Angiographic-based assistance for coronary sinus reducer procedure

Abstract

Example systems, devices, and techniques are described for imaging data-based assistance for a coronary sinus reducer implant procedure. An example medical system includes memory configured to store imaging data of a patient and processing circuitry communicatively coupled to the memory. The processing circuitry is configured to obtain the imaging data. The processing circuitry is configured to execute one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient. The processing circuitry is configured to output an indication of the likelihood of success.

Description

ANGIOGRAPHIC-BASED ASSISTANCEFOR CORONARY SINUS REDUCER PROCEDURE

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 693,907, filed September 12, 2024, and entitled, “ANGIOGRAPHIC-BASED ASSISTANCE FOR CORONARY SINUS REDUCER PROCEDURE,” the entire content of which is incorporated by reference.TECHNICAL FIELD

[0002] This disclosure relates to assistance for a coronary sinus reducer stent implant 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 and / or placing a device inside a patient, such as moving a medical instrument to an intended location inside the patient and / or placing a stent at a particular location. 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), magnetic resonance imaging (MRI) systems, positron emission tomography (PET) systems, as well as other imaging systems.

[0004] In some medical procedures, a clinician may treat a patient condition, such as refractory angina, by implanting a coronary sinus reducer stent. A coronary sinus reducer stent may include an hour-glass shaped lumen and be configured to be implanted in a coronary sinus of a patient.SUMMARY

[0005] A coronary sinus reducer stent is a device that may be used to treat coronary microvascular disease, specifically including a subset of that disease state which is termed as refractory angina. Refractory angina is a condition where blood flow to a heart muscle of a patient is insufficient to meet the demands of the heart muscle, leading to chest pain or discomfort, and where other treatments such as percutaneous coronary intervention (PCI),coronary artery bypass graft (CABG), and medical therapy have failed to alleviate the chest pain. A coronary sinus reducer stent may include an hour-glass shaped lumen which may serve to effectively reduce the size of a lumen of the coronary sinus. The use of procedures to implant coronary sinus reducer stents is increasing. However, clinical studies have shown that up to 30% of patients may be non-responsive to such procedures. See Giannini F et al., Coronary Sinus Reducer Implantation for the Treatment of Chronic Refractory Angina: A Single-Center Experience, JACC Cardio Interv 11-8 2018. Therefore, there is a need for improved screening techniques to identify for which patients a coronary sinus reducer may be desirable. There is also a need for assistance during implantation and post procedure optimization.

[0006] This disclosure describes techniques for an imaging data-based assistant system to help screen and optimize the implantation of a coronary sinus reducer stent, thus reducing the amount of coronary sinus reducer stent implant procedures which do not have effective results for the patients and improving patient outcomes for those coronary sinus reducer stent implant procedures that do occur.

[0007] In one example, the disclosure describes a medical system comprising: memory configured to store imaging data of a patient; and processing circuitry communicatively coupled to the memory, the processing circuitry being configured to: obtain the imaging data; execute one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient; and output an indication of the likelihood of success.

[0008] In another example, the disclosure describes a method comprising: obtaining, by processing circuitry, imaging data of a patient; executing, by the processing circuitry, one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient; and outputting, by the processing circuitry, an indication of the likelihood of success.

[0009] In yet another example, the disclosure describes non-transitory computer readable media comprising instructions, which, when executed, cause processing circuitry to: obtain imaging data of a patient; execute one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient; and output an indication of the likelihood of success.

[0010] 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.

[0011] 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 theapparatus 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

[0012] FIG. l is a schematic perspective view of one example of a system for providing imaging data-based assistance for a coronary sinus reducer implant procedure according to one or more aspects of this disclosure.

[0013] FIG. 2 is a schematic view of one example of a computing system of the system of FIG. 1.

[0014] FIG. 3 is a conceptual diagram illustrating an example coronary sinus reducer stent in a coronary sinus of a patient according to one or more aspects of this disclosure.

[0015] FIG. 4 is conceptual diagram illustrating example training data for one or more machine learning models according to one or more aspects of this disclosure.

[0016] FIG. 5 is a flow diagram of example techniques for determining a likelihood of success of a coronary sinus reducer stent implant procedure for a patient according to one or more aspects of this disclosure.

[0017] FIG. 6 is a conceptual diagram illustrating an example machine learning model according to one or more aspects of this disclosure.

[0018] 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.

[0019] 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

[0020] This disclosure describes a system for providing assistance to a clinician regarding coronary sinus reducer stent implant procedures. A coronary sinus reducer stent is a medical device used, for example, in the treatment of refractory angina. A coronary sinus reducer stent is typically implanted when other treatments like medications, angioplasty, or bypass surgery, have not been effective. These patients are often referred to as “no option patients.” A coronary sinus reducer stent may be implanted through a minimally invasive procedure and remains an option for patients who are not suitable candidates for other interventions or who have exhausted other treatment options.

[0021] Procedures to implant coronary sinus reducer stents are becoming more prevalent. However, a sizable portion of potential patients for such procedures may not respond to such a procedure. As such, it may be desirable to have a system that may analyze patients for the usefulness of such a procedure and, in the case the procedure is performed, may assist (e.g., plan and / or guide) the clinician in planning the implant procedure and / or implanting the coronary sinus reducer stent.

[0022] This disclosure describes techniques for providing recommendations and real-time imaging (e.g., angiographic) assistance for a clinician implanting a coronary sinus reducer stent. In some examples, a medical system may process imaging data captured pre-procedure and recommend whether or not a patient is a good candidate for a coronary sinus reducer stent. For example, a patient having coronary sinus side branch location(s) that is along the length of a coronary sinus reducer stent, after implantation of the coronary sinus reducer stent, may be less likely to experience symptom relief than a patient without such coronary sinus side branch location(s).

[0023] In some examples, the medical system may recommend a particular type of, such as the size of, a coronary sinus reducer stent for the patient, and / or recommend an implant location within anatomy of the patient. The medical system may execute one or more machine learning models to determine the recommendation(s).

[0024] The techniques of this disclosure may provide a technical solution to a technical problem. For example, current techniques to determine which patients are candidates for a coronary sinus reducer stent do not identify up to 30% of patients who may not be good candidates for a coronary sinus reducer stent implant procedure. The techniques of this disclosure identify such patients, and thereby reduce the number of coronary sinus reducer stent implant procedures which may not result in positive outcomes. Additionally, the techniques of this disclosure may provide a clinician with data-backed recommendations for which coronary sinus reducer stent to use and / or a location at which to implant the coronary sinus reducer stent. Such recommendations may increase a likelihood of a successful procedure, thereby improving patient outcomes and reducing the frequency and / or severity of angina symptoms.

[0025] A coronary sinus reducer stent may work by increasing backpressure into the precapillary arteriolar system which facilitates dilation of the constricted subendocardial capillaries. This improves capillary perfusion in the sub endocardium of the ischemic territory, which improves contractility and increases oxygen consumption. By doing so, the coronary sinus reducer may improve blood flow to the heart muscle and alleviate symptoms of angina.

