Angiography derived calcium modification recommendation tool

Abstract

Example techniques, devices and systems are disclosed. An example medical system includes memory configured to store angiography data of a patient. The medical system includes processing circuitry communicatively coupled to the memory and configured to obtain the angiography data. The medical system is configured to determine, based on the angiography data, one or more calcium characteristics of a lesion. The medical system is configured to determine, based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent. The medical system is configured to output an indication of the determined likelihood the stent will provide the at least the level of expansion.

Description

ANGIOGRAPHY DERIVED CALCIUM MODIFICATIONRECOMMENDATION TOOL

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 631,792, filed April 9, 2024, the entire content of each application is incorporated herein by reference.TECHNICAL FIELD

[0002] This disclosure relates to the imaging such as imaging used during a medical procedure.BACKGROUND

[0003] During a medical procedure, a clinician may use an imaging system to be able to visualize internal anatomy of a patient. Such an imaging system may display anatomy, medical instruments, or the like, and may be used to diagnose a patient condition or assist in guiding a clinician in navigating a device inside a patient, such as moving a medical instrument to an intended location inside the patient. Imaging systems may use sensors to capture image data which may be displayed during the medical procedure. Imaging systems include angiography systems, computed tomography (CT) scan systems (including coronary computed tomography angiography (CCTA) systems), fluoroscopic systems (e.g., isocentric C-arm fluoroscopic systems), intravascular ultrasound (IVUS) systems, other ultrasound imaging systems, optical coherence tomography (OCT) fractional flow reserve (FFR) systems, magnetic resonance imaging (MRI) systems, positron emission tomography (PET) systems, as well as other imaging systems.

[0004] Sometimes a patient may have lesions including plaque, such as calcium deposits, in an artery. Medical procedures may be used to treat such lesions, such as a percutaneous coronary intervention (PCI), which may attempt to restore blood flow through a vessel that is occluded by the lesion.SUMMARY

[0005] When planning a PCI of a lesion, such as a calcified lesion, a clinician may expect a certain amount of stent expansion after the stent is implanted to improve blood flow in the occluded vessel. However, in some examples, a stent may not actually expand adequately after implantation to meet the clinician’s expectations and the clinician may wish they had spent more time doing lesion preparation prior to implanting the stent. Such preparation could include vessel modification techniques, such as using a high-pressure balloon, cutting balloon, intravascular lithotripsy, atherectomy, or the like. The techniques of this disclosure include using an algorithm to identifylesions that are likely to require lesion preparation to achieve best, optimal, or acceptable post-PCI results, based on pre-PCI (e.g., diagnostic angiogram) angiography data.

[0006] Calcified lesions are generally very difficult to treat. The techniques of this disclosure may assist in the planning of PCI of a calcified lesion, potentially reducing the need for intravascular imaging and improving outcomes by reducing instances of under-expanded stents due to lack of adequate lesion preparation.

[0007] This disclosure describes techniques, systems, and devices for providing information to a clinician regarding estimated chances of success for stent expansion for a PCI treating a calcified lesion. A system may generate a 3D model including a coronary artery based on diagnostic angiogram data and overlay lesion calcium characteristics onto the 3D model. Diagnostic angiogram data may be angiogram data captured or generated during a diagnostic medical procedure. In some examples, the system may execute a machine learning model to determine such calcium characteristics. The system may also determine and provide predictions of chances of success of a plurality of potential PCI treatments to a clinician to inform the clinician’s planning of the PCI procedure to treat the calcified lesion. In some examples, the system may execute one or more machine learning models to determine such chances of success.

[0008] Additionally, it may be desirable to determine an atherosclerotic plaque phenotype associated with a patient, as such information may be used by a clinician when determining treatment of the patient. As such, the techniques of this disclosure may include determining an atherosclerotic plaque phenotype associated with a patient based on FFR data derived from angiography data and pullback pressure gradient (PPG) data derived from angiography data.

[0009] In one example, the disclosure describes a medical system comprising: memory configured to store angiography data of a patient; and processing circuitry communicatively coupled to the memory, the processing circuitry being configured to: obtain the angiography data; determine, based on the angiography data, one or more calcium characteristics of a lesion; determine, based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent; and output an indication of the determined likelihood the stent will provide the at least the level of expansion.

[0010] In another example, the disclosure describes a method comprising: obtaining, by processing circuitry, angiography data of a patient; determining, by the processing circuitry and based on the angiography data, one or more calcium characteristics of a lesion; determining, by the processing circuitry and based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calciummodification of the lesion prior to implanting the stent; and outputting, by the processing circuitry, an indication of the determined likelihood the stent will provide the at least the level of expansion.

[0011] In yet another example, the disclosure describes a non-transitory computer readable medium comprising instructions, which, when executed, cause processing circuitry to: obtain angiography data; determine, based on the angiography data, one or more calcium characteristics of a lesion; determine, based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent; and output an indication of the determined likelihood the stent will provide the at least the level of expansion.

[0012] These and other aspects of the present disclosure will be apparent from the detailed description below. In no event, however, should the above summaries be construed as limitations on the claimed subject matter, which subject matter is defined solely by the attached claims.

[0013] This summary is intended to provide an overview of the subject matter described in this disclosure. It is not intended to provide an exclusive or exhaustive explanation of the apparatus and methods described in detail within the accompanying drawings and description below. Further details of one or more examples are set forth in the accompanying drawings and the description below.BRIEF DESCRIPTION OF DRAWINGS

[0014] FIG. 1 is a schematic perspective view of one example of a system for performing a medical procedure according to one or more aspects of this disclosure.

[0015] FIG. 2 is a schematic view of one example of a computing system of the system of FIG. 1.

[0016] FIG. 3 is a conceptual diagram illustrating a cross-section of an example culprit vessel according to one or more aspects of this disclosure.

[0017] FIGS. 4A-4B are conceptual diagrams illustrating example calcium characteristics according to one or more aspects of this disclosure.

[0018] FIG. 5 is a flow diagram illustrating example calcium modification recommendation techniques according to one or more aspects of this disclosure.

[0019] FIG. 6 is a conceptual diagram illustrating an example machine learning model according to one or more aspects of this disclosure.

[0020] FIG. 7 is a conceptual diagram illustrating an example training process for a machine learning model according to one or more aspects of this disclosure.

[0021] FIG. 8 is a conceptual diagram illustrating another example training process for a machine learning model according to one or more aspects of this disclosure.DETAILED DESCRIPTION

[0022] As discussed above, when planning a PCI of a calcified lesion, a clinician may expect an amount of stent expansion after the PCI procedure. However, in some examples, a stent may not expand adequately after the PCI to meet the clinician’ s expectations and the clinician may wish they had spent more time doing lesion preparation prior to implanting the stent.

[0023] With a calcified lesion, it may be beneficial for a clinician to utilize intravascular imaging, such as OCT or IVUS, to help better observe the calcium deposit(s) and help with planning the PCI procedure. For example, measuring calcium with OCT or IVUS may help identify calcific lesions that would benefit from plaque modification before stent implantation and to determine which type of plaque modification treatment may be most desirable. However, OCT and IVUS are more invasive than typical angiography and not all clinicians employ such techniques, particularly during a diagnostic procedure.

[0024] As such it may be desirable to provide techniques, such as the techniques of this disclosure, for determining calcium characteristics and / or providing a clinician with a plurality of treatment options and corresponding chances of success based on diagnostic angiography data. As such, a clinician may perform a diagnostic angiography on a patient without the use of more invasive imaging techniques on the patient, such as OCT or IVUS, and be informed of the calcium characteristics of one or more lesion in arteries of the patient. In the event that a lesion is calcified to the extent that the implantation of a stent is less likely than desired to permit stent expansion without treatment of the calcified lesion, the system may inform the clinician that treatment may be desirable. In some examples, the system may inform the clinician of a plurality of treatment options and corresponding chances of successfully treating the lesion such that the desired stent expansion can be achieved.

