Plaque shift and / or carina shift during stent placement and associated systems, devices, and methods

Abstract

An apparatus includes a processor circuit that communicates with an intravascular imaging catheter. The processor circuit controls the intravascular imaging catheter to obtain intravascular images during movement through a first blood vessel of a patient before a stent is deployed inside the first blood vessel. The processor circuit determines first data representative of plaque within the first blood vessel or a second data representative of a second blood vessel connected to the first blood. The processor circuit determines a likelihood for a carina between the first blood vessel and the second vessel to shift when the stent is deployed and / or the plaque to shift when the stent is deployed. The processor circuit outputs, to a display in communication with the processor circuit, a screen display based on the likelihood for at least one of the carina to shift or the plaque to shift.

Description

PLAQUE SHIFT AND / OR CARINA SHIFT DURING STENT PLACEMENT AND ASSOCIATED SYSTEMS, DEVICES, AND METHODS.TECHNICAL FIELD

[0001] The present disclosure relates generally to intravascular imaging (e.g., intravascular ultrasound (IVUS), optical coherence tomography (OCT), etc.) using an intravascular imaging catheter for generating images of a blood vessel. In particular, measurements derived from the intravascular images may be used to predict the likelihood of a plaque shift of carina shift during stent placement.BACKGROUND

[0002] Intravascular imaging (IVI) (such as intravascular ultrasound (IVUS) or optical coherence tomography (OCT) imaging) is widely used in interventional cardiology as a diagnostic tool for assessing a diseased vessel, such as an artery, within the human body to determine the need for treatment, to guide the intervention, and / or to assess its effectiveness. An IVI device including one or more ultrasound transducers is passed into the vessel and guided to the area to be imaged. The transducers emit ultrasonic energy in order to create an image of the vessel of interest. Ultrasonic waves are partially reflected by discontinuities arising from tissue structures (such as the various layers of the vessel wall), red blood cells, and other features of interest. Echoes from the reflected waves are received by the transducer and passed along to an IVI (e.g., IVUS or OCT) imaging system. The imaging system processes the received ultrasound echoes to produce a cross-sectional image of the vessel where the device is placed.

[0003] Peripheral and coronary vascular procedures, such as stenting, often involve IVI. A stent is a dense (e.g., metallic) object that may be placed in a vessel or lumen to hold the vessel or lumen open to a particular diameter, to counteract the effects of an occlusion, plaque, or compression. Placing a stent can have secondary effects that may be harmful to a patient. For example, plaque may shift or flow beyond the ends of the stent, and / or cause a carina to shift.

[0004] Currently, determining the likelihood of plaque shift or carina shift is accomplished by error-prone human decision-making. A physician may have to judge by eye, looking at various intravascular images, and / or evaluating physical measurements that are not directly related to the likelihood of a shift or that are not easily combined using professional human judgment.

[0005] Thus, a need exists for improved prediction of plaque shift and carina shift during stent placement.

[0006] The information included in this Background section of the specification, including any references cited herein and any description or discussion thereof, is included for technical reference purposes only and is not to be regarded as subject matter by which the scope of the disclosure is to be bound.SUMMARY

[0007] Disclosed herein are systems, devices, and methods for predicting the likelihood of a plaque shift or a carina shift during stent placement. Predicting the likelihood of a plaque or carina shift allows a physician or other medical professional to revise a treatment plan, such as stenting or balloon angioplasty, to mitigate plaque shifting and / or carina shifting and the associated harmful effects. Prediction of a plaque or carina shift may be based on a number of features, including lumen and vessel borders of a main vessel and side branch of a blood vessel, main vessel or side branch lumen volume / area, side branch angle, plaque type / softness, plaque shape, etc. In some instances, prediction is facilitated by a look-up table and / or machine learning model. In addition to predicting the likelihood of plaque or carina shift, a region where such a shift is likely may be determined. A display may be provided to a physician which may include various intravascular images and longitudinal views. The plaque and carina shift prediction system may also indicate regions on the displayed images where a plaque or carina shift is likely, and a flag / warning may be included with an explanation warning about the likelihood of plaque or carina shift. The plaque and carina shift prediction system has particular but not exclusive utility to predict the likelihood of a plaque or carina shift if a stent is placed in a blood vessel. Thus, a physician planning a treatment utilizing a stent may increase the odds of a successful treatment by updating the treatment plan based on the likelihood of plaque or carina shifting.

[0008] This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to limit the scope of the claimed subject matter. A more extensive presentation of features, details, utilities, and advantages of the image co-registration system, as defined in the claims, is provided in the following written description of various aspects of the disclosure and illustrated in the accompanying drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Illustrative aspects of the present disclosure will be described with reference to the accompanying drawings, of which:

[0010] Figure 1 is a diagrammatic schematic view of an intraluminal imaging system, according to aspects of the present disclosure.

[0011] Figure 2 is a schematic diagram of a processor circuit, according to aspects of the present disclosure.

[0012] Figure 3A illustrates a blood vessel with a plaque, according to aspects of the present disclosure.

[0013] Figure 3B illustrates a blood vessel with a plaque, as depicted in Fig. 3A, with a stent expanded inside a main vessel to restore flow, according to aspects of the present disclosure.

[0014] Figure 4A illustrates a blood vessel 400 with plaques of varying softness and / or composition, according to aspects of the present disclosure.

[0015] Figure 4B illustrates the blood vessel with plaques of varying softness and / or composition, as depicted in Fig. 4A, with a stent expanded inside a main vessel to restore flow, according to aspects of the present disclosure.

[0016] Figure 5A illustrates a blood vessel with a focal lesion, according to aspects of the present disclosure.

[0017] Figure 5B illustrates a blood vessel with a focal lesion with a stent expanded in the blood vessel, according to aspects of the present disclosure.

[0018] Figure 5C illustrates a blood vessel with a diffuse lesion, according to aspects of the present disclosure.

[0019] Figure 5D illustrates a blood vessel with a diffuse lesion with a stent expanded in the blood vessel, according to aspects of the present disclosure.

[0020] Figure 6 is an example display of blood vessel images with a warning for plaque or carina shift, according to aspects of the present disclosure.

[0021] Figure 7 is a diagrammatic schematic view of a plaque and carina shift prediction system, according to aspects of the present disclosure.

[0022] Figure 8A illustrates a blood vessel with a first side branch angle, according to aspects of the present disclosure.

[0023] Figure 8B illustrates a blood vessel with a second side branch angle, according to aspects of the present disclosure.

[0024] Figure 9A is an illustration of the cross-sectional image of a blood vessel at the first equivalent imaging planes, according to aspects of the present disclosure.

[0025] Figure 9B is an illustration of the cross-section of a blood vessel at the second imaging plane of Fig. 8 A, according to aspects of the present disclosure.

[0026] Figure 9C is an illustration of the cross-section of a blood vessel at the second imaging plane of Fig. 8B, according to aspects of the present disclosure.

[0027] Figure 10 is a diagrammatic schematic view of carina shift and / or plaque shift prediction module including look-up tables and / or decision trees, according to aspects of the present disclosure.

[0028] Figure 11 is a diagrammatic schematic view of carina shift and / or plaque shift prediction module including user-defined thresholds, according to aspects of the present disclosure.

[0029] Figure 12 is a look-up table based on side branch angle and lumen volume / area, according to aspects of the present disclosure.

[0030] Figure 13 is a look-up table based on degree of plaque shape and plaque type, according to aspects of the present disclosure.

[0031] Figure 14 is a diagrammatic schematic view of a training system for a machine learning model according to aspects of the present disclosure.

[0032] Figure 15 is a second diagrammatic schematic view of a training system for a machine learning model, according to aspects of the present disclosure.

[0033] Figure 16 is a diagrammatic schematic view of a plaque and carina shift prediction system including a machine learning model receiving a plurality of vessel features and / or measurements as input, according to aspects of the present disclosure.

[0034] Figure 17 is a diagrammatic schematic view of a plaque and carina shift prediction system including a machine learning model receiving images and / or views, according to aspects of the present disclosure.

[0035] Figure 18 is a diagrammatic schematic view of a plaque and carina shift prediction system including a machine learning model for object detection or tissue characterization, according to aspects of the present disclosure.

[0036] Figure 19 is a diagrammatic schematic of a deep learning network configuration, according to aspects of the present disclosure.DETAILED DESCRIPTION

[0037] When formulating a treatment plan, a physician will often employ intravascular imaging as part of the treatment planning procedure. For example, imaging may be needed for identifying a location for stent placement. By placing a stent, blood flow through a vessel that was previously obstructed by plaque. However, placement of a stent may cause plaque to shift or flow beyond the ends of the stent. In addition, when a stent is placed near a bifurcation of a vessel into a main vessel and side branch, a carina may be deflected by the stent and impede the flow of blood to the side branch. In scenarios where plaque is located within or near a side branch, placing a stent may cause the plaque and / or carina shift, resulting in reduction of blood flow through the side branch. In other words, the shift may occlude the main vessel or side branch. To evaluate the response of plaque and / or carina to stenting or related procedures, physician’s may often be required to employ heuristic assessments and use measurements that do not directly relate to the likelihood of occlusion of a branch of the blood vessel.

[0038] The systems, methods, and devices for plaque and carina shift prediction, as described herein, predict the likelihood of a plaque or carina shift based on various vessel features. For sake of explanation, the description herein focuses on a stent placement procedure, but other treatments are conceived of and encompassed, e.g., balloon angioplasty. To predict a plaque or carina shift features of a blood vessel are identified and measurements are made based on those features. For example, the contours of the blood vessel or lumen of the blood vessel may be used to compute various lumen or vessels areas / volumes. Many other features may be used and measurements made.

[0039] By systematically predicting the likelihood of a plaque or carina shift treatment plans, such as stent placement, may be formulated to avoid harmful secondary effects. For example, when plaque flow is likely, a longer stent may be employed. In addition, automated prediction of plaque and carina shift will reduce the time required to develop treatment plans.

[0040] Examples of calculation or estimation of pullback speed can be found for example in U.S. Application No. 16 / 542,001, filed August 15, 2019, and U.S. Application No. 16 / 662,847, filed October 24, 2019, incorporated by reference as though fully set forth herein.

[0041] The devices, systems, and methods described herein can include one or more features described in U.S. Provisional App. No. 63 / 600,110, filed November 17, 2023, which is hereby incorporated by reference in its entirety as though fully set forth herein.

[0042] The devices, systems, and methods described herein can include one or more features described in U.S. Provisional App. No. 62 / 750,983, filed 26 October 2018, U.S. Provisional App. No. 62 / 751,268, filed 26 October 2018, U.S. Provisional App. No. 62 / 751,289, filed 26 October 2018, U.S. Provisional App. No. 62 / 750,996, filed 26 October 2018, U.S. Provisional App. No. 62 / 751,167, filed 26 October 2018, and U.S. Provisional App. No. 62 / 751,185, filed 26 October 2018, each of which is hereby incorporated by reference in its entirety as though fully set forth herein.

[0043] The devices, systems, and methods described herein can also include one or more features described in U.S. Provisional App. No. 62 / 642,847, filed March 14, 2018, U.S. Provisional App. No. 62 / 712,009, filed July 30, 2018, U.S. Provisional App. No. 62 / 711,927, filed July 30, 2018, and U.S. Provisional App. No. 62 / 643,366, filed March 15, 2018, each of which is hereby incorporated by reference in its entirety as though fully set forth herein.

[0044] These descriptions are provided for exemplary purposes only, and should not be considered to limit the scope of the plaque and carina shift prediction system. Certain features may be added, removed, or modified without departing from the spirit of the claimed subject matter.