[0026] The techniques of this disclosure may utilize data regarding a pressure gradient across the coronary sinus reducer stent in the long term so as to predict the likelihood of success of theprocedure for a given patient. The techniques of this disclosure may use imaging data-based image analysis to understand the pressure gradient. For example, the degree of pressure drop across a coronary sinus reducer stent may correlate with the symptomatic relief experienced by patients with refractory angina. A more substantial pressure drop, which may indicate improved coronary blood flow, may result in better alleviation of angina symptoms and improved quality of life for the patient. For example, the techniques may predict a fractional flow reserve (FFR) measurement and / or pressure gradient across the coronary sinus reducer stent as part of determining a likelihood of success, a recommended type of coronary sinus reducer stent, and / or a recommended implant location.

[0027] FIG. l is a schematic perspective view of one example of a system for guiding navigation of a medical instrument in the vasculature of a patient 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 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. System 100 may be used during a medical procedure, such as a medical procedure to implant a coronary sinus reducer stent into a coronary sinus of a patient.

[0028] 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 guidance workstation. Computing device 150 may control the operation of imager 140 and receive the output of imager 140 and may receive imaging data from imager 140. Computing device 150 may execute one or more machine learning models to determine procedural recommendations to provide to a clinician.

[0029] Display device 110 may be configured to output instructions, images, and messages relating to the medical procedure(s), such as any procedural recommendations determined by computing device 150. For example, display device 110 may display imaging data obtained through imager 140 and / or one or more procedural recommendations. Table 120 may be, for example, an operating table or other table suitable for use during a medical procedure.

[0030] In the example of FIG. 1, imager 140, such as an angiography imager, fluoroscopy imager, a CT 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, a medical instrument, and / or a device to be implanted, such as a coronary sinus reducer stent issues inside the patient’sbody through the generation of imaging data. While primarily described herein 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 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. In some examples, the system may include c-arm video processing equipment or a connected video capture system.

[0031] 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 coronary sinus reducer stent implant procedure, the region of interest may include the coronary sinus of the patient or a portion thereof.

[0032] 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, 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) and / or perform one or more of, or a portion of one or more of, the determinations associated therewith.

[0033] 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 imaging data, e.g., via imager 140. Computing device 150, imager 140, and / or server 160 may execute one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient. Computing device 150, imager 140, and / or server 160 may output an indication of the likelihood of success. In some examples, computing device 150, imager 140, and / or server 160 may execute the one or more machine learning models to determine, based on the imaging data, a recommended coronary sinus reducer stent and a recommended location within anatomy of the patient for implanting the recommended coronary sinus reducer stent. Computing device 150, imager 140, and / or server 160 may output an indication of the recommended coronary sinus reducer stent and a recommended location.

[0034] By outputting an indication of the likelihood of success of a coronary sinus reducer stent implant procedure and / or outputting a recommended coronary sinus reducer stent and a recommended location, system 100 may assist clinicians in more effectively determining on which patients to perform a coronary sinus reducer stent implant procedure, and / or which coronary sinus reducer stent to use and where to implant the coronary sinus reducer stent. Assuch, the techniques of this disclosure may reduce risk to patients whose outcomes are likely to be unsuccessful, improve patient outcomes, and / or improve medical facility efficiency.

[0035] FIG. 2 is a schematic view of one example of a computing device 150 of system 10 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.

[0036] Computing device 150 may be configured to perform processing, control and other functions associated with imager 140. In some examples, 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.

[0037] 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 some examples, 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.

[0038] 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).

[0039] 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 non-removable media implemented in any method or technology for storage of informationsuch as computer-readable instructions, data structures, program modules, or other data. For example, computer-readable storage media includes RAM, ROM, EPROM, EEPROM, flash memory or other solid state memory technology, CD-ROM, DVD, Blu-Ray or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, 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.

[0040] Memory 202 may store imaging data 214, other data 216, one or more recommendation(s) 226, and / or three-dimensional (3D) model 228. Imaging data 214 may include angiography data obtained, for example, from imager 140 during the coronary sinus reducer stent implant procedure (e.g., peri-procedure). In some examples, imaging data 214 may also include data obtained during a prior medical procedure of the patient, such as a diagnostic angiography procedure. Such imaging data may be considered pre-implant procedure imaging data.

[0041] While the coronary sinus reducer stent implant procedure is underway (e.g., during the coronary sinus reducer stent implant procedure), additional imaging data may be obtained from imager 140 and stored in imaging data 214 (e.g., as peri-implantation procedure data). Such imaging data may be displayed via display 206 and / or display device 110 and may be used by a clinician when implanting the coronary sinus reducer stent in the coronary sinus and / or navigating a medical instrument used to deliver the coronary sinus reducer stent through anatomy of a patient. In some examples, imaging data 214 may include data captured by imager 140 after the implantation of the coronary sinus reducer stent (e.g., as post-implant procedure data and / or long-term data). Such data may include imaging data captured immediately after the implantation of the coronary sinus reducer stent, but prior to the end of the medical procedure, and / or imaging data captured at a later date in time, such as a number of months after the coronary sinus reducer stent implant procedures, like approximately 3 months, 6 months, 9, months, 12 months, 24 months, and / or the like.

[0042] Imaging 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, imaging data 214 may be captured by imager 140 (FIG. 1). Processing circuitry 204 may obtain imaging data 214 from imager 140 and store imaging data 214 in memory 202.

[0043] Other data 216 may include imaging data from one or more additional imagers. For example, if imager 140 is an angiography imager, other data 216 may include CT data, OCT data, and / or IVUS data. Other data 216 may include wire-based FFR data, image-based FFR data, and / or patient meta data (which may include demographic data, co-morbidities, etc.).

[0044] Recommendation(s) 226 may include a likelihood of success of a coronary sinus reducer stent implant procedure for a patient, a recommended type of coronary sinus reducer stent, a recommended location for implantation of the coronary sinus reducer stent, and / or a recommended adjustment to the coronary sinus reducer stent once the coronary sinus reducer stent is implanted. Computing device 150 may execute one or more machine learning model(s) 222 on imaging data 214 to determine recommendation(s) 226. Computing device 150 may output recomm endation(s) 226 or indications thereof. In some examples, computing device 150 may output recommendation(s) 226 or indications thereof for viewing on display 206 and / or display device 110. In some examples, recomm endation(s) 226 may be editable by a clinician. For example, a clinician may use input device(s) 210 to edit a recommended coronary sinus reducer stent, for example, to match a coronary sinus reducer stent present in the inventory of a medical facility. The clinician may edit the recommended location of the implant. In some examples, if the clinician edits any of recommendation(s) 226, computing device 150 may redetermine one or more other recommendation(s) of recommendation(s) 226. For example, if a clinician edits a recommended coronary sinus reducer stent, computing device 150 may redetermine a likelihood of success and / or a recommended location.

[0045] Memory 202 may also store one or more machine learning model(s) 222. Machine learning model(s) 222 may be configured to determine, when executed by processing circuitry 204, recommendation(s) 226.