[0025] A recent study showed a correlation between wire based FFR, quantified by pullback pressure gradient (PPG), with the associated atherosclerotic plaque phenotypes. The atherosclerotic plaque phenotypes were identified using coronary CT angiography and OCT intravascular imaging. Knowledge of the atherosclerotic plaque phenotype associated with a patient may inform clinician PCI planning. As such, according to the techniques of this disclosure, a machine learning model may utilize angiography data to derive an FFR and / or PPG. The machine learning model may determine the atherosclerosis plaque phenotype of a lesion using the angiography derived FFR, angiography derived PPG, coronary CT angiography, and / or OCTintravascular imaging and may augment the angiography data with the determined atherosclerotic plaque phenotype. For example, an indication of the atherosclerotic plaque phenotype may be overlayed on the angiography data and / or a 3D model representation of the angiography data.

[0026] FIG. 1 is a schematic perspective view of one example of a system for performing a medical procedure according to one or more aspects of this disclosure. System 100 includes a display device 110, a table 120, an imager 140, and a computing device 150. System 100 may be an example of a system for use in an emergency room or a Catheterization laboratory (Cath lab). In some examples, system 100 may include other devices, not shown for simplicity purposes. In some examples, system 100 may also include server 160, which may be co-located with the other devices of system 100 or may be located elsewhere. In some examples, server 160 represents one or more servers located in a cloud-computing environment.

[0027] System 100 may be used during a medical procedure, such as a diagnostic medical procedure and / or a PCI. During a diagnostic medical procedure, a clinician may collect and / or view diagnostic angiography data. During a PCI, a clinician may perform one or more procedures to address lesions, which in some examples, may be found during a diagnostic medical procedure. Some of such procedures may include implantation of a stent in the area of a lesion to increase the diameter of the affected vessel in the area of the lesion to improve blood flow through the vessel.

[0028] Sometimes a stent may not open the vessel as much as desired after implantation, for example, when the lesion is calcified. Calcification of a lesion may increase the resistance and / or pressure a stent may experience from the vessel which may impede the expansion of the vessel with the implanted stent. There are calcium modification treatments or procedures that a clinician may perform during the PCI, prior to implantation of the stent, on a calcified lesion which may improve the chances that the stent will expand as desired once implanted. However, not all calcium modification treatments may be equally effective for given calcium characteristics of a particular lesion. If no treatment is performed prior to implantation of the stent, or if a selected calcium modification treatment is not as effective as thought, there is a chance that the stent may not expand as desired after implantation. System 100 may be used to determine whether a stent is likely to expand as desired once implanted in a vessel having a calcified lesion and / or to determine and present, to a clinician, a plurality of treatment options with corresponding chances of success to aid the clinician in planning and performing the PCI procedure such that the implanted stent may expand as desired.

[0029] Computing device 150 may include, for example, an off-the-shelf device such as a laptop computer, desktop computer, tablet computer, smart phone, or other similar device or may include a specific purpose device. Computing device 150 may perform various control functions with respect to imager 140. In some examples, computing device 150 may include a guidanceworkstation. Computing device 150 may control the operation of imager 140 and receive the output of imager 140 and may receive angiography data from imager 140. Computing device 150 may execute one or more machine learning models to determine calcium characteristics of a lesion, determine whether a stent is likely to expand as desired once implanted true lumen, and / or determine a plurality of treatment options and corresponding likelihoods of success to prepare the lesion for the stent such that the stent expands as desired once implanted.

[0030] Display device 110 may be configured to output instructions, images, and messages relating to the medical procedure(s). For example, display device 110 may display angiography data obtained through imager 140 and / or a representation of the navigation path. Table 120 may be, for example, an operating table or other table suitable for use during a medical procedure.

[0031] In the example of FIG. 1, imager 140, such as an angiography imager (or other imaging device), may be used to image relevant portions of the patient’s anatomy during a medical procedure to visualize the anatomy, characteristics and locations of lesions or other issues inside the patient’s body through the generation of imaging data. While described herein primarily as an angiography imager, imager 140 may be any type of imaging device, such as an angiography device, a fluoroscopy device, a CT device, a CCTA device, an IVUS device, an OCT device, an OCT - FFR device, an MRI device, a PET device, an ultrasound device, or the like. In some examples, imager 140 may represent more than one imaging device, such as a plurality of any of the aforementioned devices.

[0032] Imager 140 may image a region of interest in the patient’s body. The particular region of interest may be dependent on anatomy, the medical procedure, patient symptoms, and / or the like. For example, when performing a cardiac medical procedure, a portion of the vasculature and / or the heart may be within the region of interest.

[0033] Computing device 150 may be communicatively coupled to imager 140, display device 110 and / or server 160, for example, by wired, optical, or wireless communications. Server 160 may be a hospital server which may or may not be located in an emergency room or Cath lab of a hospital, a cloud-based server(s), or the like. Server 160 may be configured to store patient imaging data (such as angiography data), electronic healthcare or medical records, or the like. In some examples, server 160 may be configured to execute the machine learning model(s) of this disclosure and / or perform one or more of, or a portion of one or more of, the determinations associated therewith.

[0034] Any of, or any combination of, computing device 150, imager 140, and / or server 160 may include one or more machine learning model(s). For example, computing device 150, imager 140, and / or server 160 may obtain angiography data, e.g., via imager 140. Computing device 150, imager 140, and / or server 160 may determine, based on the angiography data, one or more calciumcharacteristics of a lesion. For example, computing device 150, imager 140, and / or server 160 may determine the one or more calcium characteristics of the lesion by executing a machine learning model. Computing device 150, imager 140, and / or server 160 may determine a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent. Computing device 150, imager 140, and / or server 160 may output, based on the determined likelihood that the stent will provide at least the level of expansion, an indication of the determined likelihood the stent will provide the at least the level of expansion.

[0035] In some examples, computing device 150, imager 140, and / or server 160 may determine FFR data based on the angiography data and PPG data based on the angiography data. In some examples, computing device 150, imager 140, and / or server 160 may determine an atherosclerotic plaque phenotype of the patient based on the FFR data and the PPG data and output an indication of the atherosclerotic plaque phenotype for display.

[0036] By determining a likelihood of whether a stent will expand as desired, and suggesting a plurality of treatment options and corresponding chances of success, system 100 may assist clinicians in more effectively planning and treating lesions, such as calcified lesions during a PCI medical procedure. As such, the techniques of this disclosure may improve patient outcomes and / or medical facility efficiency, reducing a need for a repeated PCI to address an under-expanded, implanted stent, so as to better restore blood flow through a vessel of a patient. Additionally, providing a clinician with the atherosclerotic plaque phenotype associated with a patient may aid the clinician in planning the PCI procedure and / or other treatment of the patient.

[0037] FIG. 2 is a schematic view of one example of a computing device 150 of system 100 of FIG. 1. Computing device 150 may include a workstation, a desktop computer, a laptop computer, a smart phone, a tablet, a dedicated computing device, or any other computing device capable of performing the techniques of this disclosure.

[0038] Computing device 150 may be configured to perform processing, control and other functions associated with imager 140. As shown in FIG. 2, computing device 150 may represent multiple instances of computing devices, each of which may be associated with imager 140. Computing device 150 may include, for example, a memory 202, processing circuitry 204, a display 206, a network interface 208, input device(s) 210, and / or output device(s) 212, each of which may represent any of multiple instances of such a device within the computing system, for ease of description.