[0045] For the purposes of promoting an understanding of the principles of the present disclosure, reference will now be made to the aspects illustrated in the drawings, and specific language will be used to describe the same. It is nevertheless understood that no limitation to the scope of the disclosure is intended. Any alterations and further modifications to the described devices, systems, and methods, and any further application of the principles of the present disclosure are fully contemplated and included within the present disclosure as would normally occur to one skilled in the art to which the disclosure relates. In particular, it is fully contemplated that the features, components, and / or steps described with respect to one aspect may be combined with the features, components, and / or steps described with respect to other aspects of the present disclosure. For the sake of brevity, however, the numerous iterations of these combinations will not be described separately.

[0046] Figure 1 is a diagrammatic schematic view of an intraluminal imaging system, according to aspects of the present disclosure. The intraluminal imaging system 100 can be an intravascular ultrasound (IVUS) imaging system in some aspects. The intraluminal imaging system 100 may include an intraluminal device 102, a patient interface module (PIM) 104, aconsole or processing system 106, a monitor 108, and an external imaging system 132 which may include angiography, ultrasound, X-ray, computed tomography (CT), magnetic resonance imaging (MRI), or other imaging technologies, equipment, and methods. The intraluminal device 102 is sized and shaped, and / or otherwise structurally arranged to be positioned within a body lumen of a patient. For example, the intraluminal device 102 can be a catheter, guide wire, guide catheter, pressure wire, and / or flow wire in various aspects. In some circumstances, the system 100 may include additional elements and / or may be implemented without one or more of the elements illustrated in Figure 1. For example, the system 100 may omit the external imaging system 132.

[0047] The intraluminal imaging system 100 (or intravascular imaging system) can be any type of imaging system suitable for use in the lumens or vasculature of a patient. In some aspects, the intraluminal imaging system 100 is an intravascular ultrasound (IVUS) imaging system. In other aspects, the intraluminal imaging system 100 may include systems configured for forward looking intravascular ultrasound (FL-IVUS) imaging, intravascular photoacoustic (IVPA) imaging, intracardiac echocardiography (ICE), transesophageal echocardiography (TEE), and / or other suitable imaging modalities.

[0048] It is understood that the system 100 and / or device 102 can be configured to obtain any suitable intraluminal imaging data. In some aspects, the device 102 may include an imaging component of any suitable imaging modality, such as optical imaging, optical coherence tomography (OCT), etc. In some aspects, the device 102 may include any suitable non-imaging component, including a pressure sensor, a flow sensor, a temperature sensor, an optical fiber, a reflector, a mirror, a prism, an ablation element, a radio frequency (RF) electrode, a conductor, or combinations thereof. Generally, the device 102 can include an imaging element to obtain intraluminal imaging data associated with the lumen 120. The device 102 may be sized and shaped (and / or configured) for insertion into a vessel or lumen 120 of the patient.

[0049] The system 100 may be deployed in a catheterization laboratory having a control room. The processing system 106 may be located in the control room. Optionally, the processing system 106 may be located elsewhere, such as in the catheterization laboratory itself. The catheterization laboratory may include a sterile field while its associated control room may or may not be sterile depending on the procedure to be performed and / or on the health care facility. The catheterization laboratory and control room may be used to perform any number ofmedical imaging procedures such as angiography, fluoroscopy, CT, IVUS, virtual histology (VH), forward looking IVUS (FL-IVUS), intraluminal photoacoustic (IVPA) imaging, a fractional flow reserve (FFR) determination, a coronary flow reserve (CFR) determination, optical coherence tomography (OCT), computed tomography, intracardiac echocardiography (ICE), forward-looking ICE (FLICE), intraluminal palpography, transesophageal ultrasound, fluoroscopy, and other medical imaging modalities, or combinations thereof. In some aspects, device 102 may be controlled from a remote location such as the control room, such than an operator is not required to be in close proximity to the patient.

[0050] The intraluminal device 102, PIM 104, monitor 108, and external imaging system 132 may be communicatively coupled directly or indirectly to the processing system 106. These elements may be communicatively coupled to the medical processing system 106 via a wired connection such as a standard copper link or a fiber optic link and / or via wireless connections using IEEE 802.11 Wi-Fi standards, Ultra Wide-Band (UWB) standards, wireless FireWire, wireless USB, or another high-speed wireless networking standard. The processing system 106 may be communicatively coupled to one or more data networks, e.g., a TCP / IP-based local area network (LAN). In other aspects, different protocols may be utilized such as Synchronous Optical Networking (SONET). In some cases, the processing system 106 may be communicatively coupled to a wide area network (WAN). The processing system 106 may utilize network connectivity to access various resources. For example, the processing system 106 may communicate with a Digital Imaging and Communications in Medicine (DICOM) system, a Picture Archiving and Communication System (PACS), and / or a Hospital Information System (HIS) via a network connection.

[0051] At a high level, an ultrasound imaging intraluminal device 102 emits ultrasonic energy from a transducer array 124 included in scanner assembly 110 mounted near a distal end of the intraluminal device 102. The ultrasonic energy is reflected by tissue structures in the medium (such as a lumen 120) surrounding the scanner assembly 110, and the ultrasound echo signals are received by the transducer array 124. The scanner assembly 110 generates electrical signal(s) representative of the ultrasound echoes. The scanner assembly 110 can include one or more single ultrasound transducers and / or a transducer array 124 in any suitable configuration, such as a planar array, a curved array, a circumferential array, an annular array, etc. For example, the scanner assembly 110 can be a one-dimensional array or a two-dimensional array insome instances. In some instances, the scanner assembly 110 can be a rotational ultrasound device. The active area of the scanner assembly 110 can include one or more transducer materials and / or one or more segments of ultrasound elements (e.g., one or more rows, one or more columns, and / or one or more orientations) that can be uniformly or independently controlled and activated. The active area of the scanner assembly 110 can be patterned or structured in various basic or complex geometries. The scanner assembly 110 can be disposed in a side-looking orientation (e.g., ultrasonic energy emitted perpendicular and / or orthogonal to the longitudinal axis of the intraluminal device 102) and / or a forward-looking looking orientation (e.g., ultrasonic energy emitted parallel to and / or along the longitudinal axis). In some instances, the scanner assembly 110 is structurally arranged to emit and / or receive ultrasonic energy at an oblique angle relative to the longitudinal axis, in a proximal or distal direction. In some aspects, ultrasonic energy emission can be electronically steered by selective triggering of one or more transducer elements of the scanner assembly 110.

[0052] The ultrasound transducer(s) of the scanner assembly 110 can be a piezoelectric micromachined ultrasound transducer (PMUT), capacitive micromachined ultrasonic transducer (CMUT), single crystal, lead zirconate titanate (PZT), PZT composite, other suitable transducer type, and / or combinations thereof. In an aspect the ultrasound transducer array 124 can include any suitable number of individual transducer elements or acoustic elements between 1 acoustic element and 1000 acoustic elements, including values such as 2 acoustic elements, 4 acoustic elements, 36 acoustic elements, 64 acoustic elements, 128 acoustic elements, 500 acoustic elements, 812 acoustic elements, and / or other values both larger and smaller.

[0053] The PIM 104 transfers the received echo signals to the processing system 106 where the ultrasound image (including the flow information) is reconstructed and displayed on the monitor 108. The console or processing system 106 can include a processor and a memory. The processing system 106 may be operable to facilitate the features of the intraluminal imaging system 100 described herein. For example, the processor can execute computer readable instructions stored on the non-transitory tangible computer readable medium.

[0054] The PIM 104 facilitates communication of signals between the processing system 106 and the scanner assembly 110 included in the intraluminal device 102. This communication may include providing commands to integrated circuit controller chip(s) within the intraluminal device 102, selecting particular element(s) on the transducer array 124 to be used for transmitand receive, providing the transmit trigger signals to the integrated circuit controller chip(s) to activate the transmitter circuitry to generate an electrical pulse to excite the selected transducer array element(s), and / or accepting amplified echo signals received from the selected transducer array element(s) via amplifiers included on the integrated circuit controller chip(s). In some aspects, the PIM 104 performs preliminary processing of the echo data prior to relaying the data to the processing system 106. In examples of such aspects, the PIM 104 performs amplification, filtering, and / or aggregating of the data. In an aspect, the PIM 104 also supplies high- and low- voltage DC power to support operation of the intraluminal device 102 including circuitry within the scanner assembly 110.

[0055] The processing system 106 receives echo data from the scanner assembly 110 by way of the PIM 104 and processes the data to reconstruct an image of the tissue structures in the medium surrounding the scanner assembly 110. Generally, the device 102 can be utilized within any suitable anatomy and / or body lumen of the patient. The processing system 106 outputs image data such that an image of the vessel or lumen 120, such as a cross-sectional IVUS image of the lumen 120, is displayed on the monitor 108. Lumen 120 may represent fluid filled or fluid-surrounded structures, both natural and man-made. Lumen 120 may be within a body of a patient. Lumen 120 may be a blood vessel, such as an artery or a vein of a patient’s vascular system, including cardiac vasculature, peripheral vasculature, neural vasculature, renal vasculature, and / or or any other suitable lumen inside the body. For example, the device 102 may be used to examine any number of anatomical locations and tissue types, including without limitation, organs including the liver, heart, kidneys, gall bladder, pancreas, lungs; ducts; intestines; nervous system structures including the brain, dural sac, spinal cord and peripheral nerves; the urinary tract; as well as valves within the blood, chambers or other parts of the heart, and / or other systems of the body. In addition to natural structures, the device 102 may be used to examine man-made structures such as, but without limitation, heart valves, stents, shunts, filters and other devices.

[0056] The controller or processing system 106 may include a processing circuit having one or more processors in communication with memory and / or other suitable tangible computer readable storage media. The controller or processing system 106 may be configured to carry out one or more aspects of the present disclosure. In some aspects, the processing system 106 and the monitor 108 are separate components. In other aspects, the processing system 106 and themonitor 108 are integrated in a single component. For example, the system 100 can include a touch screen device, including a housing having a touch screen display and a processor. The system 100 can include any suitable input device, such as a touch sensitive pad or touch screen display, keyboard / mouse, joystick, button, etc., for a user to select options shown on the monitor 108. The processing system 106, the monitor 108, the input device, and / or combinations thereof can be referenced as a controller of the system 100. The controller can be in communication with the device 102, the PIM 104, the processing system 106, the monitor 108, the input device, and / or other components of the system 100.

[0057] In some aspects, the intraluminal device 102 includes some features similar to traditional solid-state IVUS catheters, such those disclosed in U.S. Patent No. 7,846,101, hereby incorporated by reference in its entirety. For example, the intraluminal device 102 may include the scanner assembly 110 near a distal end of the intraluminal device 102 and a transmission line bundle 112 extending along the longitudinal body of the intraluminal device 102. The cable or transmission line bundle 112 can include a plurality of conductors, including one, two, three, four, five, six, seven, or more conductors.

[0058] The transmission line bundle 112 terminates in a PIM connector 114 at a proximal end of the intraluminal device 102. The PIM connector 114 electrically couples the transmission line bundle 112 to the PIM 104 and physically couples the intraluminal device 102 to the PIM 104. In an aspect, the intraluminal device 102 further includes a guidewire exit port 116. Accordingly, in some instances the intraluminal device 102 is a rapid-exchange catheter. The guidewire exit port 116 allows a guidewire 118 to be inserted towards the distal end in order to direct the intraluminal device 102 through the lumen 120.

[0059] The monitor 108 may be a display device such as a computer monitor or other type of screen. The monitor 108 may be used to display selectable prompts, instructions, and visualizations of imaging data to a user. In some aspects, the monitor 108 may be used to provide a procedure-specific workflow to a user to complete an intraluminal imaging procedure. This workflow may include performing a pre-stent plan to determine the state of a lumen and potential for a stent, as well as a post-stent inspection to determine the status of a stent that has been positioned in a lumen.