[0046] In some examples, imaging data 214 includes angiography and / or CT data. Imaging data 214 may include imaging data obtained pre-implant procedure, peri-implant procedure, postimplant procedure, and / or over a long-term. In some examples, processing circuitry 204 may use imaging data 214 to generate 3D model 228 of the anatomy of the patient which processing circuitry 204 may cause display 206 and / or display device 110 to visually display.

[0047] Processing circuitry 204 may execute machine learning model(s) 222, which may include a deep learning model, such as a convolutional neural network (CNN), or other type of machine learning model, to determine recommendation(s) 226. Machine learning model(s) 222 may be trained based on pre-implant procedure data, peri-implant procedure data, post-implant procedure data, long-term data, and / or the like, as is discussed further herein with respect to FIG. 4. In some examples, training data may include pre-implant procedure, post-implant procedure,and / or long-term angina ratings, which may be used to determine improvements to quality of life and / or pain reduction over time.

[0048] Processing circuitry 204 may obtain imaging data 214. Processing circuitry 204 may execute one or more machine learning model(s) 222 to determine, based on imaging data 214, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient (e.g., of recommendation(s) 226). Processing circuitry 204 may output an indication of the likelihood of success. In some examples, processing circuitry 204 may execute one or more machine learning model(s) to determine, based on imaging data 214, a recommended coronary sinus reducer stent and a recommended location within anatomy of the patient for implanting the recommended coronary sinus reducer stent (e.g., of recomm endation(s) 226). Processing circuitry 204 may output an indication of the recommended coronary sinus reducer stent and a recommended location. In some examples, processing circuitry 204 may provide real-time angiographic assistance to help optimize the implantation of a coronary sinus reducer stent.

[0049] For example, pre-procedure, processing circuitry 204 may identify patients who should be discounted for a coronary sinus reducer stent implant procedure based on anatomical considerations. For example, the anatomy of one patient may negatively affect the ability of a coronary sinus reducer stent to generate a sufficient pressure gradient across the coronary sinus reducer stent to cause the capillary perfusion in the subendocarium of the ischemic territory to improve blood flow to the heart muscle and alleviate the symptoms of angina.

[0050] Processing circuitry 204 may also determine and recommend a type of coronary sinus reducer stent, such as coronary sinus reducer stent dimensions and / or interior diameter, for example, based on the size and location of the coronary sinus of a patient. For example, computing device 150 may determine, based on imaging data 214 the size, location, and / or anatomical characteristics of the coronary sinus of the patient. The determined size, location, and / or anatomical characteristics of the coronary sinus may be used by processing circuitry 204, for example, executing machine learning model(s) 222 to determine recommended dimensions for the coronary sinus reducer stent to be implanted. In some examples, a clinician may input to processing circuitry 204 via input device(s) 210 dimensions for the coronary sinus reducer stent to be used and / or the size, location, and / or anatomical characteristics of the coronary sinus of the patient.

[0051] In some examples, processing circuitry 204 may determine, based on imaging data 214, variations in the size and location of the coronary sinus. Variations in the size and location of the coronary sinus can impact the placement and functioning of a coronary sinus reducer stent. As such, variations in the size and location of the coronary sinus may be used as input to machine learning model(s) 222.

[0052] Anatomical anomalies or abnormalities in the coronary sinus may affect the ability of a clinician to deliver the coronary sinus reducer stent effectively. As such, processing circuitry 204 may determine, based on imaging data 214, anatomical anomalies and / or abnormalities that may impact the ability of a clinician to implant the coronary sinus reducer stent in a preferred location in the coronary sinus. Such anatomical anomalies and / or abnormalities may be used as input to machine learning model(s) 222.

[0053] Processing circuitry 204 may determine, based on imaging data 214, collateral circulation caused by alternative blood vessels that may provide a bypass route for blood flow of the coronary sinus. Such collateral circulation or alternative blood vessels may affect the pressure gradient across the coronary sinus reducer stent once implanted and may be used as input to machine learning model(s) 222.

[0054] It is conceivable that the arterial side of a patient anatomy may also impact the effectiveness of a coronary sinus reducer stent. Processing circuitry 204 may determine, based on imaging data 214, the coronary artery anatomy. Such coronary artery anatomy may also be used as an input to machine learning model(s) 222.

[0055] Processing circuitry 204, executing machine learning model(s) 222, may identify a recommended optimal device position for implantation, geometry, and / or size, based on imaging data 214. Machine learning model(s) 222 may be trained using a longitudinal training data set with a ground truth of previous patients. Computing device 150, executing machine learning model(s) 222, may determine or predict an FFR and / or pressure gradient across the coronary sinus reducer stent at various potential locations within the coronary sinus and identify a recommended location, for example, based on the predicted FFR and / or pressure gradient. Processing circuitry 204 may store the recommended location in recommendation(s) 226 and may output an indication identifying the recommended location, for example, via output device(s) 212 and / or display 206.

[0056] In some examples, processing circuitry 204, executing machine learning model(s) 222, may recommend a specific minimum internal diameter for the coronary sinus reducer stent for the specific patient to obtain an optimal or preferred pressure gradient. The minimum internal diameter of coronary sinus reducer stent may be a minimum diameter inside of the coronary sinus reducer stent and may correspond to a location within the “waist” of an hourglass shaped coronary sinus reducer stent. It should be noted that a coronary sinus reducer stent need not be of an hourglass shape, but may have a difference in interior diameters along a length of the coronary sinus reducer stent. The minimum internal diameter is discussed further herein with respect to FIG. 3.

[0057] In some examples, a clinician may use a noncompliant (NC) plain old balloon angioplasty (POBA) balloon for a balloon implantable coronary sinus reducer stent to adjust the minimum internal diameter of the coronary sinus reducer stent based on the recommended minimum internal diameter. In some examples, processing circuitry 204 executing machine learning model(s) 222 may, for a self-expanding coronary sinus reducer stent having multiple waist size options available, recommend and identify a particular model or stock keeping unit (SKU) code to achieve the recommended minimum internal diameter.

[0058] In some examples, processing circuitry 204 executing machine learning model 222 may predict a likelihood of success of a coronary sinus reducer stent implant procedure for a particular patient based on imaging data 214. For example, processing circuitry 204 executing machine learning model 222 may determine a recommended location and / or internal diameter of a coronary sinus reducer stent. Processing circuitry 204 executing machine learning model 222 may predict a likelihood of relief of angina symptoms if a coronary sinus reducer stent of the recommended size were implanted at the recommended location. Processing circuitry 204 may output an indication of the predicted likelihood of success for a clinician and / or patient to consider when determining whether to proceed with a coronary sinus reducer stent implant procedure. Such an indication may inform the clinician and patient such that cases with little chances of success are not performed, thus reducing risk to the patient and improving efficiency of medical resource usage.