[0039] While processing circuitry 204 appears in computing device 150 in FIG. 2, in some examples, features attributed to processing circuitry 204 may be performed by processing circuitry of any of computing device 150, imager 140, or server 160, or combinations thereof. In someexamples, one or more processors associated with processing circuitry 204 in computing system may be distributed and shared across any combination of computing device 150, imager 140, and server 160. Computing device 150 may be used to perform any of the techniques described in this disclosure, and may form all or part of devices or systems configured to perform such techniques, alone or in conjunction with other components, such as components of computing device 150, imager 140, server 160, or a system including any or all of such systems / devices.

[0040] Memory 202 of computing device 150 includes any non-transitory computer-readable storage media for storing data or software that is executable by processing circuitry 204 and that controls the operation of computing device 150 and / or imager 140, as applicable. In one or more examples, memory 202 may include one or more solid-state storage devices such as flash memory chips. In one or more examples, memory 202 may include one or more mass storage devices connected to the processing circuitry 204 through a mass storage controller (not shown) and a communications bus (not shown).

[0041] Although the description of computer-readable media herein refers to a solid-state storage, it should be appreciated by those skilled in the art that computer-readable storage media may be any available media that may be accessed by the processing circuitry 204. That is, computer readable storage media includes non-transitory, volatile and non-volatile, removable and nonremovable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. For example, computer-readable storage media includes RAM, ROM, EPROM, EEPROM, flash memory and / or other solid state memory technology, CD-ROM, DVD, Blu-Ray and / or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage and / or other magnetic storage devices, and / or any other medium that may be used to store the desired information and that may be accessed by computing device 150. In one or more examples, computer-readable storage media may be stored in the cloud or remote storage and accessed using any suitable technique or techniques through at least one of a wired or wireless connection.

[0042] Memory 202 may store angiography data 214, phenotype data 216, calcium characteristics 226, 3D model 228, suggestions 230, and threshold(s) 232. Angiography data 214 may include a plurality of images obtained, for example, from imager 140 during a medical procedure, such as a diagnostic medical procedure. As such, in some examples, angiography data 214 may be referred to as diagnostic angiography data. In some examples, angiography data 214 may include images obtained during a prior PCI medical procedure. In some examples, angiography data 214 may include images obtained during a current PCI medical procedure. During a medical procedure, while the medical procedure is proceeding in time, additional angiography data may be obtained from imager 140 and be stored in angiography data 214. Suchangiography data may be displayed via display 206 and / or display device 110 and may be used by a clinician when performing a treatment and / or navigating a medical instrument through anatomy of a patient. Memory 202 may also store applications 217, which may include user interface 218 and / or one or more machine learning model(s) 222. In some examples, rather than, or in addition to, being located on computing device 150, machine learning model(s) 222 may be located on server 160 of FIG. 1, such as in a cloud computing environment. In such examples, machine learning model(s) 222 may be trained with new data after each use of machine learning model(s) 222 occurs, for example, on different patients across one or more Cath labs, which may improve an accuracy of estimations of machine learning model(s) 222.

[0043] Angiography data 214 may be generated by imager 140 of anatomy of the patient and obtained by computing device 150 via network interface 208 which may be communicatively coupled to imager 140. In some examples, imager 140 may generate other types of imaging data, such as when imager 140 represents more than one imaging device. For example, imager 140 may generate OCT data and / or IVUS data, which may be used to train machine learning model(s) 222.

[0044] For example, angiography data 214 may be captured by imager 140 (FIG. 1). Processing circuitry 204 may obtain angiography data 214 from imager 140 and store angiography data 214 in memory 202. Processing circuitry 204 may execute user interface 218 so as to cause display 206 (and / or display device 110 of FIG. 1) to present user interface 218 to one or more clinicians performing the medical procedure. User interface 218 may display angiography data 214.

[0045] For example, processing circuitry 204 may generate a 3D model 228 (e.g., 3D reconstruction) of the coronary artery based on diagnostic angiography data. For example, processing circuitry 204 may generate 3D model 228 based on angiography data 214. In some examples, processing circuitry 204 may generate 3D model 228 during the diagnostic angiogram procedure. In some examples, processing circuitry 204 may generate 3D model 228 after the diagnostic angiogram. 3D model 228 may include a 3D model of one or more vessels (or portions thereof) of a patient. In some examples, processing circuitry 204 may execute machine learning model(s) 222 to generate 3D model 228. 3D model 228 may include a 3D model of the anatomy of the patient in at least an area proximate to a lesion. For example, as part of generating 3D model 228, processing circuitry 204 may execute one or more machine learning model(s) 222 to determine sizes (e.g., diameters, lengths, etc.), locations, and / or other characteristics of anatomy of the patient based on angiography data 214.

[0046] Processing circuitry 204 may execute one or more machine learning model(s) 222 to determine calcium characteristics 226 based on actigraphy data 214. Calcium characteristics 226 may include various characteristics of calcium deposits of a calcified lesion. For example, calciumcharacteristics 226 may include a depth, angle, length, density, hardness, and / or the like of calcium of a calcified lesion. Processing circuitry 204 may include calcium characteristics 226 in 3D model 228, such that when a clinician views 3D model 228, for example, on display 206 or display device 110, the clinician may see a representation of calcium characteristics 226. This may aid the clinician in determining whether to perform one or more calcium modification treatments on the lesion prior to implanting a stent, performing such treatments, and / or implanting the stent.

[0047] In some examples, processing circuitry 204 may overlay calcium characteristics 226 onto 3D model 228. Processing circuitry 204 may generate calcium characteristics 226, for example, based on diagnostic angiography data. For example, processing circuitry 204 may execute machine learning model(s) 222 to determine calcium characteristics 226 and / or other lesion characteristics based on angiography data 214. Machine learning model(s) 222 may use, as a ground truth, OCT data and / or IVUS data and the post-PCI angiogram data, for example, from a plurality of PCI procedures.

[0048] In some examples, a clinician may determine a particular size or type of stent to use to treat a lesion. For example, a clinician may determine the particular size of the stent in an attempt to match a diameter of a vessel and to surpass the length of the lesion. In some examples, processing circuitry 204 may execute one or more of machine learning model(s) 222 to determine the particular size or type of stent to use based on angiography data 214, 3D model 228, and / or calcium characteristics 226. The clinician or processing circuitry 204 executing one or more of machine learning model(s) 222 may select (or suggest) a specific stent to use during the PCI based on the determined particular size.

[0049] Processing circuitry 204 may determine how likely the selected stent is to expand at least as much as desired when implanted in a patient based on angiography data 214, without the performance of a calcium modification treatment or procedure. The desired amount of expansion may be programmable by a clinician, or may be determined by processing circuitry 204 executing one or more of machine learning model(s) 222 based on angiography data 214, 3D model 228, and / or calcium characteristics 226. For example, if the diameter of the vessel is 4mm outside of the area of the lesion, a 4 mm stent may be selected, and a desired amount of expansion may be at least 3.8 mm.

[0050] Processing circuitry 204 may execute one or more of machine learning model(s) 222 and determine a likelihood that the stent expands at least as much as desired, without the performance of a calcium modification treatment. Processing circuitry 204 may compare the determined likelihood to a calcium modification threshold (of threshold(s) 232) to determine whether the determined likelihood satisfies or meets the calcium modification threshold. As used herein, a threshold may be satisfied or met if the information to which it is compared is greaterthan, greater than or equal to, less than, or less than or equal to, depending on the information and / or the threshold. For example, where the determined likelihood that the stent expands at least as much as desired is a number where a higher value indicates a higher likelihood, then the calcium modification threshold may be satisfied if the determined likelihood is greater than (or greater than or equal to) the calcium modification threshold. In such cases, the calcium modification threshold may not be satisfied if the determined likelihood is less than or equal to (or less than) the calcium modification threshold.