[0060] The external imaging system 132 can be configured to obtain x-ray, radiographic, angiographic / venographic (e.g., with contrast), and / or fluoroscopic (e.g., without contrast)images of the body of a patient (including the vessel 120). External imaging system 132 may also be configured to obtain computed tomography images of the body of the patient (including the vessel 120). The external imaging system 132 may include an external ultrasound probe configured to obtain ultrasound images of the body of the patient (including the vessel 120) while positioned outside the body. In some aspects, the system 100 includes other imaging modality systems (e.g., MRI) to obtain images of the body of the patient (including the vessel 120). The processing system 106 can utilize the images of the body of the patient in conjunction with the intraluminal images obtained by the intraluminal device 102.

[0061] Figure 2 is a schematic diagram of a processor circuit 250, according to aspects of the present disclosure. The processor circuit 250 may be implemented in the intraluminal imaging system 100, or other devices or workstations (e.g., third-party workstations, network routers, etc.), or on a cloud processor or other remote processing unit, as necessary to implement the method. As shown, the processor circuit 250 may include a processor 260, a memory 264, and a communication module 268. These elements may be in direct or indirect communication with each other, for example via one or more buses.

[0062] The processor 260 may include a central processing unit (CPU), a digital signal processor (DSP), an ASIC, a controller, or any combination of general-purpose computing devices, reduced instruction set computing (RISC) devices, application-specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other related logic devices, including mechanical and quantum computers. The processor 260 may also comprise another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein. The processor 260 may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.

[0063] The memory 264 may include a cache memory (e.g., a cache memory of the processor 260), random access memory (RAM), magnetoresistive RAM (MRAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), flash memory, solid state memory device, hard disk drives, other forms of volatile and non-volatile memory, or a combination of different types of memory. In an aspect, the memory 264 includesa non-transitory computer-readable medium. The memory 264 may store instructions 266. The instructions 266 may include instructions that, when executed by the processor 260, cause the processor 260 to perform the operations described herein. Instructions 266 may also be referred to as code. The terms “instructions” and “code” should be interpreted broadly to include any type of computer-readable statement(s). For example, the terms “instructions” and “code” may refer to one or more programs, routines, sub-routines, functions, procedures, etc. “Instructions” and “code” may include a single computer-readable statement or many computer-readable statements.

[0064] The communication module 268 can include any electronic circuitry and / or logic circuitry to facilitate direct or indirect communication of data between the processor circuit 250, and other processors or devices. In that regard, the communication module 268 can be an input / output (I / O) device. In some instances, the communication module 268 facilitates direct or indirect communication between various elements of the processor circuit 250 and / or the intraluminal imaging system 100. The communication module 268 may communicate within the processor circuit 250 through numerous methods or protocols. Serial communication protocols may include but are not limited to United States Serial Protocol Interface (US SPI), Inter- Integrated Circuit (I2C), Recommended Standard 232 (RS-232), RS-485, Controller Area Network (CAN), Ethernet, Aeronautical Radio, Incorporated 429 (ARINC 429), MODBUS, Military Standard 1553 (MIL-STD-1553), or any other suitable method or protocol. Parallel protocols include but are not limited to Industry Standard Architecture (ISA), Advanced Technology Attachment (ATA), Small Computer System Interface (SCSI), Peripheral Component Interconnect (PCI), Institute of Electrical and Electronics Engineers 488 (IEEE-488), IEEE- 1284, and other suitable protocols. Where appropriate, serial and parallel communications may be bridged by a Universal Asynchronous Receiver Transmitter (UART), Universal Synchronous Receiver Transmitter (US ART), or other appropriate subsystem.

[0065] External communication (including but not limited to software updates, firmware updates, preset sharing between the processor and central server, or readings from the annular ultrasound imaging array) may be accomplished using any suitable wireless or wired communication technology, such as a cable interface such as a universal serial bus (USB), micro USB, Lightning, or FireWire interface, Bluetooth, Wi-Fi, ZigBee, Li-Fi, or cellular data connections such as 2G / GSM (global system for mobiles) , 3G / UMTS (universal mobiletelecommunications system), 4G, long term evolution (LTE), WiMax, or 5G. For example, a Bluetooth Low Energy (BLE) radio can be used to establish connectivity with a cloud service, for transmission of data, and for receipt of software patches. The controller may be configured to communicate with a remote server, or a local device such as a laptop, tablet, or handheld device, or may include a display capable of showing status variables and other information. Information may also be transferred on physical media such as a USB flash drive or memory stick.

[0066] It will also be understood that one or more of the steps of the methods described above can be performed by one or more components of an ultrasound imaging system, such as the processing system, a multiplexer, a beamformer, a signal processing unit, an image processing unit, or any other suitable component of the system. For example, activating the scan sequences may be carried out by a processor in communication with a multiplexer configured to select or activate one or more elements of an ultrasound transducer array. In some aspects, generating the ultrasound images may include beamforming incoming signals from the ultrasound imaging device and processing the beamformed signals by an image processor. The processing components of the system can be integrated within the ultrasound imaging device, contained within an external console, or may be a separate component.

[0067] Figure 3A illustrates a blood vessel 300 with a plaque, according to aspects of the present disclosure. The blood vessel 300 includes a main vessel 302, side branch 304, and carina 306. As depicted in Fig. 3A, a plaque 308 may be disposed in the main vessel 302 of the blood vessel 300, and a plaque 310 may be disposed in the side branch 304 of the blood vessel 300. Plaque 310 may reduce the cross-sectional area of the lumen of the blood vessel’s main vessel 302, resulting in reduced blood flow. Similarly, plaque 310 may reduce the cross-sectional area of the lumen of the blood vessel’s side branch 304. The angle 312 defined by the wall of the side branch 304 and wall of the main vessel 302 may impact the likelihood of the carina 306 deflecting during a stenting procedure, reducing blood flow through the side branch 304. The size, shape, and number of plaques depicted in Fig. 3A are merely illustrative; the plaque burden in a blood vessel may be lower or higher, the plaque may be more or less focal, and more or fewer distinct plaques may be presents in a blood vessel. For example, the side branch 304 may not contain plaque 310. In addition, plaques may be of varying softness and composition. Forexample, some plaques are soft (i.e., fatty) while others are more fibrous or comprised by calcium.

[0068] Figure 3B illustrates the blood vessel 300 with a plaque, as depicted in Fig. 3A, with a stent 320 expanded inside a main vessel 302 to restore flow, according to aspects of the present disclosure. The blood vessel 300 includes a main vessel 302, side branch 304, and carina 306. As depicted in Fig. 3B, the stent 320 has been deployed to increase the lumen area of the main vessel 302. In deploying the stent 320, the plaque 308 formerly impeding the flow, as shown in Fig. 3 A, has been shifted. However, deployment of the stent 320 has also deflected the carina 306. By deflecting the carina 306, blood flow through the side branch 304 may be reduced because the carina, at least partially, blocks the side branch 304. Blood flow may be further reduced where the deflection of the carina 306 by a stent 320 occurs at a side branch 304 with a plaque 310 located near the branch point (e.g., confluence, bifurcation, etc.) of the side branch 304 from the main vessel 302.

[0069] In planning a stenting procedure, it important that a physician be aware of the possibility of a carina shift and the possibility of negative impacts to blood flow in the side branch. It is one of the benefits of this disclosure that a warning / flag may be automatically generated while conducting intravascular imaging to warn a physician of the possibility of a carina shift and / or impact on blood flow. In some embodiments, providing a warning / flag may be based on the angle a side branch makes with the main vessel.

[0070] Figure 4A illustrates a blood vessel 400 with plaques of varying softness and / or composition, according to aspects of the present disclosure. The blood vessel 400 includes a main vessel 402 and side branch 404. As depicted in Fig. 4A, the main vessel 402 includes a first plaque 406 and at least part of a second plaque 408, and the side branch 404 includes, at least part of, the second plaque 408 and a third plaque 410. First plaque 406 and second plaque 408 may reduce the cross-sectional area of the lumen of the blood vessel’s main vessel 402, resulting in reduced blood flow. Similarly, second plaque 408 and third plaque 410 may reduce the cross- sectional area of the lumen of the blood vessel’s side branch 404. The second plaque 408 may be softer than the first plaque 406 and third plaque 410. Generally, softer plaques flow more easily, i.e., when compressed, softer plaques will flow in directions perpendicular to the compression direction instead of deflecting a vessel wall in the direction of compression. Thus, when a stent is placed, softer plaques may be displaced by the stent in unintended ways. Thesize, shape, and number of plaques depicted in Fig. 4A are merely illustrative; the plaque burden in a blood vessel may be lower or higher, the plaque may be more or less focal, and more or fewer distinct plaques may be presents in a blood vessel. For example, the side branch 304 may not contain plaque 310. In addition, plaques may be of varying softness and composition. For example, some plaques are soft (i.e., fatty) while others are more fibrous or comprised by calcium.

[0071] Figure 4B illustrates the blood vessel 400 with plaques of varying softness and / or composition, as depicted in Fig. 4A, with a stent 420 expanded inside a main vessel 402 to restore flow, according to aspects of the present disclosure. As depicted in Fig. 4B, the stent 420 has been deployed to increase the lumen area of the main vessel 402. In deploying the stent 420, the plaque 406 formerly impeding the flow, as shown in Fig. 4A, has been successfully shifted, increasing the cross-sectional area of the lumen the main vessel 402. However, deployment of the stent 420 has also caused the soft second plaque 408 to flow into the side branch 404. Because of the flow of second plaque 408 into the side branch 404, blood flow through the side branch 404 may be reduced because second plaque 408 at least partially blocks the side branch 404. Blood flow may be significantly reduced where the flow of second plaque 408 by the stent 420 occurs in a side branch 404 with a third plaque 410 located near the branch point (e.g., confluence, bifurcation, etc.) of the side branch 404 from the main vessel 302.

[0072] In planning a stenting procedure, it important that a physician be aware of the possibility of plaque flow and the possibility of negative impacts to blood flow in the side branch. It is one of the benefits of this disclosure that a warning / flag may be automatically generated while conducting intravascular imaging to warn a physician of the possibility of a plaque flow and / or impact on blood flow. In some embodiments, providing a warning / flag may be based on the brightness of plaques appearance in an ultrasound-based image.

[0073] Figures 5A-5D depicts various severities of lesion in the form of plaque. Figs. 5A and 5B depict a focal lesion, while Figs. 5C and 5D depict a diffuse lesion. As a general matter, after stenting, focal lesions present a greater risk of plaque flow / shift compared to diffuse lesions, other factors being equal.

[0074] Figure 5A illustrates a blood vessel 500 with a focal lesion 502, according to aspects of the present disclosure. The focal lesion 502 of blood vessel 500 comprises a significant quantity of plaque that is confined to a relatively short length of the blood vessel 500. A focallesion 502 may be identified by a large slope for the thickness / area of the plaque along the vessel. For example, a high plaque burden may be indicative of a focal lesion. Plaque burden of a given location or image frame may be defined as:(Vessel Area - Lumen Area) / Vessel Area (EQN. 1)

[0075] In some instances, focal lesions are more likely to shift / flow when a stent is placed. The amount and / or likelihood of shift or slow of a focal lesion 502 may be increased when the plaque is relatively soft.

[0076] Figure 5B illustrates a blood vessel 500 with a focal lesion 502 with a stent 505 expanded in the blood vessel 500, according to aspects of the present disclosure. After placement of the stent 505, focal lesion 502 has shifted / flowed beyond the edges of the stent 505. As a result, the shift of plaque, the stent 505 may be less effective in restoring blood flow in the vessel 500.