[0059] During a coronary sinus reducer stent implant procedure (peri-procedure), processing circuitry 204 executing machine learning model(s) 222 may provide real time feedback, including metrics versus a generated implant procedure plan or strategy (e.g., a recommended implant location). In some examples, processing circuitry 204 may provide an indication of when the coronary sinus reducer stent is in or at the recommended location. Such indication may include instructions for the clinician to stop maneuvering the coronary sinus reducer stent within the patient and / or may include visual effects on a display, such as changing a color of a representation of the recommended location.

[0060] For example, during the procedure, processing circuitry 204 may obtain periprocedural imaging data (e.g., captured by imager 140 peri-procedure) of imaging data 214 and provide feedback to the clinician regarding the placement of the coronary sinus reducer stent to improve the likelihood that the coronary sinus reducer stent is optimally or preferably positioned. For example, processing circuitry 204 may control display 206 and / or display device 110 to overlay virtual markers on imaging data (e.g., fluoroscopy images obtained by imager 140 in real-time) being displayed on display 206 and / or display device 110 to denote or mark a “landing zone” or recommended implantation location. In some examples, when coronary sinus reducerstent distal and / or proximal radiopaque marker bands align with (or nearly align with) the virtual markers, processing circuitry 204 may change the color of the virtual markers, for example, from a flashing red to a solid green, to denote proper placement of a coronary sinus reducer stent and / or to inform the clinician that inflation of a delivery balloon can begin.

[0061] Processing circuitry 204 may analyze imaging data 214 acquired by imager 140 after a coronary sinus reducer stent is in position at the implantation location. Processing circuitry 204 may use such imaging data to predict the success of the implant procedure based on the actual device position and predicted pressure drop. For example, processing circuitry 204 may estimate the pressure drop from fluid dynamic calculations that estimate blood flow rates upstream and downstream of the narrowing within a coronary sinus reducer stent, for example, using frame count analysis of contrast clearance and the deployed waist diameter. In another example, processing circuitry 204 may estimate the pressure drop using an electrical resistance model similar to resistors in an electrical circuit. Processing circuitry 204 may use the electrical resistance model to determine the hemodynamic impact that the coronary sinus reducer stent has on the coronary sinus flow.

[0062] In some examples, processing circuitry 204 may provide a post-implantation analysis, for example, once the coronary sinus reducer stent is in place, based on imaging data 214 and / or other data 216. In some examples, other data 216 includes wire-based FFR data captured proximally and distally to the coronary sinus reducer stent post-implantation (which, in some examples, may be during the same procedure to implant the coronary sinus reducer stent, but after the coronary sinus reducer stent is in place). In some examples, other data 216 includes image-based FFR data proximal and distal to the coronary sinus reducer stent post-implantation. For example, processing circuitry 204 may processing imaging data 214 to determine the imagebased FFR data. In some examples, the post-implantation analysis may include one or more recommendations for post procedure optimization which may be stored in recommendation(s) 226. For example, processing circuitry 204 may execute machine learning model(s) 222 after the coronary sinus reducer stent is implanted (e.g., any deliver instrument has delivered and / or deployed the coronary sinus reducer stent to the implant location) to determine whether any adjustment to the minimum interior diameter or waist the coronary sinus reducer stent is desirable. This determination may be based on imaging data 214, and / or other data 216, such as the wire-based FFR data mentioned above. For example, processing circuitry 204 executing machine learning model(s) 222 may recommend an increase in the waist of the coronary sinus reducer stent by 0.25mm. Processing circuitry 204 may control display 206 and / or display device 110 to display an indication of the recommendation of the increase.

[0063] In some examples, machine learning model(s) 222 may be continuously trained, using data, such as imaging data 214 from procedures occurring after initial deployment of machine learning model(s) 222 to further train machine learning model(s) 222, thereby continuously improving machine learning model(s) 222. In some examples, machine learning model(s) may not be continuously trained, but may be trained on a closed data set that does not change over time.

[0064] For example, processing circuitry 204 may obtain imaging data 214, which may include angiographic imaging data acquired by imager 140 prior to a coronary sinus reducer stent implant procedure. Processing circuitry 204 may, executing machine learning model(s) 222, determine a whether a procedure on the patient is likely to succeed (e.g., reduce or remove angina symptoms). Processing circuitry 204 may determine a recommended procedure strategy and / or plan. Such strategy and / or plan may include a recommended location for the coronary sinus reducer stent, for example, in the coronary sinus of the patient. The strategy and / or plan may include a device type, such as a device geometry and / or size (e.g., length, outer maximum diameter, inner minimum diameter, etc.). In some examples, the strategy and / or plan may include a specific device make, model, and / or SKU. For example, processing circuitry 204 may, based on imaging data 214 of a patient acquired by imager 140 prior to a coronary sinus reducer stent implant procedure, may identify a patient as being a good candidate for the procedure, recommend an implantation location and device type or geometry, and indicate a predicted chance or level of success. For example, such an indication may include that the patient is 90% likely to have a reduction in angina symptoms and / or the patient is likely to have at least a 75% reduction in frequency and / or intensity of symptoms.

[0065] Processing circuitry 204 may execute one or more machine learning model(s) 222. Machine learning model(s) 222 may be trained using real world data and procedural success data, for example, out to 12 months post-procedure. Processing circuitry 204 may correlate long-term data, such as 12-month post-procedure data back to acute procedural data and provide real time guidance to the clinician.

[0066] 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. In various 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 thatcause 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.

[0067] 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 204 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.

[0068] 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.

[0069] 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 imaging 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.

[0070] 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.

[0071] 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.

[0072] 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 mayexecute user interface 218, which may display imaging data 214 and / or recommendation(s) 226 on display 206 and / or display device 110. A clinician may use recommendation(s) 226 to inform their actions during the medical procedure.

[0073] FIG. 3 is a conceptual diagram illustrating an example coronary sinus reducer stent in a coronary sinus of a patient according to one or more aspects of this disclosure. Coronary sinus reducer stent 310 is depicted in coronary sinus 300. Coronary sinus reducer stent 310 may have a length 302, may have a maximum exterior diameter 304 equal to the largest diameter from one exterior edge of coronary sinus reducer stent 310 to an opposite exterior edge of coronary sinus reducer stent 310. Coronary sinus reducer stent 310 may also have a minimum interior diameter 306 equal to a minimum interior diameter (e.g., within coronary sinus reducer stent 310 itself) from one interior edge of coronary sinus reducer stent 310 to an opposite interior edge of coronary sinus reducer stent 310. In some examples, minimum interior diameter 306 is fixed. In other examples, minimum interior diameter 306 is adjustable. Coronary sinus reducer stent 310 may have a nominal length 312 during which coronary sinus reducer stent 310 has minimum interior diameter 306. In some examples, nominal length 312 may be within a range of lengths having a value of greater than zero to a value less than length 302.