[0051] In some examples, the calcium modification threshold may be a programmable threshold, permitting a clinician to set a level for which the clinician desires to differentiate between cases in which the clinician would like to perform calcium modification treatments and cases in which the clinician would like to forego performing calcium modification treatments and instead perform the stent implantation without performing a calcium modification treatment. In some examples, the calcium modification threshold may be determined by processing circuitry 204 executing one or more of machine learning model(s) 222 to balance risk a stent not expanding as much as desired against risk(s) inherent in performing calcium modification treatments.

[0052] In the instances where the determined likelihood that the stent expands as much as desired does not satisfy the calcium modification threshold, processing circuitry 204 may executed one or more of machine learning model(s) 222 to determine suggestions 230. Suggestions 230 may include a plurality of calcium modification treatments and corresponding likelihoods of success. For example, suggestions 230 may include an indication that direct stenting has a 21% probability of success, high pressure balloon has a 45% probability of success, atherectomy and non-compliant ballon has a 78% probability of success, and lithotripsy has an 82% probability of success.

[0053] For example, a clinician may perform diagnostic angiography. For example, the diagnostic angiography may include 3 different angiographs at a specific angulation apart which processing circuitry 204 may be used to create 3D model 228. The physiology of vessels, including any calcium deposits, may be derived from 3D model 228. For example, calcium is visible in angiography data, especially moderate and severe calcium deposits.

[0054] Processing circuitry 204 may execute machine learning model(s) 222 which may overlay an indication of calcium characteristics 226 onto the displayed angiogram data providing an indicator, such as a color-coded indicator (or other visible indicator), to assist the clinician in making the decision to prepare the lesion with one or more calcium modification treatments. In some examples, processing circuitry 204 may use pre-defined thresholds to determine when to suggest a calcium modification preparation using one or more medical devices.

[0055] For example, processing circuitry 204 executing machine learning model(s) 222 may suggest a course of action for the clinician. For example, processing circuitry 204 executingmachine learning model(s) 222 may estimate a respective probability of success of a PCI for a patient using a plurality of different calcium modification strategies. Processing circuitry 204 may store such information in suggestions 230. Processing circuitry 204 may control display 206 and / or display device 110 to display suggestions 230. For example, processing circuitry 204 may control display 206 to display an estimated success rate for each of a plurality of different calcium modification strategies. Such information presented to the clinician may inform the clinician treatment strategy decision.

[0056] In some examples, computing device 150 does not include a calcium modification threshold. In such cases, processing circuitry 204 may execute one or more of machine learning model(s) 222 to generate suggestions 230 without regard to the likelihood that the stent will expand as desired.

[0057] Memory 202 may also store one or more machine learning model(s) 222 and user interface 218. Machine learning model(s) 222 may be configured to, when executed by processing circuitry 204, determine 3D model 228, determine calcium characteristics 226, whether a stent is likely to expand as desired, and / or to determine suggestions 230. By determining calcium characteristics, whether the stent is likely to expand, and determining and providing a plurality of suggestions for calcium treatment and corresponding likelihoods of success, the techniques of this disclosure may provide for a greater chance of success of an implanted stent.

[0058] In some examples, machine learning model(s) 222 may be trained using pre-treatment calcium scores and / or assessments; pre-PCI angiography data along with OCT data, IVUS data, FFR data, and / or other physiology assessment; decisions to treat or not to treat; thresholds to treat; and / or the like. If a clinician does treat a calcification, machine learning model(s) 222 may record details and feed back information into the training data set for further accuracy improvements. Such information may include post-PCI angiography data which may be used to measures success of the treatment. As such, the training data set may include pre-PCI, treatment, and post-PCI success metrics to improve machine learning model(s) 222 overtime after each case. Alternatively, machine learning model(s) 222 may not be further trained after each case, but may be relatively static after initial training. In such a case, an update may be pushed to machine learning model(s) 222, for example, from server 160, in a decided cadence (e.g., every 3 months) using the recorded cases as new training data.

[0059] In some examples, machine learning model(s) 222 may include a plurality of machine learning models. In some examples, each respective threshold may be used by a respective machine learning model of the plurality of machine learning model(s). For example, one machine learning model may analyze thickness and compare thickness to a thickness threshold, another machinelearning model may analyze angle and compare the angle to a vessel arc threshold, and yet another machine learning model may analyze length and compare the length to a length threshold.

[0060] Machine learning model(s) 222, which may include a neural network or other type of machine learning model, may be trained based on angiography data of medical procedures, such as diagnostic and / or interventional medical procedures, and corresponding IVUS data and / or OCT data, and / or the like. For example, during a medical procedure, an IVUS sensor and / or an OCT sensor may actually be located within a culprit vessel and the data generated by the IVUS sensor and / or the OCT sensor may be used to determine calcium characteristics of a lesion. Such procedures may also include angiography data, which, when correlated with the IVUS and / or OCT data, may provide some insight into the calcium characteristics of the lesion. In another example, during a post- PCI medical procedure, the data generated by the IVUS sensor and / or the OCT sensor may show an actual expansion of the vessel or stent. In the case where no calcium modification procedure was performed prior to implantation of the stent, such information, when correlated with the calcium characteristics, may be indicative of the likelihood the stent will expand as desired. In the case where one or more calcium modification procedures were performed prior to implantation of the stent, such information, when correlated with the any calcium modification procedure(s) and the calcium characteristics, may be indicative of the likelihood of a successful stent implantation after such calcium modification procedure(s).

[0061] By training machine learning model(s) 222 using the angiography data of such procedures, and OCT data and / or IVUS data, machine learning model(s) 222 may be trained to suggest when to perform calcium modification procedures and to suggest a likelihood of success of such procedure(s) using angiography data 214 alone, thus reducing or eliminating the need for OCT and / or IVUS imaging during diagnostic medical procedures.

[0062] In some examples, machine learning model(s) 222 may be built utilizing ground truth data such as angiography data, OCT data and / or IVUS data, and post-PCI angiogram data from a plurality of PCI procedures. The ground truth data may be utilized to train one or more of machine learning model(s) 222 to predict calcium characteristics of a lesion, the likelihood that a stent will expand as desired when implanted given the calcium characteristics of the lesion if the lesion is not treated with any calcium modification treatments, and / or the probability that a stent will expand as desired if the lesion is treated using one or more given calcium modification techniques. Ground truth data may also include wire-based FFR, PPG data, and atherosclerotic plaque phenotype data. Such ground truth data may be additionally used to train one or more of machine learning model(s) 222 to determine an atherosclerotic plaque phenotype of a patient.

[0063] Angiography data 214 alone generally only provides a low level of lesion plaque characteristics for coronary artery disease (CAD) patients. For example, angiography data 214alone may not provide an identification of an atherosclerotic plaque phenotype of a given lesion. Example, atherosclerotic plaque phenotypes include lipid-rich plaque, which may have a large plaque burden, and calcium, which may have a large calcium burden.

[0064] Different atherosclerotic plaque phenotypes may be representative of focal CAD or diffuse CAD. Not only may different lesions including atherosclerotic plaque of different phenotypes warrant different treatment during a PCI, focal CAD and diffuse CAD may involve different treatments outside of the Cath lab, such as different medications and / or dosages of medications. For example, the chance of a plaque rupture may be greater with focal CAD than with diffuse CAD and may warrant a more aggressive anti-rupture treatment than for diffuse CAD.