[0077] In planning a stenting procedure, it important that a physician be aware of the possibility of plaque flow. It is one of the benefits of this disclosure that a warning / flag may be automatically generated while conducting intravascular imaging to warn a physician of the possibility of a plaque flow and / or impact on blood flow. In some embodiments, providing a warning / flag may be based on how focal or diffuse a lesion is, determined using an ultrasoundbased image.

[0078] Figure 5C illustrates a blood vessel 510 with a diffuse lesion, according to aspects of the present disclosure. The diffuse lesion of blood vessel 500 comprises a quantity of plaque 502 that is spread over a significant length of the blood vessel 500. For example, a low plaque burden may be indicative of a diffuse lesion. Plaque burden may be defined by EQN. 1.

[0079] Figure 5D illustrates a blood vessel 510 with a diffuse lesion 512 with a stent 515 expanded in the blood vessel 510, according to aspects of the present disclosure. After placement of the stent 515, diffuse lesion 502 is confined to the radial exterior of the stent. In other words, no flow of plaque beyond the end points are the stent 515 has been observed. As a result, the stent 515 is effective in restoring blood flow in the vessel 510. In some instances, diffuse lesions present less concern for plaque shifting / flow.

[0080] Figure 6 is an example display 600 of blood vessel images with a warning for plaque or carina shift, according to aspects of the present disclosure. Example display 600 may be provided to a physician during the planning for a stent placement procedure. Example display 600 may include a tomographic or radial cross-sectional image 610, an image-based longitudinal view 620, and / or an angiogram 630. A warning flag 635 may be included in any of a tomographic of radial cross-sectional image 610, an image-based longitudinal view 620, and / or an angiogram 630. A warning flag 635 may include an explanation such “plaque shift possible” or “carina shift possible.” The explanations generated with the warning flag 635 may also account for the likelihood of a plaque or carina shift. For example, an explanation may indicate that a shift is “possible,” “more likely than not,” or “very likely.” In some embodiments, numerical probabilities may be included in the explanations accompanying a warning flag 635.

[0081] The image-based longitudinal view 620 may include a bounding box 640, frame marker 645, and / or warning flag 635. A bounding box 640 may be placed on the image-based longitudinal view 620 and / or angiogram 630. The bounding box 640 indicates the section of the blood vessel where plaque and / or carina shift is likely to occur if a stent were to be placed there. As an example, a frame 645 contained within the bounding box 640 may be selected, and the corresponding tomographic or radial-cross sectional image 610. The display may indicate which frame has been selected, e.g., “Frame 330.” Furthermore, information 650 about the vessel area, lumen area, and / or plaque burden may be displayed.

[0082] The angiogram 630 may include bounding box 640, frame marker 645, and / or warning flag 635. Display 600 may or may not include angiogram 630. To include a bounding box 640, frame marker 645, and warning flag on angiogram 630 requires co-registration of the angiogram 630 with the intravascular image 610 and / or longitudinal view 620. The angiogram may be processed from x-ray imaging data generated using the external imaging system 132. Aspects of co-registration are described for example in U.S. Patent No. 7,930,014, titled “Vascular image co-registration”, and U.S. Publication No. 2012 / 0004537, titled “Co-use of endoluminal data and extraluminal imaging”, each of which is incorporated by reference as though fully set forth herein.

[0083] A tomographic or radial cross-sectional image 610 may be a frame from a sequence of frames generated by an intravascular imaging pullback using the intraluminal imaging system 100. A tomographic or radial cross-sectional image 610 may be associated with frame marker645 included in the longitudinal view 620 and / or angiogram 630. A frame number identifying the frame 610 in the in the sequence of frames from a pullback may also be provided, e.g., “Frame 330” as depicted in Fig. 6. A tomographic or radial cross-sectional image 610 may depict plaque 652, lumen contour 654, and vessel contour 656. From these and other vessel features may be computed vessel measurements 650, such the vessel area, lumen area, and plaque burden. The plaque burden may be computed according to EQN. 1.

[0084] Figure 7 is a diagrammatic schematic view of a plaque and carina shift prediction system 700, according to aspects of the present disclosure. System 700 may determine a likelihood for carina shift and / or plaque shift and generate a display on which a physician may review the areas of a vessel where a shift is likely. The plaque and carina shift prediction system 700 may include intravascular imaging 710, border identification 720, and carina shift and / or plaque shift prediction 740.

[0085] Intravascular imaging 710 may include use of an intraluminal imaging system 100 to gather intravascular imaging data to generate intravascular images 712 (e.g., cross-sectional radial and / or tomographic views). Intraluminal imaging system 100 may process the intravascular images 712 into longitudinal views 714 (e.g., image-based longitudinal view as depicted in the display of Fig. 6 and / or a graphical longitudinal view). Longitudinal locations along an intravascular cross-sectional longitudinal view can be respectively correspond to individual intravascular images (radial and / or tomographic cross-sectional views). Longitudinal views are described in, for example, U.S. Publication No. 2020 / 0129158, U.S. Publication No. 2022 / 0079563, U.S. Patent No. 9,367,965, and U.S. Patent No. 10,070,827, each of which is incorporated by reference herein in its entirety. Intravascular images 712 and longitudinal views 714 may be output to a border identification module 720 and / or screen display 750.

[0086] Border identification module 720 process the images and view from intravascular imaging 710 to identify various features in the image. Lumen borders 722, such as those associated with the main vessel and / or side branch of a blood vessel, may be identified using techniques known in the art. Alternatively or in addition to lumen borders 722, vessel borders may be identified using techniques known in the art. The borders of the plaque 726 present in the intravascular images and longitudinal views may be identified separately from the lumen borders 722 and vessel borders 724. Alternatively, plaque 726 may be identified using the shape / size of the space between the lumen border and vessel border. Aspects of detecting a lumen borderand / or a vessel border and / or calculating lumen / vessel dimensions based on the detected borders are described in U.S. Publication No. 2007 / 0201736, U.S. Patent No. 11,272,845, U.S. Patent No. 7,463,759, U.S. Patent No. 9,295,447, U.S. Patent No. 11,744,527, U.S. Publication No. 2020 / 0029932, U.S. Publication No. 2019 / 0282211, each of which is hereby incorporated by reference as though fully set forth herein.

[0087] After borders have been identified using border identification module 720, various vessel measurements may be determined. Vessel measurements may include main vessel and / or side branch lumen volume / area 732, side branch angle 734, plaque type and / or plaque softness 736, and / or plaque shape 738. Main vessel and / or side branch lumen volume / area 732 may be determined in various ways as described herein. Like the vessel measurements 650 in display 600 of Fig. 6, the main vessel and / or side branch lumen volume / area 732 may be included in a display. Side branch angle 734, e.g., as described further with respect to Figs. 8A-8B, may be computed from distances between imaging planes and distances between the wall of the main vessel and side branch of a blood vessel. In some instances, the distance between the wall of the main vessel and side branch is defined to be the closest distance between the wall of the main vessel and side branch of a blood vessel as measured within an imaging plane. Plaque type and / or softness 736 may be determined from the contrast in an intravascular image 712 because the opacity of plaque to, for example, ultrasound depends on its softness. In some embodiments, virtual histology may be used to further identify plaque composition. For example, plaque identified in intravascular images may fall into four different composition types (from hardest to softest): necrotic core, dense calcium, fibrous, and fibro-fatty. These composition types may be indicated using color in intravascular images. Aspects of tissue characterization and / or virtual histology are described in U.S. Patent No. 8,449,465, U.S. Patent No. 9,615,878, U.S. Patent No. 9,592,027, each of which is hereby incorporated by reference as though fully set forth herein. Plaque shape 738 may include determining whether a plaque is focal, diffuse, or irregular. Examples of various plaque shapes are depicted in and described with respect to Figs. 3-5.

[0088] Carina shift and / or plaque shift prediction module 740 may receive main vessel and / or side branch lumen volume / area 732, side branch angle 734, plaque type and / or plaque softness 736, and / or plaque shape 738. Carina shift and / or plaque shift prediction module 740 may generate a probability / likelihood of plaque and / or carina shifting, e.g., shifting as described in Figs. 3-5. Carina shift and / or plaque shift prediction module 740 may identify a region of ablood vessel where a carina or plaque shift may occur. The probability of a plaque or carina shift and / or the region where the shift may occur may be used to determine if an indicator should be presented to a user, e.g., a physician. Carina shift and / or plaque shift prediction module 740 is further described with respect to and depicted in Figs. 10-11 and 16-18, e.g., modules 1000, 1100, 1640, 1720, 1830.

[0089] Screen display 750 may provide various images and carina and plaque shift prediction to a user such as a physician. Screen display 750 may include intravascular images 712, longitudinal view 714, and indicator for carina or plaque shift prediction 752, e.g., as generated by carina shift and / or plaque shift prediction module 740. Indicator 752 may include a region of the blood vessel where a plaque or carina shift is predicted to occur. Screen display may be similar to display 600 depicted in Fig. 6.

[0090] Figure 8A illustrates a blood vessel 800 with a first side branch angle 810, according to aspects of the present disclosure. The blood vessel 800 includes a major branch 802, side branch 804, and carina 806. A first side branch angle is defined by the angle between the side branch wall 812 and major branch wall 814. For sake of explanation, first imaging plane 820, second imaging plane 822, and third imaging plane 824 are spaced equidistantly along the length of the main vessel. In some embodiments, imaging planes may correspond to intravascular imaging frames as produced from intravascular imaging data collected by intraluminal imaging system 100. Depending on the first side branch angle 810, the distance 816 between the main vessel and side branch may be larger or smaller for second imaging plane 822. In practice, the distance 816 may be measured from an intravascular image. Using the distance 816 and the distance 830 between imaging planes, the angle may be determined by a simple trigonometric calculation. The third imaging plane 824 has an intersection 818 with the side branch wall 812. Larger angles will result in intersections of the side branch wall 812 with an imaging plan being further from the main vessel wall 814. The imaging plans and distances of Fig. 8A are merely illustrative. Imaging data from a real intravascular pullback may require the use non- equi distantly spaced imaging planes and / or averaging of distances to account for noise in the data.

[0091] In some instances, more acute angles increase the likelihood that the carina 806 will be deflected by stent placing, reducing the blood flow to the side branch 804.

[0092] Figure 8B illustrates a blood vessel 850 with a second side branch angle 860, according to aspects of the present disclosure. The blood vessel 850 includes a major branch 852, side branch 854, and carina 856. A first side branch angle is defined by the angle between the side branch wall 862 and major branch wall 864. For sake of explanation, first imaging plane 870, second imaging plane 872, and third imaging plane 874 are spaced equidistantly along the length of the main vessel. The imaging planes 870, 872, 874 of Fig. 8B are equivalent to the imaging planes 820, 822, 824 of Fig. 8A. In some embodiments, imaging planes may correspond to intravascular imaging frames as produced from intravascular imaging data collected by intraluminal imaging system 100. Depending on the second side branch angle 860, the distance 866 between the main vessel and side branch may be larger or smaller for second imaging plane 872. In practice, the distance 866 may be measured from an intravascular image. Using the distance 866 and the distance 830 between imaging planes, the angle may be determined by a simple trigonometric calculation.

[0093] Compared to Fig. 8 A, the second side branch angle 860 is larger than the first side branch angle 810. The larger angle can be observed from the lack of equivalent intersection between side branch wall 862 and third imaging plane 874 in Fig. 8B to intersection 818 in Fig. 8A. Larger angles will result in intersections of the side branch wall 862 with an imaging plan being further from the main vessel wall 864. The imaging plans and distances of Fig. 8B are merely illustrative. Imaging data from a real intravascular pullback may require the use non- equi distantly spaced imaging planes and / or averaging of distances to account for noise in the data. In some instances, less acute angles (angles closer to 90 degrees) decrease the likelihood that the carina 806 will be deflected by stent placing, reducing the blood flow to the side branch 804.