[0074] Coronary sinus reducer stent 310 may be a recommended coronary sinus reducer stent or a recommended type of coronary sinus reducer stent. For example, one of recommendation(s) 226 may include a recommendation for coronary sinus reducer stent 310. In some examples, the recommendation may be for a coronary sinus reducer stent having a specific length, such as length 302. In some examples, the recommendation may be for a coronary sinus reducer stent having a specific maximum exterior diameter, such as maximum exterior diameter 304. In some examples, the recommendation may be for a coronary sinus reducer stent having a specific minimum interior diameter, such as minimum interior diameter 306. In some examples, the recommendation may be for a specific make, model, and / or SKU of a coronary sinus reducer stent. As such, a type of coronary sinus reducer stent should be understood to include a length, a maximum exterior diameter, a minimum interior diameter, a make, a model, a SKU, and / or the like of a coronary sinus reducer stent.

[0075] FIG. 3 also depicts recommended location 308 of recommendation(s) 226. Recommended location 308 may be a recommended location for coronary sinus reducer stent 310 to be implanted in coronary sinus 300. In some examples, processing circuitry 204 may output an indication of recommended location, for example, such as the dashed lines depicted within coronary sinus 300. For example, processing circuitry 204 may control display 206 and / or display device 110 to overlay virtual markers on imaging data (e.g., fluoroscopy images obtained by imager 140 in real-time) being displayed on display 206 and / or display device 110 to denoteor mark a “landing zone” or recommended implantation location. Processing circuitry 204 may control display 206 and / or display device 110 to overlay an indicator or color over imaging data 214 being displayed to identify recommended location 308. In some examples, processing circuitry 204 may change the color of the virtual markers, for example, from a flashing red to a solid green, to denote proper placement of a coronary sinus reducer stent and / or to inform the clinician that inflation of a delivery balloon can begin.

[0076] In some examples, processing circuitry 204 may track coronary sinus reducer stent 310 when it is being implanted in coronary sinus 300 (e.g., peri-procedure) and may output an indication that coronary sinus reducer stent 310 is at or within the recommended location when coronary sinus reducer stent 310 reaches recommended location 308. For example, processing circuitry 204 may utilize image processing techniques to determine a location of coronary sinus reducer stent 310 within imaging data 214, may monitor the location of one or more device radiopaque markers of coronary sinus reducer stent 310 (not shown) or of a delivery device for coronary sinus reducer stent 310, and / or may use other techniques to determine the location of coronary sinus reducer stent 310 with respect to target location 308. For example, processing circuitry 204 may control display 206 and / or display device 110 to change a color of target location 308 and / or otherwise change the appearance of target location 308 once coronary sinus reducer stent 310 is within target location 308.

[0077] In some examples, recommended location 308 may be longer than length 302 of coronary sinus reducer stent 310 to allow some variance in actual implantation of coronary sinus reducer stent 310, for example, due to inherent imprecision of implanting a device in coronary sinus of a living being. In some examples, recommended location 308 may be a same length as length 302.

[0078] Once coronary sinus reducer stent 310 is in place at the recommended location 308, a clinician may deploy or implant coronary sinus reducer stent 310. Coronary sinus reducer stent 310 should then increase backpressure into the precapillary arteriolar system which facilitates dilation of the constricted subendocardial capillaries. This improves capillary perfusion in the subendocardium of the ischemic territory, which improves contractility and increases oxygen consumption. By doing so, coronary sinus reducer stent 310 may improve blood flow to the heart muscle and alleviate symptoms of angina. It should be noted that the existence of branches off of the coronary sinus along length 302 of coronary sinus reducer stent 310 (not shown) may reduce the effectiveness of the stent in alleviating angina symptoms. By predicting a pressure gradient or pressure drop across coronary sinus reducer stent 310, for example, based on imaging data 214, processing circuitry 204 may determine recommendation(s) 226.

[0079] In some examples, coronary sinus reducer stent 310 may include mesh material such that the effect of coronary sinus reducer stent 310 on the pressure gradient across coronary sinus reducer stent 310 is not immediately complete and that such effect may increase over time, such as 6-8 weeks, while the mesh material becomes endothelialized.

[0080] FIG. 4 is conceptual diagram illustrating example training data for one or more machine learning models according to one or more aspects of this disclosure. Computing device 150, server 160, imager 140, and / or any other device or system capable of doing so may train machine learning model(s) 222 using training data 400. In some examples, training data 400 may be longitudinal data including data associated with a group of previous patients captured at different times.

[0081] For example, training data 400 may include pre-implant procedure data 402. Preimplant procedure data 402 may be data for a patient of the group of previous patients captured prior to a coronary sinus reducer stent implant procedure of that patient. For example, for patients of the group, pre-implant procedure data 402 may include a pre-procedure angiographic assessment. The pre-procedure angiographic assessment may include pre-implant procedure angiographic data and may include a 3D model (e.g., reconstruction) of the full coronary sinus tree. In some examples, the 3D model may include an image-based FFR, wire-based FFR, and / or pressure gradient data including full vessel pull back. Pre-implant procedure data 402 may include anatomical characteristics such as main coronary sinus vessel length, diameter, collateral vessels, vessel disease state, and / or the like.

[0082] Pre-implant procedure data 402 may include pre-procedure CT data and / or other imaging data. Pre-implant procedure data 402 may include pre-procedure patient meta data, such as patient demographics and any comorbidity data. Such comorbidity data may include any cardiovascular comorbidity data, such as the existence of or state of vascular disease.

[0083] Training data 400 may include peri-implant procedure data 404. Peri-implant procedure data 404 may be data for a patient of the group of previous patients captured during a coronary sinus reducer stent implant procedure of that patient. In some examples, peri-implant procedure data 404 includes imaging data captured during the coronary sinus reducer stent implant procedure. In some examples, imaging data 214 captured during a coronary sinus reducer stent implant procedure, such as peri-implant procedure data 404 may be recorded, for example, by processing circuitry 204, and co-registered with pre-implant procedure data 402 and / or data captured prior to the coronary sinus reducer stent implant procedure (e.g., of pre-implant procedure data 402) and with data captured after the coronary sinus reducer stent implant procedure (e.g., of post-implant procedure data 406 and / or long-term data 408).

[0084] Coronary sinus reducer stent treatment data of peri-implant procedure data 404 may include a coronary sinus reducer type, make, model, SKU, device details such as maximum external diameter, minimum internal diameter, device length, etc. Coronary sinus reducer stent treatment data may include specific coronary sinus reducer FFR wire data, which, in some examples, may be collected specifically for the purpose of training machine learning model(s) 222. For example, wire-based FFR data may be collected prior to a coronary sinus reducer stent implant procedure (e.g., of pre-implant procedure data 402) and may include a pressure pull back gradient recording. Then, after the coronary sinus reducer stent placement, a FFR wire may used to determine distal and proximal implant pressure data, which may include pressure pull back gradient recordings (e.g., of post-implant procedure data 406 and / or long-term data 408). This data may be used as ground truth data for imaging data 214 (e.g., angiography data) based predictions. In some examples, the coronary sinus reducer stent placement FFR wire data collected after the coronary sinus reducer stent implant procedure may include long-term data 408 recorded at a plurality of periodic intervals after implantation, such as at 3, 6 and / or 12- month intervals. In some examples, in lieu of, or in addition to wire-based FFR data, such training data may include image-based FFR data.

