WO2025106526A2

WO2025106526A2 - System and method for vessel evaluation using virtual flow reserve

Info

Publication number: WO2025106526A2
Application number: PCT/US2024/055694
Authority: WO
Inventors: Wei C. Chiu; Ajay Gopinath; Mark H. Hoeveler; Christopher E. Griffin; Vivek N. BHATIA; Humphrey Chen; Songyuan TANG; Matthew BRINKMAN
Original assignee: LightLab Imaging Inc
Current assignee: LightLab Imaging Inc
Priority date: 2023-11-17
Filing date: 2024-11-13
Publication date: 2025-05-22
Anticipated expiration: 2026-05-17
Also published as: WO2025106526A3

Abstract

The technology generally relates to systems and methods for predicting the impact of a potential treatment option for a blood vessel based on characteristics of the blood vessel. The characteristic may include, for example, virtual flow reserve (VFR). Different regions of interest may be selected and the value of the characteristic may be predicted based on a potential treatment option being implemented in the selected region. The technology further relates to system and methods for automatically reviewing a classification of a segment of a vessel as stenotic. An artificial intelligence (Al) model may be implemented to generate a lumen mask of the blood vessel within a region of interest. Based on a measure of similarity between the lumen contour determined by a lumen algorithm, different than the Al model, and the lumen mask, an indication that the classification of the segment as stenotic is a false positive may be provided.

Description

SYSTEM AND METHOD FOR VESSEL EVALUATION USING VIRTUAL FLOW RESERVE

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of the filing date of U.S. Provisional Patent Application No. 63/712,057, filed October 25, 2024, U.S. Provisional Patent Application No. 63/667, 172, filed July 3, 2024, U.S. Provisional Patent Application No. 63/557,883, filed February 26, 2024, and U.S. Provisional Patent Application No. 63/600,373, filed November 17, 2023, the disclosures of which are incorporated herein by reference.

BACKGROUND

[0002] The placement of stents in coronary arteries requires a significant amount of planning. Such planning may be accomplished by the physician with longitudinal photographs of the coronary vessel and a ruler. This has inherent limitations. Further, in the case of complex lesions, the optimal deployment location and stent size cannot be determined from viewing a cross-sectional presentation of the vessel alone. Various factors can change which stent should be used and where it should be placed that are not apparent based on a manual review of images. Even an experienced cardiologist may find it challenging to predict the stent size to use and to select a placement location that would result in the best outcome.

BRIEF SUMMARY

[0003] The technology described herein generally relates to systems and methods for predicting the impact of a potential treatment option for a blood vessel based on characteristics of the blood vessel. The characteristic may include, for example, virtual flow reserve (VFR). Different regions of interest may be selected and the value of the characteristic may be predicted based on a potential treatment option being implemented in the selected region. The technology further relates to system and methods for automatically reviewing a classification of a segment of a vessel as stenotic. An artificial intelligence (Al) model may be implemented to generate a lumen mask of the blood vessel within a region of interest. Based on a measure of similarity between the lumen contour determined by a lumen algorithm, different than the Al model, and the lumen mask, an indication that the classification of the segment as stenotic is a false positive may be provided

[0004] One aspect of this disclosure is directed to a computer implemented method of predicting an impact of a treatment option for a blood vessel characteristic, the method comprising receiving, by one or more processors, image data for a blood vessel; receiving, by the one or more processors, measurements of a characteristic of the blood vessel; outputting for display the measured characteristic; receiving a selection of a zone for implementing a potential treatment option; and predicting, by the one or more processors responsive to receiving the selection, an updated characteristic based on the selected zone.

[0005] The measurements may be received at, or in respect of, multiple points along the blood vessel. Outputting the measured characteristic for display may comprise outputting for display the characteristic in correlation with the multiple points along the blood vessel. Outputting for display the measured characteristic may comprise outputting at least one of a numerical value or a graphical representation. The potential d option may include placement of a stent having a size and placement location corresponding to the selected zone. Receiving the selection of the zone may comprise receiving adjustments to a boundary of an overlay depicted in relation to a representation of the blood vessel.

[0006] The method may further comprise outputting for display a first depiction of the zone relative to a first representation of the blood vessel and a second depiction of the zone relative to a second representation of the blood vessel. The method may further comprise receiving input modifying a boundary of the first depiction of the zone; and automatically updating the second depiction of the zone based on the input.