[0094] Figures 9A-9C depicts cross-sectional images of a blood vessel along two of the equivalent imaging planes as described in Figs. 8A and 8B. A mask 902 is included to reflect how the intravascular image frames comprising the cross-section images are typically presented. The mask 902 covers the portion of the image where the imaging catheter would be located.

[0095] Figure 9A is an illustration of the cross-sectional image 900 of a blood vessel at the first equivalent imaging planes 820, 870, according to aspects of the present disclosure. Cross- sectional image 900 depicts a cross-section of the main vessel 802 and side branch 804. As shown in Figs. 8A and 8B, the first imaging plane 820, 870 crosses near the intersection of theside branch 804 and main vessel 802. Thus, there is almost no separation between the lumens of the main vessel 802 and side branch 804.

[0096] Figure 9B is an illustration of the cross-section 920 of a blood vessel at the second imaging plane 822 of Fig. 8A, according to aspects of the present disclosure. Cross-sectional image 920 depicts a cross-section of the main vessel 802 and side branch 804. As shown in Figs. 8B at the second imaging plane 820, the lumens of the main vessel 802 and side branch 804 are separated by the distance 816, which may be measured perpendicular from the main vessel 802 wall.

[0097] Figure 9C is an illustration of the cross-section 940 of a blood vessel at the second imaging plane 872 of Fig. 8B, according to aspects of the present disclosure. Cross-sectional image 940 depicts a cross-section of the main vessel 802 and side branch 804. As shown in Figs. 8B at the second imaging plane 870, the lumens of the main vessel 802 and side branch 804 are separated by the distance 866, which may be measured perpendicular from the main vessel 852 wall.

[0098] Figure 10 is a diagrammatic schematic view of carina shift and / or plaque shift prediction module 1000 including look-up tables and / or decision trees 1010, according to aspects of the present disclosure. Carina shift and / or plaque shift prediction module 1000 includes individual / combined look-up table(s) or decision tree(s) 1010, scoring and / or weights 1012, threshold for indicator or no indicator 1016. In some embodiments, carina shift and / or plaque shift prediction module 1000 may receive as input main vessel and / or side branch lumen volume / area 732, side branch angle 734, plaque type and / or plaque softness 736, and / or plaque shape 738, as described in Fig. 7.

[0099] Individual / combined look-up table(s) or decision tree(s) 1010 include one or more look-up tables and / or decision trees. Look-up tables may comprise tables which may be functions of the vessel measurements, e.g., main vessel and / or side branch lumen volume / area 732, side branch angle 734, plaque type and / or plaque softness 736, and / or plaque shape 738. A value of one or more vessel measurements selects an entry in a look-up. The entries in a look-up table may be indicative of a plaque shift or carina shift. Examples of look-up tables are provided in Figs. 12-13. Decision trees may include one or more statements of propositional logic that result in a numerical value or decision. For example, a decision tree might include a proposition “Is the side branch angle less than 45 degrees?” If the answer is yes, the decision tree may returna value of “1,” whereas if the answer is no, then the decision tree may return a value of “0.” Propositions may be similarly formulated for each of the vessel measurements or a combination of the vessel measurements. In some embodiments, there may be multiple and / or or a sequence of propositional logic questions for each measurement separately or combined with another measurement.

[0100] The numerical values output from the individual / combined look-up table(s) or decision tree(s) 1010 may be combined using scoring and / or weights 1012. For example, the entries selected from multiple look-up tables may be summed together in a linear combination using weights. Weights may either be pre-programmed or received as a user input. For example, a physician may be more concerned about occlusions of side branches as a result of stent placement and thus may prefer the look-up table or decision tree value derived from the side branch angle be more heavily weighted. To effect this preference, a physician may be prompted to enter one or more weights for various vessel features and / or measurements, including the side branch angle. When scores are used, entries in the look-up table may correspond to scores which may be tallied to create a final score. The final score may be compared to a threshold score and / or mapped to a probability for the likelihood of a plaque or carina shift. In other words, after application of the scoring or weights, the results may be mapped to a likelihood of carina shift and / or plaque shift 1014. The likelihood of carina shift and / or plaque shift 1014 may determine whether an indicator 1018 is generated. For example, likelihoods greater than (or equal to) 50% may result in an indicator being generated, whereas likelihoods less than 50% may result in no indicator. These probability thresholds are only examples, other ranges may be used to determine whether not indicator 1018 is generated.

[0101] The results from applying the scoring and / or weights 1012 may be compared against a threshold for indicator or no indicator 1016. A threshold for the indicator may be predetermined (or pre-programmed) or provided by a user, e.g., a physician. An indicator 1018 may or may not be generated based on whether the threshold was met.

[0102] In some embodiments, the values selected from the individual or combined look-up table(s) or decision tree(s) 1010 may be compared directly to the threshold 1016 for indicator 1018. For example, a look-up table may include probabilities of a plaque or carina shift and a threshold 1016 may be in the form of a minimum probability of shift to generate an indicator 1018. Thus, if the probability from the look-up table exceeds the threshold, then an indicator willbe generated. An indicator 1018 may or may not be generated based on whether the threshold was met.

[0103] In some embodiments, the values selected from the individual or combined look-up table(s) or decision tree(s) 1010 may directly decide whether an indicator 1018 is generated. A decision tree may include a threshold for plaque shape that immediately results in an indicator. For example, if the plaque is a focal lesion, then the decision tree may be configured to generate an indicator.

[0104] Scoring and / or weighting 1012 may be optional.

[0105] Figure 11 is a diagrammatic schematic view of carina shift and / or plaque shift prediction module 1100 including user-defined thresholds 1112, 1114, 1116, 1118, according to aspects of the present disclosure. Carina shift and / or plaque shift prediction module 1100 includes user-defined thresholds 1112, 1114, 1116, 1118, scoring and / or weights 1120, threshold for indicator or no indicator 1124. In some embodiments, carina shift and / or plaque shift prediction module 1100 may receive as input vessel measurements, including main vessel and / or side branch lumen volume / area 732, side branch angle 734, plaque type and / or plaque softness 736, and / or plaque shape 738, as described in Fig. 7.

[0106] For each vessel measurement, a user-defined threshold may be provided. Threshold 1112 applies to main vessel and / or side branch lumen volume / area 732. Threshold 1114 applies to side branch angle 734. Threshold 1116 applies to plaque type / softness 736. Threshold 1118 applies to plaque shape 738. For example, a physician may be prompted to respond to “Please enter a lower bound for the lumen volume / area below which the lumen volume / area will be factored into scoring / weighting.” The physician’s choice for the volume / area of the lumen will be informed by the target region of the blood vessel and the expected lumen volume / area. e.g., how large should the lumen for a particular blood vessel. A physician may similarly be prompted to input threshold for the other vessel measurements, which the physician informed by knowledge of the patient’s unique anatomy and the general features of the anatomy of the target region. A physician may input thresholds and respond to system-provided prompts through a display or other user interface, e.g., monitor 108.

[0107] After applying thresholds 1112, 1114, 1116, 1118, scoring and / or weighting 1120 may be applied. In some embodiments, scoring 1120 may count the number of thresholds met by the vessel measurements. For example, if three vessel measurements met user-definedthresholds, then the score is 3. The score may be compared to a separate threshold for indicator 1124. For example, threshold 1124 may require a score of 2 or higher. Thus, when the score is 2 or higher (i.e., at least two vessel measurements have met user-defined thresholds) an indicator 1126 may be generated. In some embodiments, the score resulting from scoring 1120 may be mapped to a likelihood of carina shift and / or plaque shift 1122. For example, analysis of historical data may show that satisfying two out of the four thresholds results in a likelihood of 25%, while satisfying three out of the four thresholds results in a likelihood of 90%. In such an example, when three out four thresholds are satisfied an indicator should be generated.

[0108] In some embodiments weights 1120 may be applied to the vessel measurements filtered by user-defined thresholds. For example, for each threshold that is met by a vessel measurement, the weight associated with the measurements is included in a sum. Thus, if the weights for the four vessel measurements 732, 734, 736, 738 were 0.25, 0.35, 0.2, 0.2, respectively and the threshold was only met for side branch angle 734 and plaque type 736, then weights 0.35 and 0.2 would be selected. The sum of the weights may be compared against a threshold for indicator 1124. Using the previous example, the sum of weights would be 0.55. If the threshold for an indicator 1124 was set at 0.6, then no indicator would be generated. Alternatively, if the threshold for indicator was set at 0.5, then an indicator would be generated.

[0109] A threshold for the indicator 1124 may be pre-determined (or pre-programmed) or provided by a user, e.g., a physician. An indicator 1126 may or may not be generated based on whether the threshold was met. Scoring and / or weighting 1120 may be optional. Threshold for indicator 1124 may be optional.

[0110] Figure 12 is a look-up table 1200 based on side branch angle 734 and lumen volume / area 732, according to aspects of the present disclosure. Look-up table 1200 includes a plurality of entries, e.g., numerical values, each entry identified by a value of the main vessel and / or side branch lumen volume / area 732 and side branch angle 734. In some embodiments, look-up table 1200 is two dimensional. Look-up table 1200 includes a horizontal axis 1202 marking off values of the main vessel and / or side branch lumen volume / area 732, with values increasing in the direction of the arrow of horizontal axis 1202. Look-up table 1200 includes a vertical axis 1204 marking off values of the side branch angle 734, with values increase in the direction of the arrow of vertical axis 1204, i.e., more acute angles are smaller angles and more perpendicular angles are larger angles. In look-up table 1200 darker entries correspond to agreater likelihood of plaque or carina shift. Look-up tables may be structured organized in any number of ways and may include more than two dimensions (or only include one dimension).

[0111] As depicted, look-up table comprises three regions 1210, 1212, 1214 of differing scores / probabilities for different values of the lumen volume area 732 and side branch angle 734. A first region 1210 has entries which correspond to probabilities or scores that correspond to a greater likelihood of plaque or carina shift. First region 1210 generally corresponds to smaller side branch angles 734 and smaller branch lumens 732, both of which tend to result in plaque or carina shift more often than for large values. A second region 1212 has entries which correspond to probabilities or scores that correspond to a lesser likelihood of plaque or carina shift compared to the first region 1210 but a greater likelihood of plaque or carina shift compared to a third region 1214. Second region 1212 generally corresponds to smaller side branch angles 734 and smaller branch lumens 732 than those of third region 1214, but larger side branch angles 734 and larger branch lumens 732 than those of first region 1210. A third region 1214 has entries which correspond to probabilities or scores that correspond to a lesser likelihood of plaque or carina shift than the first region 1210 and second region 1212. Third region 1214 generally corresponds to larger side branch angles 734 and larger branch lumens 732, both of which tend to result in plaque or carina shift less often than for smaller values. The entries and regions depicted in lookup table 1200 are merely illustrative. In some embodiments, the entries may take any number of discrete values, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10, or any number of continuous values, e.g., in the interval [0,1],

[0112] Entry 1220 may correspond to a value of the look-up table for a single image frame, a group of frames, or an entire pullback. Entry 1220 is selected when side branch angles 734 are more acute and lumen volumes / areas are smaller. Thus, if the vessel measurements for a patient corresponded to entry 1220, then a plaque or carina shift is more likely. However, as shown for a relatively small increase in either of the side branch angle 734 or lumen volume / area 732, the likelihood of a plaque or carina shift becomes less likely.