[0085] Training data may include post-implant procedure data 406. For example, postimplant procedure data 406 may include imaging data 214, such as angiography data, CT data, a 3D reconstruction with image-based FFR data and full vessel pull back, wire-based FFR data, and / or the like.Post-coronary sinus reducer stent implant data may include success criteria, which may include a patient assessment of reduction of angina symptom frequency and / or intensity, coronary sinus reducer stent position, and / or coronary sinus reducer stent dimensions. It should be noted that post-implant procedure data 406 and long-term data 408 are both data collected after the coronary sinus reducer stent is implanted (e.g., deployed) at an implantation location within the patient. It should be noted that as used herein, data collected after the coronary sinus reducer stent has been delivered and deployed to the implant location, even during the same medical procedure (e.g., during the removal of any delivery instrument from the anatomy of the patient) is considered post-implant procedure data 406 because the coronary sinus reducer stent is already implanted. As such, in some examples, post-implant procedure data 406 and long-term data 408 may be represented as a single block of data.

[0086] Training data may include long-term data 408. Long-term data 408 may be data captured or gathered from patients at least a period of time after the coronary sinus reducer stent implant procedure, for example after 3 months, after 6 months, or after 12 months. Long-term data 408 may include imaging data 214, such as CT data and / or angiography data. Long-termdata 408 may include post-procedure angiography and / or CT data collected at approximately a pre-defined point in time post-procedure, such as 3 months, 6 months, 12 months, or other length of time. In some examples, a 3D model may be generated based on long-term angiography data, including image-based FFR and full vessel pull back. In some examples, long-term data 408 may include wire-based FFR measurements obtained of the distal and proximal implant pressure and may include a pressure pull back gradient recording. In some examples, long-term data 408 may further include OCT data and / or IVUS data.

[0087] In some examples, training data 400 may include angina ratings, such as Canadian Cardiovascular Society (CCS) Angina Classification System ratings and / or Seattle Angina Questionnaire (SAQ) ratings, for example, collected by a clinician at multiple time points, such as pre-procedure, post-procedure, and 1 or more time periods after the procedure, such as 3 months, 6 months, 12 months, 24 months and / or the like. As such, angina ratings may form part of pre-implant procedure data 402, post-implant procedure data 406 and / or long-term data 408. Long-term data 408 may, in some cases, demonstrate improvements over time of CCS and / or SAQ ratings.

[0088] Training data 400 thus includes pre-implant procedure data 402, peri-implant procedure data 404, post-implant procedure data 406, and long-term data 408 which may be used to train machine learning model(s) 222 to inform success metrics and determine recommendation(s) 226. In some examples, machine learning model(s) 222 may be continuously trained as data from new patients and / or procedures becomes available.

[0089] FIG. 5 is a flow diagram of example techniques for determining a likelihood of success of a coronary sinus reducer stent implant procedure for a patient 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 performed by any of, or any combination of, processing circuitry of devices depicted in FIG. 1 or capable of performing such techniques.

[0090] Processing circuitry 204 may obtain the imaging data (500). For example, processing circuitry 204 may read imaging data 214 from memory 202 or receive imaging data 214 from imager 140 via, for example, network interface 208.

[0091] Processing circuitry 204 may execute one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient (502). For example, processing circuitry 204 may execute machine learning model(s) 222, using imaging data 214 as input data, to determine a likelihood of success of a potential coronary sinus reducer stent implant procedure for the patient. In some examples, processing circuitry 204 may store the likelihood of success in recommendation(s) 226.

[0092] Processing circuitry 204 may output an indication of the likelihood of success (504). For example, processing circuitry 204 may control display 206 and / or display device 110 to display the indication of the likelihood of success. In some examples, the indication of the likelihood of success may include a recommendation of whether or not to perform the coronary sinus reducer stent implant procedure. In some examples, the indication of the likelihood of success may include an indication that the patient is a borderline patient for a coronary sinus reducer stent implant procedure (e.g., the patient is neither a good candidate for the procedure or a poor candidate for the procedure). In some examples, the indication of the likelihood of success may include an alphabetic, numeric, or alphanumeric representation of the likelihood of success. In some examples, the indication of the likelihood of success may include a percentage of the likelihood of success (e.g., an 80% likelihood of success).

[0093] In some examples, processing circuitry 204 may execute one or more machine learning model(s) 222 to determine, based on imaging data 214, a recommended coronary sinus reducer stent and a recommended location within anatomy of the patient for implanting the recommended coronary sinus reducer stent. In such examples, processing circuitry 204 may output an indication of the recommended coronary sinus reducer stent and a recommended location.

[0094] In some examples, the recommended coronary sinus reducer stent includes at least one of a recommended length, a recommended maximum exterior diameter, a recommended minimum interior diameter, a recommended make, a recommended model, or a recommended stock keeping unit (SKU) code.

[0095] In some examples, imaging data 214 includes pre-procedure imaging data, the preprocedure imaging data being obtained prior to the coronary sinus reducer stent implant procedure, and peri-procedure imaging data, the peri-procedure imaging data being obtained during the coronary sinus reducer stent implant procedure. In some examples, processing circuitry 204 may execute one or more machine learning model(s) 222 to determine, based on the peri-procedure imaging data, that the coronary sinus reducer stent is in the recommended location. In some examples, processing circuitry 204 may output an indication that the coronary sinus reducer stent is in the recommended location.

[0096] In some examples, processing circuitry 204 may execute the one or more machine learning models to determine, based at least one of the imaging data, image-based FFR data, or wire-based FFR data of the patient, a recommended adjustment to a minimum diameter of the coronary sinus reducer stent. Processing circuitry 204 may output an indication of the adjustment. In some examples, imaging data 214 includes at least one of angiography data or computed tomography data.

[0097] In some examples, one or more machine learning model(s) 222 are trained on training data 400 obtained from a plurality of patients, training data 400 comprising, for each of the plurality of patients, imaging data obtained prior to a coronary sinus reducer stent implant procedure (e.g., pre-implant procedure data 402), imaging data obtained during the coronary sinus reducer stent implant procedure (e.g., peri-implant procedure data 404), and imaging data obtained after the coronary sinus reducer stent implant procedure (e.g., post-implant procedure data 406 and / or long-term data 408). In some examples, the imaging data obtained after the coronary sinus reducer stent implant procedure includes imaging data obtained at least 3 months after the coronary sinus reducer stent implant procedure (e.g., long-term data 408). In some examples, training data 400 further includes, for each of the plurality of patients, wire-based FFR data or image-based FFR data obtained prior to the coronary sinus reducer stent implant procedure, and wire-based FFR data or image-based FFR data obtained after the coronary sinus reducer stent implant procedure. In some examples, training data 400 further includes, for each of the plurality of patients, at least one of patient meta data, an angina classification rating prior to the coronary sinus reducer stent implant procedure, an angina classification rating after the coronary sinus reducer stent implant procedure, optical coherence tomography data obtained after the coronary sinus reducer stent implant procedure, or intravascular ultrasound data obtained after the coronary sinus reducer stent implant procedure.