[0007] The method may further comprise determining whether the predicted updated characteristic meets a predetermined threshold. The method may further comprise outputting a representation of the predicted updated characteristic in a first mode if the predicted updated characteristic meets the predetermined threshold, and outputting the representation of the predicted updated characteristic in a second mode if the predicted updated characteristic does not meet the predetermined threshold.

[0008] In some examples, the computer implemented method of predicting an impact of a treatment option for a blood vessel characteristic may comprise receiving, by one or more processors, image data of the blood vessel, determining using the image data, by the one or more processors, one or more actual values of the characteristic, outputting for display one or more of the actual values of the characteristic, receiving a user selection of a zone of the blood vessel for implementing a potential treatment option, and determining, by the one or more processors, one or more predicted values of the characteristic which would occur if the potential treatment option was implemented as defined by the selected zone.

[0009] The actual and/or predicted values of the characteristic may be determined at the multiple positions along the blood vessel. Outputting for display the one or more of the actual values of the characteristic may comprise outputting for display a graphical representation of the actual values in spatial correlation with a graphical representation of the blood vessel. Outputting for display the one or more of the actual values of the characteristic may comprise outputting at least one of a numerical depiction of an actual value of the characteristic or a graphical representation of actual values of the characteristic.

[0010] The potential treatment option may include placement of a stent having a size and placement location corresponding to the selected zone. Receiving the user selection of the zone may comprise receiving from the user adjustments to a boundary of an overlay depicted in relation to a graphical representation of the blood vessel.

[0011] The method may further comprise outputting for display a first depiction of the user selected zone relative to a first representation of the blood vessel and a second depiction of the user selected zone relative to a second representation of the blood vessel. The method may further comprise receiving user input modifying a boundary of the first depiction of the zone; and automatically updating the second depiction of the zone based on the user input.

[0012] The method may further comprise determining whether one or more of the predicted values of the characteristic meet a predetermined threshold. The method may further comprise outputting a representation of the predicted values of the characteristic in a first mode if the predicted updated values characteristic meet the predetermined threshold, and outputting the representation of the predicted values of the characteristic in a second mode if the predicted updated measures do not meet the predetermined threshold.

[0013] The image data used to determine the one or more actual values of the characteristic may be intravascular image data of the blood vessel. The characteristic may be virtual flow reserve. Outputting for display the one or more of the actual values of the characteristic may comprise providing for output, by the one or more processors, a color map on at least one of an extraluminal image or a graphical representation of the actual values. The color map comprises one or more colors, each of the one or more colors corresponding to a respective value or range of values of the characteristic.

[0014] Another aspect of this disclosure is directed to a computer implemented method, comprising receiving, by one or more processors, intravascular imaging data of a vessel; determining, by the one or more processors, a lumen contour for at least one frame of the intravascular imaging data; determining, by the one or more processors executing an artificial intelligence model, a lumen mask for the at least one frame of the intravascular imaging data; comparing, by the one or more processors, the determined lumen contour and the lumen mask; determining, by the one or more processors based on the comparison, an overlap value; and when the overlap value for the at least one frame of the intravascular imaging data is below a threshold, providing for output, by the one or more processors, an indication that the at least one frame of the intravascular imaging data is a false positive.

[0015] The method may further comprise thresholding, by the one or more processors, the overlap value, and when the thresholded overlap value for the at least one frame of the intravascular imaging data is below the threshold, providing for output, by the one or more processors, the indication that the at least one frame of the intravascular imaging data is the false positive. The false positive may correspond to a frame of intravascular imaging data that was wrongly identified as being stenotic. The overlap value may be a Jaccard index value. The lumen mask may identify regions of the at least one frame within a lumen of the vessel and regions of the at least one frame outside the lumen of the vessel.

[0016] The method may further comprise generating, by the one or more processors based on the intravascular imaging data, a two-dimensional representation of the vessel, wherein the two-dimensional representation is symmetrical about a longest axis of the two-dimensional representation. The method may further comprise receiving, by the one or more processors, an input relative to the two-dimensional representation. The input may correspond to a selection of a plurality of frames of the intravascular imaging data, and the selection of the plurality of frames may include the at least one frame of the intravascular imaging data that is the false positive. The method may further comprise interpolating, by the one or more processors, the plurality of frames to taper the lumen profile for the selected plurality of frames. The method may further comprise updating, by the one or more processors, additional values associated with the vessel based on the interpolation of the plurality of frames. The additional values includes at least one of virtual flow reserve (VFR) value for a selected frame, a baseline VFR value, or a maximum VFR value. The method may further comprise generating, by the one or more processed based on the intravascular imaging data, a graphical representation; and updating, by the one or more processors based on the interpolated plurality of frames, the graphical representation. The graphical representation is a representation of virtual flow reserve (VFR) values or pressure values.