[0113] Figure 13 is a look-up table based on degree of plaque shape 738 and plaque type 736, according to aspects of the present disclosure. Look-up table 1300 includes a plurality of entries, e.g., numerical values, each entry identified by a value of the plaque type 736 and plaque shape 738. Plaque type 736 may be represented as degree of plaque softness, relative amount (%) of soft plaque, or plaque type. Plaque shape 738 may be represented as degree of diffuseness. Insome embodiments, look-up table 1300 is two dimensional. Look-up table 1300 includes a horizontal axis 1302 marking off values of the degree of plaque softness, relative amount (%) of soft plaque, or plaque type 736, with values indicating softer plaque in the direction of the arrow of horizontal axis 1302. The degree of plaque softness and relative amount of soft plaque may take continuous, or nearly continuous, values, and plaque type may take on discrete values. In some embodiments, plaque type may have four discrete values, associated with necrotic core, dense calcium, fibrous, and fibro-fatty (going from harder to softer plaque). Look-up table 1300 includes a vertical axis 1304 marking off values of the degree of plaque diffuseness, with values indicating less diffuse (more focal) plaques in the direction of the arrow of vertical axis 1304. In some embodiments, the degree of plaque diffuseness 738 may be determined from slope of the increase in plaque volume / area / diameter as a function of distance along the vessel or it may be determined from the longitudinal area density. In other words, plaque diffuseness measures how quickly the plaque burden increases along the length of a vessel. In look-up table 1300 darker entries correspond to a greater likelihood of plaque or carina shift. Look-up tables may be structured organized in any number of ways and may include more than two dimensions (or only include one dimension).

[0114] As depicted, look-up table comprises three regions 1310, 1312, 1314 of differing scores / probabilities for different values of the plaque type 736 and plaque shape 738. A first region 1310 has entries which correspond to probabilities or scores that correspond to a greater likelihood of plaque or carina shift. First region 1310 generally corresponds to less diffuse plaque 738 and softer plaques 736, both of which tend to result in plaque or carina shift more often than for more diffuse and harder plaques. A second region 1312 has entries which correspond to probabilities or scores that correspond to a lesser likelihood of plaque or carina shift compared to the first region 1310 but a greater likelihood of plaque or carina shift compared to a third region 1314. Second region 1312 generally corresponds to less diffuse plaques 738 and softer plaques 736 than those of third region 1314, but more diffuse plaques 738 and harder plaques 736 than those of first region 1310. A third region 1314 has entries which correspond to probabilities or scores that correspond to a lesser likelihood of plaque or carina shift than the first region 1310 and second region 1312. Third region 1314 generally corresponds to more diffuse plaques 738 and harder plaques 736, both of which tend to result in plaque or carina shift less often than for less diffuse and softer plaques. The entries and regions depicted in look-up table1300 are merely illustrative. In some embodiments, the entries may take any number of discrete values, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10, or any number of continuous values, e.g., in the interval [0,1].

[0115] Entry 1320 may correspond to a value of the look-up table for a single image frame, a group of frames, or an entire pullback. Entry 1320 is selected when plaque is relatively less diffuse and sifter. Thus, if the vessel measurements for a patient corresponded to entry 1320, then a plaque or carina shift is more likely. However, as shown for a relatively more diffuse and harder plaque, the likelihood of a plaque or carina shift becomes less likely.

[0116] Figure 14 is a diagrammatic schematic view of a training system 1400 for a machine learning model 1430, according to aspects of the present disclosure. Training system 1400 trains the machine learning model 1430 to accurately predict the likelihood / probability of a plaque and / or carina shift. In some embodiments, machine learning model 1430 is trained by providing historic information from completed stent placements alongside associated vessel features and vessel measurements. Machine learning model 1430 may be trained using input of any of the intermediate outputs of plaque and carina shift prediction system 700. For example, training may done be with input comprising some or all the output of intravascular imaging 710, input comprising some or all the output of border identification module 720 (e.g., any of the various vessel borders or vessel contours), and / or input including one or more vessel measurements 732, 734, 736, 738 as described in Fig. 7. A machine learning model 1430 may take as input any combination of the previously listed inputs. In some embodiments, the machine learning model 1430 may be trained to do border identification and / or image segmentation. Training machine learning model 1430 aims to improve the performance of the model. Training system 1400 includes training data 1410, machine learning model 1430, and model objectives / functions 1450.

[0117] Training data 1410 may include data for a plurality of patient records 1420. Each patient record may contain a plurality of images or views, treatment outcomes, and / or a plurality of user annotations. In some embodiments, patient record 1420 may include pre-stent pullback intravascular images (radial / tomographic) 1480, pre-stent longitudinal view (image-based or graphical) 1482, post-stent pullback intravascular images (radial / tomographic) 1484, post-stent longitudinal view (image-based or graphical) 1486, post-stent outcomes 1488, user annotation of main vessel and / or side branch lumen volume / area 1490, user annotation of side branch angle 1492, user annotation of plaque type / softness 1494, user annotation of plaque shape (e.g., focal,diffuse, etc.) 1496, and / or user annotation of expectation of carina shift and / or plaque shift 1498. It should be appreciated that not all the patient records 1420 depicted in Fig. 14 needed to train the machine learning model 1430. Furthermore, additional medical records from a patient’s medical history may be included in the training data 1410.

[0118] Machine learning model 1430 may receive the training data 1410. From the training data 1410, machine learning model 1430 may output a prediction of the likelihood of carina shift and / or plaque shift 1440. For example, the output may be a number between 0 and 1. In some embodiments, the machine learning model may also output which of the vessel measurements contributed the most to the prediction. In some embodiments, machine learning model may receive pre-stent images 1480 and views 1482 from the training data 1410. From the pre-stent images 1480 and views 1483, machine learning model may generate vessel and / or lumen borders for main and / or side branches. Alternatively, machine learning model may generate vessel measurements, e.g., 732, 734, 736, 738, from images 1480 and views 1483.

[0119] Using model objectives / functions 1450 training system 1400 compares the prediction 1440 with associated ground truth predictions. For example, in some embodiments, a user annotation of a post-stent image or view may note a plaque and / or carina shift. The annotation identifying whether a shift occurred may act as the ground truth label. If a post-stent image or view is available but lacks an annotation in associated medical records, then ground truth labels can be generated by having a physical review the image. Alternatively, a separate machine learning model trained for object detection and / or image segmentation be used to analyze poststent images and / or views for evidence of plaque or carina shift. In some embodiments, poststent imaging or medical records recording plaque or carina shift may not be available. In such a case, the user annotation of an expectation of carina shift and / or plaque shift 1498 may be used as a ground truth label. The user annotation 1498 may generated when a physician reviews a prestent image.

[0120] In some embodiments, model objectives / functions 1450 may be used to compare predicted vessel measurements from a machine learning model 1430 with user annotated vessel measurements 1490, 1492, 1494, 1496, which represent the ground truth labels.

[0121] Model objectives / functions 1450 may include objectives / functions which penalize to a greater extent predicted likelihoods which are further from the ground truth labels. Model objectives / functions 1450 may also penalize predictions which are close to the ground truthlabels but which are based on the incorrect vessel measurements. For example, the machine learning model may predict a high probability of plaque or carina shift and explain it by identifying the side branch angle as the most important factor. However, if the ground truth data shows a high probability of a plaque or carina shift but the side branch angle is close to perpendicular, then the side branch angle is incorrectly identified as the explanatory measurement. In some embodiments, model objectives / functions 1450 may include objectives / functions which penalize deviations of predicted vessel measurements from user- annotated vessel measurements 1490, 1492, 1494, 1496.

[0122] Comparisons from the model objectives / functions 1450 may update parameters 1460 of the machine learning model 1430. In some instances, updating may be accomplished using gradient of the objective functions and backpropagation to update the parameters of the machine learning model. The structure of machine learning models is further described with respect to Fig. 19.

[0123] Figure 15 is a second diagrammatic schematic view of a training system 1500 for a machine learning model 1530, according to aspects of the present disclosure. After training, as described in Fig. 14, is complete an untrained machine learning model with parameters A 1530 is transformed into a trained machine learning model with parameters B. Parameters A and B differ between the trained and untrained models because over the course of training, parameters in the machine learning model are updated (e.g., 1460 in Fig. 14). based on comparisons between ground truth labels and predictions.

[0124] Figure 16 is a diagrammatic schematic view of a plaque and carina shift prediction system 1600 including a machine learning model 1641 receiving a plurality of vessel features and / or measurements as input, according to aspects of the present disclosure. System 1600 may determine a likelihood for carina shift and / or plaque shiftl642 and generate a display 1650 through which a physician may review the areas of a vessel where a shift is likely. The plaque and carina shift prediction system 1600 may include intravascular imaging 1610, border identification 1620, and carina shift and / or plaque shift prediction 1640.

[0125] Intravascular imaging 1610 may include use of an intraluminal imaging system 100 to gather intravascular imaging data to generate intravascular images 1612 (e.g., cross-sectional radial and / or tomographic views). Intraluminal imaging system 100 may process the intravascular images 1612 into longitudinal views 1614 (e.g., image-based longitudinal view asdepicted in the display of Fig. 6 and / or a graphical longitudinal view). Intravascular images 1612 and longitudinal views 1614 may be output to a border identification module 1620 and / or screen display 1650.

[0126] Border identification module 1620 process the images and view from intravascular imaging 1610 to identify various features in the image. Lumen borders 1622, such as those associated with the main vessel and / or side branch of a blood vessel, may be identified using techniques known in the art. Alternatively or in addition to lumen borders 1622, vessel borders 1624 may be identified using techniques known in the art. The borders of the plaque 1626 present in the intravascular images and longitudinal views may be identified separately from the lumen borders 1622 and vessel borders 1624. Alternatively, plaque 1626 may be identified using the shape / size of the space between the lumen border and vessel border. Incorporate by reference border detection.

[0127] After borders have been identified using border identification module 1620, various vessel measurements may be determined. Vessel measurements may include main vessel and / or side branch lumen volume / area 1632, side branch angle 1634, plaque type and / or plaque softness 1636, and / or plaque shape 1638. Main vessel and / or side branch lumen volume / area 1632 may be determined in various ways as described herein. Similar to the vessel measurements 650 in display 600 of Fig. 6, the main vessel and / or side branch lumen volume / area 1632 may be included in a display. Side branch angle 1634, e.g., as described further with respect to Figs. 8A- 8B, may be computed from distances between imaging planes and distances between the wall of the main vessel and side branch of a blood vessel. In some instances, the distance between the wall of the main vessel and side branch is defined to be the closest distance between the wall of the main vessel and side branch of a blood vessel as measured within an imaging plane. Plaque type and / or softness 1636 may be determined from the contrast in an intravascular image 1612 because the opacity of plaque to, for example, ultrasound depends on its softness. In some embodiments, virtual histology may be used to further identify plaque composition. For example, plaque identified in intravascular images may fall into four different composition types (from hardest to softest): necrotic core, dense calcium, fibrous, and fibro-fatty. These composition types may be indicated using color in intravascular images. Incorporate by reference virtual histology. Plaque shape 1638 may include determining whether a plaque is focal, diffuse, orirregular. Examples of various plaque shapes are depicted in and described with respect to Figs. 3-5.

[0128] Carina shift and / or plaque shift prediction module 1640 may receive main vessel and / or side branch lumen volume / area 1632, side branch angle 1634, plaque type and / or plaque softness 1636, and / or plaque shape 1638. Carina shift and / or plaque shift prediction module 1640 may generate a probability / likelihood of plaque and / or carina shifting, e.g., shifting as described in Figs. 3-5. Carina shift and / or plaque shift prediction module 1640 may identify a region of a blood vessel where a carina or plaque shift may occur. The probability of a plaque or carina shift and / or the region where the shift may occur may be used to determine if an indicator should be presented to a user, e.g., a physician. In some embodiments, carina shift and / or plaque shift prediction module 1640 includes a machine learning model 1641. Machine learning model 1641 may receive as input one or more of main vessel and / or side branch lumen volume / area 1632, side branch angle 1634, plaque type and / or plaque softness 1636, and / or plaque shape 1638 and generate a predicted likelihood of carina shift and / or plaque shift 1642. In some embodiments, the machine learning model 1641 has been trained as described in Figs. 14-15 and may have the structure of a deep learning network (e.g., a predictive network), as described in Fig. 19, support vector machine, or a multi-layer perceptron. The predicted likelihood of carina shift and / or plaque shift 1642 may be compared to a threshold 1644 for determining whether an indicator 1646 should be generated.