[0065] The techniques of this disclosure leverage the potential to use angiography-defined FFR and angiography-defined full vessel PPG to characterize the atherosclerotic plaque phenotypes between focal CAD and diffuse CAD, as defined by the coronary hemodynamics.

[0066] A recent clinical trial and associated paper concluded that “Atherosclerotic plaque phenotypes associate with intracoronary hemodynamics. Focal CAD had a higher plaque burden and was predominantly lipid-rich with a high prevalence of thin-cap fibroatheroma (TCFA), whereas calcifications were more prevalent in diffuse CAD.” By identifying an atherosclerotic plaque phenotype associated with a patient, the techniques of this disclosure may help clinicians tailor treatment to patients. This could help with choice of stent, drug coated balloon, and / or postPCI drug regimen. For example, the techniques of this disclosure augment existing angiography technology to provide additional diagnostically beneficial information to the clinician. As such, the techniques of this disclosure may result in a better treatment for the patient and better patient outcomes.

[0067] For example, processing circuitry 204 may augment angiography-derived FFR by additionally layering on atherosclerosis phenotype identification to improve PCI treatment decision making. This information may be used by a clinician in treating the patient in the Cath lab and / or post-PCI with aggressive drug therapy to treat the thin cap and / or lipid rich plaque.

[0068] In some examples, processing circuitry 204 may execute machine learning model(s) 222 to determine FFR data based on angiography data 214 and store the FFR data in phenotype data 216. Processing circuitry 204 may executed machine learning model(s) 222 to determine PPG data based on angiography data 214 and store the FFR data in phenotype data 216. Processing circuitry 204 may execute machine learning model(s) 222 to determine an atherosclerotic plaque phenotype of the patient based on the FFR data and the PPG data, and store the atherosclerotic plaque phenotype in phenotype data 216.

[0069] Processing circuitry 204 may be implemented by one or more processors, which may include any number of fixed-function circuits, programmable circuits, or a combination thereof. Invarious examples, control of any function by processing circuitry 204 may be implemented directly or in conjunction with any suitable electronic circuitry appropriate for the specified function. Fixed-function circuits refer to circuits that provide particular functionality and are preset on the operations that may be performed. Programmable circuits refer to circuits that may programmed to perform various tasks and provide flexible functionality in the operations that may be performed. For instance, programmable circuits may execute software or firmware that cause the programmable circuits to operate in the manner defined by instructions of the software or firmware. Fixed-function circuits may execute software instructions (e.g., to receive parameters or output parameters), but the types of operations that the fixed-function circuits perform are generally immutable. In some examples, the one or more of the units may be distinct circuit blocks (fixed- function or programmable), and in some examples, the one or more units may be integrated circuits.

[0070] Instructions may be executed by one or more processors, such as one or more digital signal processors (DSPs), general purpose microprocessors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), graphics processing units (GPUs) or other equivalent integrated or discrete logic circuitry. Accordingly, the term processing circuitry as used herein may refer to one or more processors having any of the foregoing processor or processing structure or any other structure suitable for implementation of the techniques described herein. In addition, in some aspects, the functionality described herein may be provided within dedicated hardware or software modules configured for encoding and decoding, or incorporated in a combined codec. Also, the techniques could be fully implemented in one or more circuits or logic elements.

[0071] Display 206 may be touch sensitive or voice activated, enabling display 206 to serve as both an input and output device. Alternatively, a keyboard (not shown), mouse (not shown), or other data input devices (e.g., input device(s) 210) may be employed.

[0072] Network interface 208 may be adapted to connect to a network such as a local area network (LAN) that includes a wired network or a wireless network, a wide area network (WAN), a wireless mobile network, a Bluetooth network, or the internet. For example, computing device 150 may obtain angiography data 214 from imager 140 during a medical procedure. Computing device 150 may receive updates to its software, for example, application(s) 217, via network interface 208. Computing device 150 may also display notifications on display 206 that a software update is available.

[0073] Input device(s) 210 may include any device that enables a user to interact with computing device 150, such as, for example, a mouse, keyboard, foot pedal, touch screen, augmented-reality input device receiving inputs such as hand gestures or body movements, or voice interface.

[0074] Output device(s) 212 may include any connectivity port or bus, such as, for example, parallel ports, serial ports, universal serial busses (USB), or any other similar connectivity port known to those skilled in the art.

[0075] Application(s) 217 may be one or more software programs stored in memory 202 and executed by processing circuitry 204 of computing device 150. Processing circuitry 204 may execute user interface 218, which may display angiography data 214, 3D model 228, calcium characteristics 226 and / or suggestions 230 on display 206 and / or display device 110. A clinician may use the displayed data, for example, to plan a PCI procedure.

[0076] FIG. 3 is a conceptual diagram illustrating a longitudinal cross-section of an example culprit vessel according to one or more aspects of this disclosure. Culprit vessel 300 may include a lumen 302. Lumen 302 may be a portion of culprit vessel 300 through which blood would normally flow. Culprit vessel 300 may include lesion 304 which may be a calcified lesion. For example, lesion 304 may include a calcium deposit 306 which impinges into lumen 302 restricting the blood flow through lumen 302. In the example of FIG. 3, calcium deposit 306 may cover the entire interior circumference of lumen 302, such as is represented by calcium deposit 306 being shown both at a top and a bottom of lumen 302. The extent of coverage of lumen 302 by calcium deposit 306 may be an example of a calcium characteristic of calcium characteristics 226. In some examples, processing circuitry 204 may control display 206 to display 3D model 228 with calcium characteristics 226 overlaid thereon. For example, display 206 may display the representation of vessel 300 as shown in FIG. 3. In some examples, calcium deposit 306 may be displayed using a different color(s) or shading than lumen 302 to help a clinician better distinguish calcium deposit 306 from lumen 302.

[0077] With a calcified lesion, it may be beneficial for a clinician to utilize intravascular imaging such as OCT to help better observe the calcium and help with planning the PCI procedure. For example, measuring calcium with OCT may help a clinician identify calcific lesions that would benefit from calcium modification before stent implantation. Typically, a clinician may look for characteristics such as calcium thickness, calcium angle, and calcium length. Their analysis of such characteristics may influence their decision on whether or not to employ calcium modification devices.

[0078] This disclosure describes using this imaging data to train machine learning model(s) 222 to augment current angiography data to provide a similar level of insight without using OCT and / or IVUS on the particular patient during a current procedure.

[0079] For example, machine learning model(s) 222 may be trained using intravascular imaging data (pre-, peri-, and / or post- PCI), such as IVUS and / or OCT data. Training data may include identified and labeled thickness and / or depth of calcification. A potential threshold torecommend a calcium modification technique may be >0.5 mm. Training data may include angle data, such as concentric and / or eccentric data. A potential threshold to recommend a calcium modification technique may be >50% vessel arc (>180 degree angle). Training data may include a length of calcification. A potential threshold to recommend a calcium modification technique may be >5 mm in length. Training data may include hardness and / or density of the calcification. A potential threshold to recommend a calcium modification technique may be based on a brightness score indicative of the hardness and / or density of the calcification.

[0080] FIGS. 4A-4B are conceptual diagrams illustrating example calcium characteristics according to one or more aspects of this disclosure. FIG. 4A is an expanded partial view of the longitudinal cross-section of FIG. 3. Calcium deposit 306 may have a length 408. Generally, a longer length 408 may be more likely to negatively impact the ability of the stent to expand vessel 300 (FIG. 3) without calcification modification prior to implantation of the stent. Calcium deposit 306 may also have a depth 410. A larger depth 410 may be more likely to negatively impact the ability of the stent to expand vessel 300 without calcification modification prior to implantation of the stent.

