from the medical images. In some embodiments, the arteries or coronary arteries are identified by size and / or location.

[0456] In some embodiments, at block 1308, the system may be configured to identify one or more regions of plaque within the medical image. In some embodiments, the system may be configured to utilize one or more AI and / or ML algorithms to automatically and / or dynamically identify one or more regions of plaque using image processing. For example, in some embodiments, one or more AI and / or ML algorithms may be trained using a convolutional neural network (CNN) on a set of medical images in which regions of plaque have been identified, thereby allowing the AI ​​and / or ML algorithms to automatically identify regions of plaque directly from the medical images. In some embodiments, the system is configured to identify the vessel wall and the lumen wall and classify everything between the vessel wall and the lumen wall as plaque.

[0457] In some embodiments, the system may be configured to analyze and / or characterize one or more regions of plaque based on density. For example, in some embodiments, the system may be configured to analyze and / or characterize one or more regions of plaque based on absolute density and / or relative density and / or radiometric density. In some embodiments, the system may be configured to classify a region of plaque as one of low-density non-calcified plaque, non-calcified plaque, and calcified plaque using any one or more processes and / or features described herein.

[0458] In some embodiments, the system may be configured to analyze and / or characterize one or more regions of plaque based on one or more distances. For example, as described herein, in some embodiments, the system may be configured to determine the distance between low-density non-calcified plaque and the lumen wall and / or vessel wall. In some embodiments, the proximity of low-density non-calcified plaque to the lumen wall may indicate a high-risk plaque and / or CAD. Conversely, in some embodiments, the location of low-density non-calcified plaque farther from the lumen wall may indicate a lower risk. In some embodiments, the system may be configured to utilize one or more predetermined thresholds in determining risk factors associated with the...

Claims

1. 1. A computer-implemented method for determining a probability threshold for ischemia of a coronary artery based at least in part on a percentage of stenosis generated from an image-based analysis, comprising: accessing, by a computer system, a first medical image of a subject, the first medical image including a region of a coronary artery of the subject; analyzing, by the computer system, the first medical image using image segmentation to identify a region of the coronary artery; identifying, by the computer system, one or more regions of plaque within the region of the coronary artery, the one or more regions of plaque being identified based at least in part on a density of one or more pixels within the first medical image corresponding to the one or more regions of plaque; determining, with the computer system, a percentage of one or more stenoses present in a region of the coronary artery resulting from the one or more regions of plaque, the percentage of one or more stenoses being determined based at least in part on interpolating a luminal volume or diameter of the region of the coronary artery in the absence of the one or more regions of plaque; determining, by the computer system, a probability threshold of ischemia for the region of the coronary artery based at least in part on the percentage of the one or more stenoses present in the region of the coronary artery and a plurality of reference values ​​of percentage of stenosis for which the presence or absence of ischemia is known, derived from a plurality of other subjects, wherein the probability threshold of ischemia comprises a statistical likelihood that the region of the coronary artery is ischemic, and wherein a probability threshold of ischemia for the region of the coronary artery higher than a predetermined threshold indicates a need for further evaluation of the subject for ischemia; The computer-implemented method, wherein the computer system comprises a computer processor and an electronic storage medium.

2. The computer-implemented method of claim 1 , further comprising determining a fractional flow reserve for the region of the coronary artery if the probability threshold of ischemia for the region of the coronary artery is higher than the predetermined threshold.

3. determining the fractional flow reserve, accessing, by the computer system, a second medical image, the second medical image including a region of the coronary artery of the subject; and determining, by the computer system, the fractional flow reserve of the region of the coronary artery using computational fluid dynamics.

4. 3. The computer-implemented method of claim 2, wherein the fractional flow reserve is determined based on one or more of invasive fractional flow reserve, computed tomography (CT) fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

5. The computer-implemented method of claim 1 , wherein the further assessment of the subject's ischemia comprises invasive fractional flow reserve.

6. The computer-implemented method of claim 1 , wherein the further assessment of the subject's ischemia includes computed tomography (CT) fractional flow reserve.

7. 10. The computer-implemented method of claim 1, wherein the further assessment of the subject's ischemia includes one or more of CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

8. 2. The computer-implemented method of claim 1, wherein the probability threshold of ischemia for the region of the coronary artery is determined using a machine learning algorithm trained on the plurality of reference values ​​of percentage stenosis with known presence or absence of ischemia derived from the plurality of other subjects.

9. 9. The computer-implemented method of claim 8, wherein the presence or absence of ischemia is derived from the plurality of other subjects using one or more of invasive fractional flow reserve, CT fractional flow reserve, calculated fractional flow reserve, virtual fractional flow reserve, vascular fractional flow reserve, or quantitative flow ratio.

10. The computer-implemented method of claim 1 , wherein the probability threshold of ischemia for the region of the coronary artery is determined based at least in part on a percentage of all stenoses identified within the region of the coronary artery.

11. 10. The computer-implemented method of claim 1, wherein the probability threshold of ischemia for the region of the coronary artery is determined based at least in part on a weighted measure of a percentage of all stenoses identified within the region of the coronary artery.

12. The computer-implemented method of claim 1 , wherein the probability threshold of ischemia for the region of the coronary artery constitutes a binary output.

13. The computer-implemented method of claim 1 , wherein the probability threshold of ischemia for the region of the coronary artery constitutes a continuous scale output.

14. The computer-implemented method of claim 1 , wherein the density comprises a radial density.

15. The computer-implemented method of claim 1 , wherein the density comprises a material density.

16. The computer-implemented method of claim 1 , wherein the first medical image is acquired using CT.

17. 10. The computer-implemented method of claim 1, wherein the first medical image is acquired using an imaging modality including one or more of CT, X-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron emission tomography (PET), single photon emission computed tomography (SPECT), or near-infrared spectroscopy (NIRS).

18. 1. A non-transitory computer-readable medium configured for determining a probability threshold for ischemia of a coronary artery based at least in part on a percentage of stenosis generated from an image-based analysis, the computer-readable medium having program instructions for causing a hardware processor to execute a method, the method comprising: accessing a first medical image of a subject, the first medical image including a region of a coronary artery of the subject; analyzing the first medical image using image segmentation to identify regions of the coronary arteries; identifying one or more regions of plaque within the region of the coronary artery, the one or more regions of plaque being identified based at least in part on a density of one or more pixels within the first medical image corresponding to the one or more regions of plaque; determining a percentage of one or more stenoses present in a region of the coronary artery resulting from the one or more regions of plaque, the percentage of one or more stenoses being determined based at least in part on interpolating a luminal volume or diameter of the region of the coronary artery in the absence of the one or more regions of plaque; determining a probability threshold of ischemia for the region of the coronary artery based at least in part on the percentage of the one or more stenoses present in the region of the coronary artery and a plurality of reference values ​​of percentage of stenosis for which the presence or absence of ischemia is known derived from a plurality of other subjects, wherein the probability threshold of ischemia comprises a statistical likelihood that the region of the coronary artery is ischemic, and wherein a probability threshold of ischemia for the region of the coronary artery higher than a predetermined threshold indicates a need for further evaluation of the subject for ischemia.

19. 20. The non-transitory computer-readable medium configured as recited in claim 18, further comprising determining a fractional flow reserve for the region of the coronary artery if the probability threshold of ischemia for the region of the coronary artery is higher than the predetermined threshold.

20. determining the fractional flow reserve, accessing a second medical image, the second medical image including a region of the coronary artery of the subject; and determining the fractional flow reserve of the region of the coronary artery using computational fluid dynamics.

Images