[0129] Screen display 1650 may provide various images and carina and plaque shift prediction to a user such as a physician. Screen display 1650 may include intravascular images 1612, longitudinal view 1614, and an indicator for carina or plaque shift prediction 1652, e.g., as generated by carina shift and / or plaque shift prediction module 1640 when threshold 1644 has been met. Indicator 1652 may include a region of the blood vessel where a plaque or carina shift is predicted to occur. Screen display 1650 may be similar to display 600 depicted in Fig. 6.

[0130] Figure 17 is a diagrammatic schematic view of a plaque and carina shift prediction system 1700 including a machine learning model 1721 receiving images and / or views, according to aspects of the present disclosure. System 1700 may determine a likelihood for carina shift and / or plaque shift and generate a display through which a physician may review the areas of a vessel where a shift is likely. The plaque and carina shift prediction system 1700 may include intravascular imaging 1710 and carina shift and / or plaque shift prediction module 1720.

[0131] Intravascular imaging 1710 may include use of an intraluminal imaging system 100 to gather intravascular imaging data to generate intravascular images 1712 (e.g., cross-sectional radial and / or tomographic views). Intraluminal imaging system 100 may process the intravascular images 1712 into longitudinal views 1714 (e.g., image-based longitudinal view as depicted in the display of Fig. 6 and / or a graphical longitudinal view). Intravascular images 1712 and longitudinal views 1714 may be output to a carina shift and / or plaque shift prediction module 1720 and / or screen display 1750.

[0132] Carina shift and / or plaque shift prediction module 1720 may receive intravascular images 1712 and longitudinal views 1714. Carina shift and / or plaque shift prediction module 1640 may generate a probability / likelihood of plaque and / or carina shifting, e.g., shifting as described in Figs. 3-5. Carina shift and / or plaque shift prediction module 1640 may identify a region of a blood vessel where a carina or plaque shift may occur. The probability of a plaque or carina shift and / or the region where the shift may occur may be used to determine if an indicator should be presented to a user, e.g., a physician. In some embodiments, carina shift and / or plaque shift prediction module 1720 includes a machine learning model 1721.

[0133] Machine learning model 1721 may receive as input intravascular images 1712 and / or longitudinal views 1714 and generate one or more of main vessel and / or side branch lumen volume / area 1722, side branch angle 1724, plaque type / softness 1726, plaque shape 1728, and / or a predicted likelihood of carina shift and / or plaque shift 1730. The machine learning model 1721 may generate vessel measurements 1722, 1724, 1726, 1728 as an intermediate step to generating the predicted likelihood of carina shift and / or plaque shift 1730. In some embodiments, a machine learning model 1721 may be trained for object detection, including detection of lumen borders and / or vessel borders of the main vessel and / or side branch. The machine learning model 1721 may, separately from or based on vessel and / or lumen borders, detect plaque in an intravascular image 1712 or longitudinal view 1714. The machine learning model may use a convolutional neural network-based architecture.

[0134] In some embodiments, the machine learning model 1721 may be trained as described in Figs. 14-15 and may have the structure of a convolutional neural network as described in Fig. 19. In some embodiments, machine learning model 1721 may comprise two models: one more object detection and / or vessel measurement and one for predicting likelihood of plaque or carinashift. The predicted likelihood of carina shift and / or plaque shift 1730 may be compared to a threshold 1732 for determining whether an indicator 1732 should be generated.

[0135] Screen display 1740 may provide various images and carina and plaque shift prediction 1742 to a user such as a physician. Screen display 1740 may include intravascular images 1712, longitudinal view 1714, and indicator for carina or plaque shift prediction 1742, e.g., as generated by machine learning model 1721. Indicator 752 may include a region of the blood vessel where a plaque or carina shift is predicted to occur. Screen display may be similar to display 600 depicted in Fig. 6.

[0136] Figure 18 is a diagrammatic schematic view of a plaque and carina shift prediction system 1800 including a machine learning model 1820 for object detection or tissue characterization, according to aspects of the present disclosure. System 1800 may determine a likelihood for carina shift and / or plaque shift 1842 and generate a display through which a physician may review the areas of a vessel where a shift is likely. The plaque and carina shift prediction system 1800 may include intravascular imaging 1810, machine learning model 1820, and carina shift and / or plaque shift prediction 1830.

[0137] Intravascular imaging 1810 may include use of an intraluminal imaging system 100 to gather intravascular imaging data to generate intravascular images 1812 (e.g., cross-sectional radial and / or tomographic views). Intraluminal imaging system 100 may process the intravascular images 1812 into longitudinal views 1814 (e.g., image-based longitudinal view as depicted in the display of Fig. 6 and / or a graphical longitudinal view). Intravascular images 1812 and longitudinal views 1814 may be output to a machine learning model 1820 and / or screen display 1850.

[0138] Machine learning model 1820 may receive as input intravascular images 1812 and / or longitudinal views 1814 and generate one or more of main vessel and / or side branch lumen volume / area 1822, side branch angle 1824, plaque type / softness 1826, or plaque shape 1828. In some embodiments, a machine learning model 1820 may be trained for object detection and / or segmentation, including detection of lumen borders and / or vessel borders of the main vessel and / or side branch. The machine learning model 1820 may, separately or based on vessel and / or lumen borders, detect / segment plaque in an intravascular image 1812 or longitudinal view 1814. The machine learning model may use a convolutional neural network-based architecture as described in Fig. 19.

[0139] Carina shift and / or plaque shift prediction module 1100 may receive main vessel and / or side branch lumen volume / area 732, side branch angle 734, plaque type and / or plaque softness 736, and / or plaque shape 738. Carina shift and / or plaque shift prediction module 1100 may generate a probability / likelihood of plaque and / or carina shifting, e.g., shifting as described in Figs. 3-5. Carina shift and / or plaque shift prediction module 1100 may identify a region of a blood vessel where a carina or plaque shift may occur. The probability of a plaque or carina shift and / or the region where the shift may occur may be used to determine if an indicator should be presented to a user, e.g., a physician. Carina shift and / or plaque shift prediction module 1100 is has structure similar to the one described in Fig. 11, with vessel measurements provided by machine learning model 1820.

[0140] Screen display 1850 may provide various images and carina and plaque shift prediction to a user such as a physician. Screen display 1850 may include intravascular images 1812, longitudinal view 1814, and indicator for carina or plaque shift prediction 1852, e.g., as generated by carina shift and / or plaque shift prediction module 1100. Indicator 1852 may include a region of the blood vessel where a plaque or carina shift is predicted to occur. Screen display may be similar to display 600 depicted in Fig. 6.

[0141] Figure 19 is a diagrammatic schematic of a deep learning network (e.g., a predictive network) configuration 1900, according to aspects of the present disclosure. The configuration 1900 can be implemented by a deep learning network. The configuration 1900 includes a deep learning network 1910 including one or more CNNs 1912. For simplicity of illustration and discussion, Fig. 19 illustrates one CNN 1912. However, the embodiments can be scaled to include any suitable number of CNNs 1912 (e.g., about 2, 3 or more). The configuration 1900 can be trained for identification of various anatomical landmarks or features within a patient anatomy, including a region of crossover of an iliac artery with an iliac vein, pelvic bone notches or other anatomical landmarks or features which may be used to identify the location of an inguinal ligament, and / or other regions of blood flow restriction (e.g., stenosis or compression) as described in greater detail below.

[0142] The CNN 1912 may include a set of N convolutional layers 1920 followed by a set of K fully connected layers 1930, where N and K may be any positive integers. The convolutional layers 1920 are shown as 1920(1) to 1920(N). The fully connected layers 1930 are shown as 1930(1) to 1930(K). Each convolutional layer 1920 may include a set of filters 1922 configuredto extract features from an input 1902 (e.g., one or more intravascular image frames, imagebased longitudinal view, graphical longitudinal view, etc.). The values N and K and the size of the filters 1922 may vary depending on the embodiments. In some instances, the convolutional layers 1920(1) to 1920(N) and the fully connected layers 1930(1) to 1930(K-l) may utilize a leaky rectified non-linear (ReLU) activation function and / or batch normalization. The fully connected layers 1930 may be non-linear and may gradually shrink the high-dimensional output to a dimension of the prediction result (e.g., the classification output 1940). Thus, the fully connected layers 1930 may also be referred to as a classifier. In some embodiments, the fully convolutional layers 1920 may additionally be referred to as perception or perceptive layers.

[0143] The classification output 1940 may indicate a confidence score for each class 1942 based on the input image 1902. The classes 1942 are shown as 1942a, 1942b, ... , 1942c. When the CNN 1912 is trained for predicting plaque shift and / or carina shift of stenosis or general venous compression, the classes 1942 may indicate a probability for plaque shift 1942a, a probability of carina shift 542b, any other outputs 1942c of machine learning models as described herein, or any other suitable class. A class 1942 indicating a high confidence score indicates that the input image 1902 or a section or pixel of the image 1902 is likely to include an anatomical object / feature of the class 1942. Conversely, a class 1942 indicating a low confidence score indicates that the input image 1902 or a section or pixel of the image 1902 is unlikely to include an anatomical object / feature of the class 1942.

[0144] The CNN 1912 can also output a feature vector 1950 at the output of the last convolutional layer 1920(N). The feature vector 1950 may indicate objects detected from the input image 1902 or other data. For example, the feature vector 1950 may indicate a region where a plaque shift and / or carina shift may occur upon stenting identified from the image 1902. The feature vector 1950 may indicate the borders of lumens associated with the blood vessels, borders of the vessels, main vessel and / or side branch lumen volume / area, side branch angle, plaque type / softness, plaque shape, etc.

[0145] The deep learning network 1910 (e.g., a predictive network) may implement or include any suitable type of learning network. For example, in some embodiments and as described in relation to Fig. 19, the deep learning network 1910 could include a convolutional neural network 1912. In addition, the convolutional neural network 1910 may additionally oralternatively be or include a multi-class classification network, an encoder-decoder type network, or any suitable network or means of identifying features within an image.

[0146] In an embodiment in which the deep learning network 1910 includes an encoderdecoder network, the network may include two paths. One path may be a contracting path, in which a large image, such as the image 1902, may be convolved by several convolutional layers 1920 such that the size of the image 1902 changes in depth of the network. The image 1902 may then be represented in a low dimensional space, or a flattened space. From this flattened space, an additional path may expand the flattened space to the original size of the image 1902. In some embodiments, the encoder-decoder network implemented may also be referred to as a principal component analysis (PCA) method. In some embodiments, the encoder-decoder network may segment the image 1902 into patches.

[0147] In an additional embodiment of the present disclosure, the deep learning network 1910 may include a multi-class classification network. In such an embodiment, the multi-class classification network may include an encoder path. For example, the image 1902 may be of a high dimensional image. The image 1902 may then be processed with the convolutional layers 1920 such that the size is reduced. The resulting low dimensional representation of the image 1902 may be used to generate the feature vector 1950 shown in Fig. 19. The low dimensional representation of the image 1902 may additionally be used by the fully connected layers 1930 to regress and output one or more classes 1942. In some regards, the fully connected layers 1930 may process the output of the encoder or convolutional layers 1920. The fully connected layers 1930 may additionally be referred to as task layers or regression layers, among other terms.