[0098] In some examples, one or more machine learning model(s) 222 are executed further based on wire-based FFR data or image-based FFR data obtained from the patient prior to the coronary sinus reducer stent implant procedure. For example, the FFR data obtained from the patient prior to the coronary sinus reducer stent implant procedure is used as input data to one or more machine learning model(s) 222, along with imaging data 214. In some examples, one or more machine learning model(s) 222 are executed further based on patient meta data of the patient. For example, patient demographic data and / or patient comorbidities may be used as input data to or more machine learning model(s) 222, along with imaging data 214.

[0099] 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 recommendation(s) 226. Recommendation(s) 226 may include recommendations on whether a patient is a good candidate for a coronary sinus reducer stent implant procedure, such as a likelihood of success of the procedure for that patient, and / or a recommended a particular type of coronary sinus reducer stent and / or a location for implantation of the coronary sinus reducer stent for that patient. One or more of computing device 150 and / or server 160 may train, store, and / or utilize machine learningmodel 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 CNN, e.g., ResNet-18, may be used. Some nonlimiting examples of machine learning techniques include Support Vector Machines, K-Nearest Neighbor algorithm, and Multi-layer Perceptron.

[0100] 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 imaging 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.

[0101] 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 imaging data 214, characteristics of a coronary sinus reducer stent, locations of the coronary sinus reducer stent, and / or measures of success of the coronary sinus reducer stent procedure. 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.

[0102] 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 of a likelihood of success of a coronary sinus reducer stent implant procedure, a recommended coronary sinus reducer stent, and / or a recommended location for implantation of the coronary sinus reducer stent.

[0103] As shown in the example above, by applying machine learning model 600 to input data such as imaging data 214, processing circuitry 204 is able to determine whether a patient is a good candidate for a coronary sinus reducer stent implant procedure and / or specific strategies forsuch a procedure including characteristic(s) of a recommended coronary sinus reducer a recommended treatment strategy for a lesion. This may screen out patients for whom a coronary sinus reducer stent is unlikely to improve their symptoms, thereby eliminating unnecessary risk to the patient and conserving medical facility resources. This may also improve the ability of a clinician to successfully treat medical conditions, such as angina, with a coronary sinus reducer stent, thereby improving patient outcomes.

[0104] 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 700 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 semi-supervised, 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, convolutional neural network (CNN), recursive neural network (RNN), long short-term memory (LSTM), ensemble network, to name only a few examples.

[0105] 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, any of, or all of, training data 400 (FIG. 4). For example, training data 772 may include imaging data, FFR data, patient meta data, coronary sinus reducer stent characteristic data, and / or the like.

[0106] In some examples, training data 772 may include images of anatomy of the current patient, images of anatomy of other patients, and / or the like. For example, training data 772 may include data from past medical procedures performed on a plurality of patients having different anatomy, different prior medical procedures, annotations or tags, other training data mentioned herein, and / or the like. In some examples, with the subsequent medical follow up of patients, treatment and outcome data may be used to train machine learning model (s) 222 or machine learning model 774 to recommend whether a potential patient is a good candidate for a coronary sinus reducer stent implant procedure, and / or a type of coronary sinus reducer stent to be used and / or a location for the coronary sinus reducer stent to be implanted.

[0107] 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 774based 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 a likelihood of success of a coronary sinus reducer stent implant procedure, a type of coronary sinus reducer stent to be used and / or a location for the coronary sinus reducer stent to be implanted is determined.

[0108] 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.

[0109] Training data 802 may include pre-implant procedure data 402, peri-implant procedure data 404, post-implant procedure data 406, and / or long-term data 408, as discussed above with respect to FIG. 4.

[0110] The input training data 802 may be pre-processed 804, for example, by a neural network. For example, imaging data 214 (e.g., angiography data and / or CT data) may be processed and / or interrogated. This may include processing wire-based FFR (with pull back) data, other pressure drop data, patient meta data, and / or the like. In some examples, such data may be matched to long-term data, such as outcomes at 6-12 months after a coronary sinus reducer stent implant procedure, to measure efficacy of an indexed procedure. For example, computing device 150 may use such automated pre-processing to prepare and / or validate the data for training.[oni] 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 a likelihood of success of a coronary sinus reducer implant procedure and / or coronary sinus reducer stent implant procedure strategies, such as a type of coronary sinus reducer stent to be used and / or a location for the coronary sinus reducer stent to be implanted. It should be noted that output of machine learning model 808 may be used for further modeling 806.

[0112] 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 foregoinglogic 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.

[0113] 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.

[0114] This disclosure includes the following non-limiting examples.

[0115] Example 1. A medical system comprising: memory configured to store imaging data of a patient; and processing circuitry communicatively coupled to the memory, the processing circuitry being configured to: obtain the imaging data; execute one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient; and output an indication of the likelihood of success.

[0116] Example 2. The medical system of example 1, wherein the processing circuitry is further configured to: execute the one or more machine learning models to determine, based on the imaging data, a recommended coronary sinus reducer stent and a recommended location within anatomy of the patient for implanting the recommended coronary sinus reducer stent; and output an indication of the recommended coronary sinus reducer stent and a recommended location.

[0117] Example 3. The medical system of example 2, wherein the recommended coronary sinus reducer stent comprises at least one of a recommended length, a recommendedmaximum exterior diameter, a recommended minimum interior diameter, a recommended make, a recommended model, or a recommended stock keeping unit (SKU) code.

[0118] Example 4. The medical system of any of examples 2-3, wherein the imaging data comprises pre- procedure imaging data, the pre-procedure imaging data being obtained prior to the coronary sinus reducer stent implant procedure, and peri-procedure imaging data, the periprocedure imaging data being obtained during the coronary sinus reducer stent implant procedure.

[0119] Example 5. The medical system of example 4, wherein the processing circuitry is further configured to: execute the one or more machine learning models to determine, based on the peri-procedure imaging data, that a coronary sinus reducer stent is in the recommended location; and output an indication that the coronary sinus reducer stent is in the recommended location.

[0120] Example 6. The medical system of any of examples 1-5, wherein the processing circuitry is further configured to: execute the one or more machine learning models to determine, based at least one of the imaging data or wire-based fractional flow reserve data of the patient, a recommended adjustment to a minimum diameter of the coronary sinus reducer stent; and output an indication of the recommended adjustment.

[0121] Example 7. The medical system of any of examples 1-6, wherein the imaging data comprises at least one of angiography data or computed tomography data.

[0122] Example 8. The medical system of any of examples 1-7, wherein the one or more machine learning models are trained on training data obtained from a plurality of patients, the training data comprising, for each of the plurality of patients, imaging data obtained prior to the coronary sinus reducer stent implant procedure, imaging data obtained during the coronary sinus reducer stent implant procedure, and imaging data obtained after the coronary sinus reducer stent implant procedure.

[0123] Example 9. The medical system of example 8, wherein the imaging data obtained after the coronary sinus reducer stent implant procedure comprises imaging data obtained at least 3 months after the coronary sinus reducer stent implant procedure.