[0081] FIG. 4B is a latitudinal cross section of the vessel 300 of FIG. 3. Calcium deposit 306 may also have an angle or vessel arc 412 at which calcium deposit 306 impinges into lumen 302 (FIG. 3). A larger angle may restrict blood flow to a greater amount than a smaller angle. As such, a larger angle or vessel arc 412 may more likely negatively impact the ability of the stent to expand vessel 300 without calcification modification prior to implantation of the stent. Length 408, depth 410, and / or angle or vessel arc 412 may be included in calcium characteristics 226.

[0082] In some examples, it may be desirable to perform one or more calcium modification procedures prior to implanting a stent when length 408 is greater than (or greater than or equal to) 5 mm, when an angle or vessel arc 412 is greater than (or greater than or equal to) 50%, and / or when depth 410 is greater than (or greater than or equal to) 0.5 mm. In some examples, the calcium modification threshold of threshold(s) 232 may include, or be based on, one or more of such measurements. In some examples, rather than executing one or more of machine learning model(s) 222 to determine whether to determine and provide suggestions 230, processing circuitry 204 may simply compare length 408 to a length threshold (e.g., 5 mm), compare angle or vessel arc 412 to a vessel arc threshold (e.g., 50%), and / or compare depth 410 to a depth threshold (e.g., 0.5 mm). In some examples, the length threshold, vessel arc threshold, and / or depth threshold may be stored in threshold(s) 232.

[0083] FIG. 5 is a flow diagram illustrating example calcium modification recommendation techniques according to one or more aspects of this disclosure. The techniques of FIG. 5 are described below with respect to processing circuitry 204, but such techniques may be performedby any of, or any combination of, processing circuitry of devices depicted in FIG. 1 or capable of performing such techniques.

[0084] Processing circuitry 204 may obtain the angiography data (500). For example, processing circuitry 204 may obtain angiography data 214 from imager 140 during a diagnostic medical procedure. The diagnostic medical procedure may be a separate medical procedure from a PCI procedure to implant a stent or may occur immediately before the PCI procedure to implant the stent. The diagnostic medical procedure may be used to diagnose one or more medical conditions of the patient, such as CAD and / or a identify one or more lesions within vasculature of the patient.

[0085] Processing circuitry 204 may determine, based on the angiography data, one or more calcium characteristics of a lesion (502). For example, processing circuitry 204 may determine calcium characteristics 226 based on angiography data 214.

[0086] Processing circuitry 204 may determine, based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent (504). For example, processing circuitry may determine a likelihood that the stent will provide a desired level of expansion if the clinician does not perform any calcium modification procedure to the lesion prior to implanting the stent.

[0087] Processing circuitry 204 may output an indication of the determined likelihood the stent will provide the at least the level of expansion (506). For example, processing circuitry 204 may output the determined likelihood that the stent will provide at least the level of expansion as a percentage likelihood (e.g., 80%). In the example where the determined likelihood is not as high as desired, processing circuitry 204 may output likelihoods that the stent will provide at least the level of expansion associated with various calcium modification procedures.

[0088] In some examples, processing circuitry 204 may, based on the determined likelihood that the stent will provide at least the level of expansion not satisfying a calcium modification threshold, determine a respective likelihood for each of a plurality of calcium modification procedures that the stent will provide the at least the level of expansion. In some examples, processing circuitry 204 may output respective indications for the determined respective likelihoods. In some examples, the indication of the likelihood the stent will provide the at least the level of expansion includes the respective indications for the respective likelihoods.

[0089] In some examples, at least one of the level of expansion or the calcium modification threshold is programmable. In some examples, the angiography data is diagnostic angiography data. In some examples, the calcium characteristics include at least one of a length, an angle, or a depth of a calcium deposit of the lesion.

[0090] In some examples, as part of at least one of determining the one or more calcium characteristics, determining the likelihood that the stent will provide at least the level of expansion, or determining a respective likelihood for each of a plurality of calcium modification procedures that the stent will provide the at least the level of expansion respective likelihoods, processing circuitry 204 may execute one or more machine learning models, medical system of claim 6, wherein the one or more machine learning models are trained using angiography data of other patients and at least one of IVUS or OCT data of other patients associated with the angiography data of other patients. In some examples, at least one of the angiography data of other patients comprises pre-PCI, peri-PCI, and post-PCI angiography data or the IVUS or OCT data of other patients comprises pre-PCI, peri-PCI, and post-PCI IVUS or OCT data.

[0091] In some examples, processing circuitry 204 may generate, based on the angiography data, a three-dimensional (3D) model of anatomy of the patient. In some examples, processing circuitry 204 may output, for display, the 3D model including at least one representation of the one or more calcium characteristics of a lesion.

[0092] In some examples, processing circuitry 204 may determine fractional flow reserved (FFR) data based on the angiography data. In some examples, processing circuitry 204 may determine pullback pressure gradient (PPG) data based on the angiography data. In some examples, processing circuitry 204 may determine an atherosclerotic plaque phenotype of a second lesion based on the FFR data and the PPG data. In some examples, processing circuitry 204 may output an indication of the atherosclerotic plaque phenotype for display.

[0093] FIG. 6 is a conceptual diagram illustrating an example machine learning model according to one or more aspects of this disclosure. Machine learning model 600 may be an example of machine learning model(s) 222. Machine learning model 600 may be an example of a deep learning model, or deep learning algorithm, trained to determine calcium characteristics 226, whether to suggest one or more calcium modification techniques, and / or suggest a plurality of calcium modification techniques and corresponding likelihoods of success. One or more of computing device 150 and / or server 160 may train, store, and / or utilize machine learning model 600, but other devices of system 100 may apply inputs to machine learning model 600 in some examples. In some examples, various types of machine learning and deep learning models or algorithms may be utilized. For examples, a convolutional neural network (CNN) model, e.g., ResNet-18, may be used. Some non-limiting examples of models that may be used for transfer learning include AlexNet, VGGNet, GoogleNet, ResNet50, or DenseNet, etc. Some non-limiting examples of machine learning techniques include Support Vector Machines, K-Nearest Neighbor algorithm, and Multi-layer Perceptron.

[0094] As shown in the example of FIG. 6, machine learning model 600 may include three types of layers. These three types of layers include input layer 602, hidden layers 604, and output layer 606. Output layer 606 comprises the output from the transfer function 605 of output layer 606. Input layer 602 represents each of the input values XI through X4 provided to machine learning model 600. In some examples, the input values may include any of the values input into the machine learning model, as described above. For example, the input values may include angiography data 214, as described above. In addition, in some examples input values of machine learning model 600 may include additional data, such as other data that may be collected by or stored in system 100.

[0095] Each of the input values for each node in the input layer 602 is provided to each node of a first layer of hidden layers 604. In the example of FIG. 6, hidden layers 604 include two layers, one layer having four nodes and the other layer having three nodes, but fewer or greater number of nodes may be used in other examples. Each input from input layer 602 is multiplied by a weight and then summed at each node of hidden layers 604. During training of machine learning model 600, the weights for each input are adjusted to establish a relationship between angiography data 214, calcium characteristics, likelihood of that a stent may expand as desired without calcium modification treatment, and likelihoods that a stent may expand as desired with corresponding calcium modification treatments. In some examples, one hidden layer may be incorporated into machine learning model 600, or three or more hidden layers may be incorporated into machine learning model 600, where each layer includes the same or different number of nodes.

[0096] The result of each node within hidden layers 604 is applied to the transfer function of output layer 606. The transfer function may be linear or non-linear, depending on the number of layers within machine learning model 600. Example non-linear transfer functions may be a sigmoid function or a rectifier function. The output 607 of the transfer function may be a classification that the calcium characteristics of a lesion do not require calcium modification prior to implantation of the stent, a classification of the likelihoods of each of a plurality of calcium modification treatments may provide for the desired expansion of the stent after implantation, and / or an atherosclerotic plaque phenotype of the patient.