[0148] Any suitable combination or variations of the deep learning network 1910 described is fully contemplated. For example, the deep learning network may include fully convolutional networks or layers or fully connected networks or layers or a combination of the two. In addition, the deep learning network may include a multi-class classification network, an encoder-decoder network, or a combination of the two.

[0149] The logical operations making up the aspects of the technology described herein are referred to variously as operations, steps, objects, elements, components, modules, etc. Furthermore, it should be understood that these may occur or be performed or arranged in any order, unless explicitly claimed otherwise or a specific order is inherently necessitated by the claim language.

[0150] As used herein, the term “blood vessel” may refer to both a main vessel and side branch or “blood vessel” may refer to the main vessel. It will be apparent from the context which use of “blood vessel” is intended.

[0151] All directional references e.g., upper, lower, inner, outer, upward, downward, left, right, lateral, front, back, top, bottom, above, below, vertical, horizontal, clockwise, counterclockwise, proximal, and distal are only used for identification purposes to aid the reader’s understanding of the claimed subject matter, and do not create limitations, particularly as to the position, orientation, or use of the aspects described herein. Connection references, e.g., attached, coupled, connected, joined, or “in communication with” are to be construed broadly and may include intermediate members between a collection of elements and relative movement between elements unless otherwise indicated. As such, connection references do not necessarily imply that two elements are directly connected and in fixed relation to each other. The term “or” shall be interpreted to mean “and / or” rather than “exclusive or.” The word “comprising” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude a plurality. Unless otherwise noted in the claims, stated values shall be interpreted as illustrative only and shall not be taken to be limiting.

[0152] The modules described herein can be software, firmware, hardware, or a combination of them. The processor circuit can include, implement, and / or execute the modules (e.g., the software, firmware, hardware, or a combination of them).

[0153] A system of one or more computers can be configured to perform particular operations or actions by virtue of having software, firmware, hardware, or a combination of them installed on the system that in operation causes or cause the system to perform the actions. One or more computer programs can be configured to perform particular operations or actions by virtue of including instructions that, when executed by data processing apparatus, cause the apparatus to perform the actions.

[0154] One general aspect includes an apparatus a processor circuit configured for communication with an intravascular imaging catheter, where the processor circuit is configured to: control the intravascular imaging catheter to obtain a plurality of intravascular images during movement through a first blood vessel of a patient before a stent is deployed inside the first blood vessel; determine, based on the plurality of intravascular images, at least one of: first data representative of plaque within the first blood vessel; or second data representative of a secondblood vessel connected to the first blood; determine, based on at least one of the first data or the second data, a likelihood for at least one of: a carina between the first blood vessel and the second vessel to shift when the stent is deployed; or the plaque to shift when the stent is deployed; and output, to a display in communication with the processor circuit, a screen display based on the likelihood for at least one of the carina to shift or the plaque to shift. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0155] Implementations may include one or more of the following features. The apparatus where the first blood vessel may include a main vessel and the second blood vessel may include a side branch extending from the main vessel. The screen display may include a first indicator representative of the likelihood for at least one of the carina to shift or the plaque to shift. The screen display may include a longitudinal view of the first blood vessel associated with the plurality of intravascular images, where the first indicator is overlaid on the longitudinal view. The first indicator may include one or more positions on the longitudinal view that identify one or more locations along the first blood vessel associated with the likelihood for at least one of the carina to shift or the plaque to shift. The first indicator may include a region of the longitudinal view corresponding to a subset of the plurality of intravascular images. The first indicator is configured to prompt a user to review at least one of the region of the longitudinal view or the subset of the plurality of intravascular images. The screen display may include an intravascular image of the plurality of intravascular images that belongs to the subset of the plurality of intravascular images. The processor circuit is configured to perform co-registration between the plurality of intravascular images and an x-ray image of the first blood vessel with contrast, where the screen display may include the x-ray image and a second indicator representative of the likelihood for at least one of the carina to shift or the plaque to shift, where the second indicator is overlaid on the x-ray image, and where the second indicator may include one or more positions on the x-ray image that identify the one or more locations along the first blood vessel associated with the likelihood for at least one of the carina to shift or the plaque to shift. The first indicator may include a text description associated with the likelihood for at least one of the carina to shift or the plaque to shift. The processor circuit is configured to: receive a user input defining a user threshold for at least one of the first data or the second data; and determinewhether or not to provide the first indicator in the screen display based on at least one of the first data or the second data satisfying the user threshold. To determine at least one of the first data or the second data, the processor circuit is configured to: provide the plurality of intravascular images as an input to a predictive network; and generate at least one the first data or the second data as an output of the predictive network. To the determine the likelihood for at least one of the carina to shift or the plaque to shift, the processor circuit is configured to: provide the plurality of intravascular images as an input to a predictive network; and generate the likelihood for at least one of the carina to shift or the plaque to shift as an output of the predictive network. The first data may include at least one of: a type of the plaque; a degree of softness of the plaque; a relative amount of the plaque that is relatively softer than other parts of the plaque that are relatively harder; a shape of the plaque along a length of the first blood vessel; or a degree of how focal or diffuse the plaque is. The second data may include at least one of: a volume of the second blood vessel; an area of the second blood vessel; or an angle between the first blood vessel and the second blood vessel. The processor circuit is configured determine the angle between the first blood vessel and the second blood vessel based on: a difference between the distances in a first intravascular image and a second intravascular image between an anatomical border of the first blood vessel and an anatomical border in the second blood vessel; a distance along a length of the vessel between the first intravascular image and the second intravascular image. The apparatus may include the intravascular imaging catheter. Implementations of the described techniques may include hardware, a method or process, or computer software on a computer-accessible medium.

[0156] One general aspect includes a method. The method also includes controlling, with the processor circuit, an intravascular imaging catheter to obtain a plurality of intravascular images during movement through a first blood vessel of a patient before a stent is deployed inside the first blood vessel; determining, with the processor circuit, at least one of: first data representative of plaque within the first blood vessel; or second data representative of a second blood vessel connected to the first blood; determining, with the processor circuit, a likelihood for at least one of: a carina between the first blood vessel and the second vessel to shift when the stent is deployed; or the plaque to shift when the stent is deployed, and outputting, to a display, a screen display based on the likelihood for at least one of the carina to shift or the plaque to shift. Other embodiments of this aspect include corresponding computer systems, apparatus, and computerprograms recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0157] The above specification, examples and data provide a complete description of the structure and use of exemplary aspects as defined in the claims. Although various aspects of the claimed subject matter have been described above with a certain degree of particularity, or with reference to one or more individual aspects, those skilled in the art could make numerous alterations to the disclosed aspects without departing from the spirit or scope of the claimed subject matter.

[0158] Still other aspects are contemplated. It is intended that all matter contained in the above description and shown in the accompanying drawings shall be interpreted as illustrative only of particular aspects and not limiting. Changes in detail or structure may be made without departing from the basic elements of the subject matter as defined in the following claims.

Claims

CLAIMSWhat is claimed is:

1. An apparatus, comprising: a processor circuit configured for communication with an intravascular imaging catheter, wherein the processor circuit is configured to: control the intravascular imaging catheter to obtain a plurality of intravascular images during movement through a first blood vessel of a patient before a stent is deployed inside the first blood vessel; determine, based on the plurality of intravascular images, at least one of: first data representative of plaque within the first blood vessel; or second data representative of a second blood vessel connected to the first blood; determine, based on at least one of the first data or the second data, a likelihood for at least one of: a carina between the first blood vessel and the second blood vessel to shift when the stent is deployed; or the plaque to shift when the stent is deployed; and output, to a display in communication with the processor circuit, a screen display based on the likelihood for at least one of the carina to shift or the plaque to shift.

2. The apparatus of claim 1, wherein the first blood vessel comprises a main vessel and the second blood vessel comprises a side branch extending from the main vessel.

3. The apparatus of claim 1, wherein the screen display comprises a first indicator representative of the likelihood for at least one of the carina to shift or the plaque to shift.

4. The apparatus of claim 3, wherein the screen display comprises a longitudinal view of the first blood vessel associated with the plurality of intravascular images,wherein the first indicator is overlaid on the longitudinal view.

5. The apparatus of claim 4, wherein the first indicator comprises one or more positions on the longitudinal view that identify one or more locations along the first blood vessel associated with the likelihood for at least one of the carina to shift or the plaque to shift.

6. The apparatus of claim 5, wherein the first indicator comprises a region of the longitudinal view corresponding to a subset of the plurality of intravascular images.

7. The apparatus of claim 6, wherein the first indicator is configured to prompt a user to review at least one of the region of the longitudinal view or the subset of the plurality of intravascular images.

8. The apparatus of claim 6, wherein the screen display comprises an intravascular image of the plurality of intravascular images that belongs to the subset of the plurality of intravascular images.

9. The apparatus of claim 5, wherein the processor circuit is configured to perform co-registration between the plurality of intravascular images and an x-ray image of the first blood vessel with contrast, wherein the screen display comprises the x-ray image and a second indicator representative of the likelihood for at least one of the carina to shift or the plaque to shift, wherein the second indicator is overlaid on the x-ray image, and wherein the second indicator comprises one or more positions on the x-ray image that identify the one or more locations along the first blood vessel associated with the likelihood for at least one of the carina to shift or the plaque to shift.

10. The apparatus of claim 3, wherein the first indicator comprises a text description associated with the likelihood for at least one of the carina to shift or the plaque to shift.

11. The apparatus of claim 3, wherein the processor circuit is configured to:receive a user input defining a user threshold for at least one of the first data or the second data; and determine whether or not to provide the first indicator in the screen display based on at least one of the first data or the second data satisfying the user threshold.

12. The apparatus of claim 11, wherein, to determine at least one of the first data or the second data, the processor circuit is configured to: provide the plurality of intravascular images as an input to a predictive network; and generate at least one the first data or the second data as an output of the predictive network.

13. The apparatus of claim 1, wherein, to the determine the likelihood for at least one of the carina to shift or the plaque to shift, the processor circuit is configured to: provide the plurality of intravascular images as an input to a predictive network; and generate the likelihood for at least one of the carina to shift or the plaque to shift as an output of the predictive network.

14. The apparatus of claim 1, wherein the first data comprises at least one of: a type of the plaque; a degree of softness of the plaque; a relative amount of the plaque that is relatively softer than other parts of the plaque that are relatively harder; a shape of the plaque along a length of the first blood vessel; or a degree of how focal or diffuse the plaque is.

15. The apparatus of claim 1, wherein the second data comprises at least one of: a volume of the second blood vessel; an area of the second blood vessel; or an angle between the first blood vessel and the second blood vessel.

16. The apparatus of claim 15, wherein the processor circuit is configured determine the angle between the first blood vessel and the second blood vessel based on: a difference between distances in a first intravascular image and a second intravascular image between an anatomical border of the first blood vessel and an anatomical border in the second blood vessel; a distance along a length of the vessel between the first intravascular image and the second intravascular image.

17. The apparatus of claim 1, further comprising the intravascular imaging catheter.

18. A method, comprising: controlling, with a processor circuit, an intravascular imaging catheter to obtain a plurality of intravascular images during movement through a first blood vessel of a patient before a stent is deployed inside the first blood vessel; determining, with the processor circuit, at least one of: first data representative of plaque within the first blood vessel; or second data representative of a second blood vessel connected to the first blood; determining, with the processor circuit, a likelihood for at least one of: a carina between the first blood vessel and the second blood vessel to shift when the stent is deployed; or the plaque to shift when the stent is deployed, and outputting, to a display, a screen display based on the likelihood for at least one of the carina to shift or the plaque to shift.

Images