[0124] Example 10. The medical system of example 8 or example 9, wherein the training data further comprises, for each of the plurality of patients, wire-based fractional flow reserve (FFR) data obtained prior to the coronary sinus reducer stent implant procedure, and wire-based FFR data obtained after the coronary sinus reducer stent implant procedure.

[0125] Example 11. The medical system of any of examples 8-10, wherein the training data further comprises, for each of the plurality of patients, at least one of patient meta data, a angina classification rating prior to the coronary sinus reducer stent implant procedure, an anginaclassification rating after the coronary sinus reducer stent implant procedure, optical coherence tomography data obtained after the coronary sinus reducer stent implant procedure, or intravascular ultrasound data obtained after the coronary sinus reducer stent implant procedure.

[0126] Example 12. The medical system of any of examples 1-11, wherein the one or more machine learning models are executed further based on wire-based fractional flow reserve (FFR) data obtained from the patient prior to the coronary sinus reducer stent implant procedure.

[0127] Example 13. The medical system of any of examples 1-12, wherein the one or more machine learning models are executed further based on patient meta data of the patient.

[0128] Example 14. A method comprising: obtaining, by processing circuitry, imaging data of a patient; executing, by the processing circuitry, one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient; and outputting, by the processing circuitry, an indication of the likelihood of success.

[0129] Example 15. The method of example 14, further comprising: executing, by the processing circuitry, the one or more machine learning models to determine, based on the imaging data, a recommended coronary sinus reducer stent and a recommended location within anatomy of the patient for implanting the recommended coronary sinus reducer stent; and outputting, by the processing circuitry, an indication of the recommended coronary sinus reducer stent and a recommended location.

[0130] Example 16. The method of example 15, wherein the recommended coronary sinus reducer stent comprises at least one of a recommended length, a recommended maximum exterior diameter, a recommended minimum interior diameter, a recommended make, a recommended model, or a recommended stock keeping unit (SKU) code.

[0131] Example 17. The method of any of examples 15-16, wherein the imaging data comprises pre- procedure imaging data, the pre-procedure imaging data being obtained prior to the coronary sinus reducer stent implant procedure, and peri-procedure imaging data, the periprocedure imaging data being obtained during the coronary sinus reducer stent implant procedure.

[0132] Example 18. The method of example 17, further comprising: executing, by the processing circuitry, the one or more machine learning models to determine, based on the periprocedure imaging data, that the coronary sinus reducer stent is in the recommended location; and outputting, by the processing circuitry, an indication that the coronary sinus reducer stent is in the recommended location.

[0133] Example 19. The method of any of examples 14-18, wherein the one or more machine learning models are trained on training data obtained from a plurality of patients, thetraining data comprising, for each of the plurality of patients, imaging data obtained prior to a coronary sinus reducer stent implant procedure, imaging data obtained during the coronary sinus reducer stent implant procedure, and imaging data obtained after the coronary sinus reducer stent implant procedure.

[0134] Example 20. A non-transitory computer-readable storage medium storing instructions, which when executed cause processing circuitry to: obtain imaging data of a patient; execute one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient; and output an indication of the likelihood of success.

[0135] 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 imaging data of a patient; and processing circuitry communicatively coupled to the memory, the processing circuitry being configured to: obtain the imaging data; execute one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient; and output an indication of the likelihood of success.

2. The medical system of claim 1, wherein the processing circuitry is further configured to: execute the one or more machine learning models to determine, based on the imaging data, a recommended coronary sinus reducer stent and a recommended location within anatomy of the patient for implanting the recommended coronary sinus reducer stent; and output an indication of the recommended coronary sinus reducer stent and a recommended location.

3. The medical system of claim 2, wherein the recommended coronary sinus reducer stent comprises at least one of a recommended length, a recommended maximum exterior diameter, a recommended minimum interior diameter, a recommended make, a recommended model, or a recommended stock keeping unit (SKU) code.

4. The medical system of any of claims 2-3, wherein the imaging data comprises preprocedure imaging data, the pre-procedure imaging data being obtained prior to the coronary sinus reducer stent implant procedure, and peri-procedure imaging data, the peri-procedure imaging data being obtained during the coronary sinus reducer stent implant procedure.

5. The medical system of claim 4, wherein the processing circuitry is further configured to: execute the one or more machine learning models to determine, based on the periprocedure imaging data, that a coronary sinus reducer stent is in the recommended location; and output an indication that the coronary sinus reducer stent is in the recommended location.

6. The medical system of any of claims 1-5, wherein the processing circuitry is further configured to: execute the one or more machine learning models to determine, based at least one of the imaging data or wire-based fractional flow reserve data of the patient, a recommended adjustment to a minimum diameter of the coronary sinus reducer stent; and output an indication of the recommended adjustment.

7. The medical system of any of claims 1-6, wherein the imaging data comprises at least one of angiography data or computed tomography data.

8. The medical system of any of claims 1-7, wherein the one or more machine learning models are trained on training data obtained from a plurality of patients, the training data comprising, for each of the plurality of patients, imaging data obtained prior to the coronary sinus reducer stent implant procedure, imaging data obtained during the coronary sinus reducer stent implant procedure, and imaging data obtained after the coronary sinus reducer stent implant procedure.

9. The medical system of claim 8, wherein the imaging data obtained after the coronary sinus reducer stent implant procedure comprises imaging data obtained at least 3 months after the coronary sinus reducer stent implant procedure.

10. The medical system of claim 8 or claim 9, wherein the training data further comprises, for each of the plurality of patients, wire-based fractional flow reserve (FFR) data or image-based FFR data obtained prior to the coronary sinus reducer stent implant procedure, and wire-based FFR data or image-based FFR data obtained after the coronary sinus reducer stent implant procedure.

11. The medical system of any of claims 8-10, wherein the training data further comprises, for each of the plurality of patients, at least one of patient meta data, a angina classification rating prior to the coronary sinus reducer stent implant procedure, an angina classification rating after the coronary sinus reducer stent implant procedure, optical coherence tomography data obtained after the coronary sinus reducer stent implant procedure, or intravascular ultrasound data obtained after the coronary sinus reducer stent implant procedure.

12. The medical system of any of claims 1-11, wherein the one or more machine learning models are executed further based on wire-based fractional flow reserve (FFR) data or imagebased FFR data obtained from the patient prior to the coronary sinus reducer stent implant procedure.

13. The medical system of any of claims 1-12, wherein the one or more machine learning models are executed further based on patient meta data of the patient.

14. A method comprising: obtaining, by processing circuitry, imaging data of a patient; executing, by the processing circuitry, one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient; and outputting, by the processing circuitry, an indication of the likelihood of success.

15. A non-transitory computer-readable storage medium storing instructions, which when executed cause processing circuitry to: obtain imaging data of a patient; execute one or more machine learning models to determine, based on the imaging data, a likelihood of success of a coronary sinus reducer stent implant procedure for the patient; and output an indication of the likelihood of success.

Images