[0097] As shown in the example above, by applying machine learning model 600 to input data such as angiography data 214, processing circuitry 204 is able to determine calcium characteristics of a lesion, determine a likelihood that a stent will provide a desired level of expansion of the vessel after implantation without calcium modification, and / or likelihoods of that a stent will provide a desired level of expansion of the vessel after implantation with each of a plurality of the calcium modification techniques. This may improve the ability of a clinician to achieve a desired level of expansion of a stent.

[0098] FIG. 7 is a conceptual diagram illustrating an example training process for a machine learning model according to one or more aspects of this disclosure. Process 770 may be used to train machine learning model(s) 222 or machine learning model 600. A machine learning model 774 (which may be an example of machine learning model 600 and / or machine learning model(s) 222) may be implemented using any number of models for supervised and / or reinforcement learning, such as but not limited to, an artificial neural network, a decision tree, naive Bayes network, support vector machine, or k-nearest neighbor model, CNN, recurrent neural network (RNN), long short-term memory (LSTM), ensemble network, to name only a few examples.

[0099] In some examples, one or more of computing device 150 and / or server 160 initially trains machine learning model 774 based on a corpus of training data 772. Training data 772 may include, for example, angiography data pre-, peri-, and post- PCI procedures. In some, examples, training data may include annotated images, such as annotations of anatomy, morphology (including calcium identificationO and / or any devices (e.g., stents) being used. Training data 772 may include outcome data and device strategies associated with successful and / or unsuccessful outcomes. Training data 772 may also include IVUS and / or OCT data, such as pre-, peri-, and / or post-PCI IVUS and / or OCT data. In some examples, training data 772 may include wire-based FFR, PPG data, and atherosclerotic plaque phenotype data.

[0100] While training machine learning model 774, processing circuitry of system 100 may compare 776 a prediction or classification with a target output 778. Processing circuitry 204 may utilize an error signal from the comparison to train (learning / training 780) machine learning model 774. Processing circuitry 204 may generate machine learning model weights or other modifications which processing circuitry 204 may use to modify machine learning model 774. For examples, processing circuitry 204 may modify the weights of machine learning model 774 based on the learning / training 780. For example, one or more of computing device 150 and / or server 160, may, for each training instance in training data 772, modify, based on training data 772, the manner in which computing device 130 and / or server 160 may determine calcium characteristics of a lesion, determine a likelihood that a stent will provide a desired level of expansion of the vessel after implantation without calcium modification, determine likelihoods of that a stent will provide a desired level of expansion of the vessel after implantation with each of a plurality of the calcium modification techniques, and / or determine an atherosclerotic plaque phenotype of a patient.

[0101] FIG. 8 is a conceptual diagram illustrating another example training process for a machine learning model according to one or more aspects of this disclosure. Process 800 may be used to train machine learning model(s) 222 or machine learning model 600. A machine learning model 808 (which may be an example of machine learning model 600 and / or machine learning model(s) 222) may include a neural network or other type of machine learning model.

[0102] Training data 802 may include patient data. In some examples, training data 802 include patient data. Such patient data may include angiography data (e.g., X-ray data) from pre-, peri-, and / or post-PCI procedures. In some examples, training data 802 includes annotated images including anatomy, morphology (including calcium identification and / or any devices (e.g., stents) being used).

[0103] Training data 802 may also include outcome data and device strategies (including calcium modification techniques) associated with successful and / or unsuccessful outcomes. Training data 802 may also include IVUS and / or OCT data, such as pre-, peri-, and / or post-PCI IVUS and / or OCT data. Such data may be used to augment decision making and procedural success.

[0104] For example, training data 802 may identify presence, severity, and location of calcium in a coronary artery aside from what a human may normally be able to see in angiography data alone. The training data may also provide a level of successful expansion of a stent associated any (or no) calcium modification procedures being performed given the presence, severity, and location of the calcium.

[0105] The input training data 802 may be pre-processed 804, for example, by a neural network. For example, angiography data may be processed and / or interrogated. This may include processing FFR, index of microvascular resistance (IMR), and / or other physiological indices. This may also include generating a 3D model of a coronary artery. IVUS and / or OCT data may be interrogated to identify presence, severity, and location of calcium, such as calcium depth, hardness or density, angle (including concentric / eccentric), lengthy. Computing device 150 and / or server 160 pre-processing input training data 802 may prepare and validate training data 802 for training.

[0106] The pre-processed data may be modeled 806 to generate machine learning model 808. For example, modeling 806 may generate a transfer function (Fx) that may be applied to input data to make a prediction regarding the calcium characteristics of a lesion and / or a likelihood of a stent providing a desired level of expansion if no, or if any number of calcium modification procedures are performed prior to implantation of the stent.

[0107] The techniques described in this disclosure may be implemented, at least in part, in hardware, software, firmware or any combination thereof. For example, various aspects of the described techniques may be implemented within one or more processors or processing circuitry, including one or more microprocessors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or any other equivalent integrated or discrete logic circuitry, as well as any combinations of such components. The terms “controller”, “processor”, or “processing circuitry” may generally refer to any of the foregoing logic circuitry, alone or in combination with other logic circuitry, or any other equivalent circuitry.A control unit comprising hardware may also perform one or more of the techniques of this disclosure. Such hardware, software, and firmware may be implemented within the same device or within separate devices to support the various operations and functions described in this disclosure. In addition, any of the described units, circuits or components may be implemented together or separately as discrete but interoperable logic devices. Depiction of different features as circuits or units is intended to highlight different functional aspects and does not necessarily imply that such circuits or units must be realized by separate hardware or software components. Rather, functionality associated with one or more circuits or units may be performed by separate hardware or software components or integrated within common or separate hardware or software components.

[0108] The techniques described in this disclosure may also be embodied or encoded in a computer-readable medium, such as a computer-readable storage medium, containing instructions. Instructions embedded or encoded in a computer-readable storage medium may cause a programmable processor, or other processor, to perform the method, e.g., when the instructions are executed. Computer readable storage media may include random access memory (RAM), read only memory (ROM), programmable read only memory (PROM), erasable programmable read only memory (EPROM), or electronically erasable programmable read only memory (EEPROM), or other computer readable media.

[0109] This disclosure includes the following non-limiting examples.

[0110] Example 1. A medical system comprising: memory configured to store angiography data of a patient; and processing circuitry communicatively coupled to the memory, the processing circuitry being configured to: obtain the angiography data; determine, based on the angiography data, one or more calcium characteristics of a lesion; determine, based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent; and output an indication of the determined likelihood the stent will provide the at least the level of expansion.

[0111] Example 2. The medical system of example 1, wherein the processing circuitry is further configured to: based on the determined likelihood that the stent will provide at least the level of expansion not satisfying a calcium modification threshold, determine a respective likelihood for each of a plurality of calcium modification procedures that the stent will provide the at least the level of expansion; and output respective indications for the determined respective likelihoods.

[0112] Example s. The medical system of example 2, wherein the indication of the likelihood the stent will provide the at least the level of expansion comprises the respective indications for the respective likelihoods.

[0113] Example 4. The medical system of any of examples 1-3, wherein at least one of the level of expansion or the calcium modification threshold is programmable.

[0114] Example 5. The medical system of any of examples 1-4 wherein, wherein the angiography data is diagnostic angiography data.

[0115] Example 6. The medical system of any of examples 1-5, wherein the calcium characteristics comprise at least one of a length, an angle, or a depth of a calcium deposit of the lesion.

[0116] Example 7. The medical system of any of examples 1-6, wherein as part of at least one of determining the one or more calcium characteristics, determining the likelihood that the stent will provide at least the level of expansion, or determining a respective likelihood for each of a plurality of calcium modification procedures that the stent will provide the at least the level of expansion respective likelihoods, the processing circuitry is configured to execute one or more machine learning models.

[0117] Example 8. The medical system of example 7, wherein the one or more machine learning models are trained using angiography data of other patients and at least one of IVUS or OCT data of other patients associated with the angiography data of other patients.

[0118] Example 9. The medical system of example 8, wherein at least one of: the angiography data of other patients comprises pre-PCI, peri-PCI, and post-PCI angiography data; or the IVUS or OCT data of other patients comprises pre-PCI, peri-PCI, and post-PCI IVUS or OCT data.

[0119] Example 10. The medical system of any of examples 1-9, wherein the processing circuitry is further configured to: generate, based on the angiography data, a three-dimensional (3D) model of anatomy of the patient; and output, for display, the 3D model including at least one representation of the one or more calcium characteristics of a lesion.

[0120] Example 11. The medical system of any of examples 1-10, wherein the processing circuitry is further configured to: determine fractional flow reserved (FFR) data based on the angiography data; determine pullback pressure gradient (PPG) data based on the angiography data; determine an atherosclerotic plaque phenotype of the patient based on the FFR data and the PPG data; and output an indication of the atherosclerotic plaque phenotype for display.

[0121] Example 12. A method comprising: obtaining, by processing circuitry, angiography data of a patient; determining, by the processing circuitry and based on the angiography data, one or more calcium characteristics of a lesion; determining, by the processingcircuitry and based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent; and outputting, by the processing circuitry, an indication of the determined likelihood the stent will provide the at least the level of expansion.

[0122] Example 13. The method of example 12, further comprising: determining, by the processing circuitry and based on the determined likelihood that the stent will provide at least the level of expansion not satisfying a calcium modification threshold, a respective likelihood for each of a plurality of calcium modification procedures that the stent will provide the at least the level of expansion; and outputting, by the processing circuitry, respective indications for the determined respective likelihoods.

[0123] Example 14. The method of example 13, wherein the indication of the likelihood the stent will provide the at least the level of expansion comprises the respective indications for the respective likelihoods.

[0124] Example 15. The method of any of examples 12-14, wherein at least one of the level of expansion or the calcium modification threshold is programmable.

[0125] Example 16. The method of any of examples 12-15, wherein the calcium characteristics comprise at least one of a length, an angle, or a depth of a calcium deposit of the lesion.

[0126] Example 17. The method of any of examples 12-16, wherein at least one of determining the one or more calcium characteristics, determining the likelihood that the stent will provide at least the level of expansion, or determining a respective likelihood for each of a plurality of calcium modification procedures that the stent will provide the at least the level of expansion respective likelihoods, comprises executing one or more machine learning models.

[0127] Example 18. The method of example 17, wherein the one or more machine learning models are trained using angiography data of other patients and at least one of IVUS or OCT data of other patients associated with the angiography data of other patients.

[0128] Example 19. The method of any of examples 12-18, further comprising: determining, by the processing circuitry and based on the angiography data, fractional flow reserved (FFR) data; determining, by the processing circuitry and based on the angiography data, pullback pressure gradient (PPG) data; determining, by the processing circuitry and based on the FFR data and the PPG data, an atherosclerotic plaque phenotype of the patient; and output, by the processing circuitry, an indication of the atherosclerotic plaque phenotype for display.

[0129] Example 20. A non-transitory computer-readable storage medium storing instructions, which when executed cause processing circuitry to: obtain angiography data; determine, based on the angiography data, one or more calcium characteristics of a lesion;determine, based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent; and output an indication of the determined likelihood the stent will provide the at least the level of expansion.

[0130] Various examples have been described. These and other examples are within the scope of the following claims.

Claims

What is claimed is:

1. A medical system comprising: memory configured to store angiography data of a patient; and processing circuitry communicatively coupled to the memory, the processing circuitry being configured to: obtain the angiography data; determine, based on the angiography data, one or more calcium characteristics of a lesion; determine, based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent; and output an indication of the determined likelihood the stent will provide the at least the level of expansion.

2. The medical system of claim 1, wherein the processing circuitry is further configured to: based on the determined likelihood that the stent will provide at least the level of expansion not satisfying a calcium modification threshold, determine a respective likelihood for each of a plurality of calcium modification procedures that the stent will provide the at least the level of expansion; and output respective indications for the determined respective likelihoods.

3. The medical system of claim 2, wherein the indication of the likelihood the stent will provide the at least the level of expansion comprises the respective indications for the respective likelihoods.

4. The medical system of any of claims 1-3, wherein at least one of the level of expansion or the calcium modification threshold is programmable.

5. The medical system of any of claims 1-4 wherein, wherein the angiography data is diagnostic angiography data.

6. The medical system of any of claims 1-5, wherein the calcium characteristics comprise at least one of a length, an angle, or a depth of a calcium deposit of the lesion.

7. The medical system of any of claims 1-6, wherein as part of at least one of determining the one or more calcium characteristics, determining the likelihood that the stent will provide at least the level of expansion, or determining a respective likelihood for each of a plurality of calcium modification procedures that the stent will provide the at least the level of expansion respective likelihoods, the processing circuitry is configured to execute one or more machine learning models.

8. The medical system of claim 7, wherein the one or more machine learning models are trained using angiography data of other patients and at least one of IVUS or OCT data of other patients associated with the angiography data of other patients.

9. The medical system of claim 8, wherein at least one of: the angiography data of other patients comprises pre-PCI, peri-PCI, and post-PCI angiography data; or the IVUS or OCT data of other patients comprises pre-PCI, peri-PCI, and post-PCI IVUS or OCT data.

10. The medical system of any of claims 1-9, wherein the processing circuitry is further configured to: generate, based on the angiography data, a three-dimensional (3D) model of anatomy of the patient; and output, for display, the 3D model including at least one representation of the one or more calcium characteristics of a lesion.

11. The medical system of any of claims 1-10, wherein the processing circuitry is further configured to: determine fractional flow reserved (FFR) data based on the angiography data; determine pullback pressure gradient (PPG) data based on the angiography data; determine an atherosclerotic plaque phenotype of the patient based on the FFR data and the PPG data; and output an indication of the atherosclerotic plaque phenotype for display.

12. A method comprising: obtaining, by processing circuitry, angiography data of a patient;determining, by the processing circuitry and based on the angiography data, one or more calcium characteristics of a lesion; determining, by the processing circuitry and based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent; and outputting, by the processing circuitry, an indication of the determined likelihood the stent will provide the at least the level of expansion.

13. The method of claim 12, further comprising: determining, by the processing circuitry and based on the determined likelihood that the stent will provide at least the level of expansion not satisfying a calcium modification threshold, a respective likelihood for each of a plurality of calcium modification procedures that the stent will provide the at least the level of expansion; and outputting, by the processing circuitry, respective indications for the determined respective likelihoods.

14. The method of claim 13, wherein the indication of the likelihood the stent will provide the at least the level of expansion comprises the respective indications for the respective likelihoods.

15. A non-transitory computer-readable storage medium storing instructions, which when executed cause processing circuitry to: obtain angiography data; determine, based on the angiography data, one or more calcium characteristics of a lesion; determine, based on the one or more calcium characteristics, a likelihood that a stent will provide at least a level of expansion in an area of the lesion without performing calcium modification of the lesion prior to implanting the stent; and output an indication of the determined likelihood the stent will provide the at least the level of expansion.

Images