[0017] The computer implemented method may be a method of automatically reviewing a classification of a segment of a blood vessel as stenotic. The method may comprise receiving, by one or more processors, one or more frames of intravascular data of the segment, determining from at least one of the frames, by the one or more processors using a lumen algorithm, a lumen contour of the blood vessel within the segment, determining from at least one of the frames, by the one or more processors using an artificial intelligence model separate to the lumen algorithm, a lumen mask of the blood vessel within the segment, determining, by the one or more processors, a measure of similarity between the lumens defined by the lumen contour and the lumen mask, and if the measure of similarity is below a threshold, providing for output, by the one or more processors, an indication that the classification of the segment as stenotic is a false positive.

[0018] The lumen algorithm may be a contour tracking algorithm and the determined lumen contour may be a determined boundary of the lumen within the segment. The lumen mask may comprise pixels each indicative of whether the pixel is inside or outside the lumen. The measure of similarity may be a Jaccard index value. The Jaccard index value may be calculated by dividing a cross sectional amount of the segment determined by both the lumen contour and the lumen mask to be within the lumen by a cross sectional amount of the segment determined by either or both of the lumen contour and the lumen mask to be within the lumen.

[0019] The method may further comprise generating, by the one or more processors based on the intravascular imaging data, a two-dimensional representation of the vessel, wherein the two-dimensional representation is symmetrical about a longest axis of the two-dimensional representation .The method may further comprise receiving, by the one or more processors, an input relative to the two-dimensional representation. The input may correspond to a selection of a plurality of frames of the intravascular imaging data, and the selection of the plurality of frames may include the at least one frame of the intravascular imaging data that is the false positive. The method may further comprise interpolating, by the one or more processors, the plurality of frames to taper the lumen profile for the selected plurality of frames. The method may further comprise updating, by the one or more processors, additional values associated with the vessel based on the interpolation of the plurality of frames. The additional values may include at least one of virtual flow reserve (VFR) value for a selected frame, a baseline VFR value, or a maximum VFR value. The method may further comprise generating, by the one or more processors based on the intravascular imaging data, a graphical representation, and updating, by the one or more processors based on the interpolated plurality of frames, the graphical representation. The graphical representation may be a representation of virtual flow reserve (VFR) values or pressure values. The graphical representation may a color map comprising one or more colors, each of the one or more colors corresponding to a respective value or range of values of the VFR values or pressure values

[0020] Aspects of the disclosure also relate to computer apparatus, or systems, arranged to carry out the above methods and other methods described herein. For example, another aspect of the disclosure is directed to a system for predicting an impact of a treatment option for a blood vessel characteristic, the system comprising one or more processors. The one or more processors may be configured to receive image data for a blood vessel, receive measurements of a characteristic of the blood vessel, output for display the measured characteristic, receive a selection of a zone for implementing a potential treatment option, and predict, responsive to receiving the selection, an updated characteristic based on the selected zone.

[0021] Similarly, another aspect of the disclosure is directed to a system comprising one or more processors. The one or more processors may be configured to receive intravascular imaging data of a vessel, determine a lumen contour for at least one frame of the intravascular imaging data, determine, by executing an artificial intelligence model, a lumen mask for the at least one frame of the intravascular imaging data, compare the determined lumen contour and the lumen mask, determine, based on the comparison, an overlap value, and when the overlap value for the at least one frame of the intravascular imaging data is below a threshold, provide for output an indication that the at least one frame of the intravascular imaging data is a false positive.

[0022] Aspects of the disclosure also relate to computer readable media carrying computer program code arranged to carry out the above methods, or other methods described herein. For example, another aspect of the disclosure is directed to one or more non-transitory computer-readable storage media encoding instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising receiving image data for a blood vessel, receiving measurements of a characteristic of the blood vessel, outputting for display the measured characteristic, receiving a selection of a zone for implementing a potential treatment option, and predicting, responsive to receiving the selection, an updated characteristic based on the selected zone.

[0023] Similarly, another aspect of the disclosure is directed to one or more non-transitory computer- readable storage media encoding instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising receiving intravascular imaging data of a vessel, determining a lumen contour for at least one frame of the intravascular imaging data, determining, by executing an artificial intelligence model, a lumen mask for the at least one frame of the intravascular imaging data, comparing the determined lumen contour and the lumen mask, determining, based on the comparison, an overlap value, and when the overlap value for the at least one frame of the intravascular imaging data is below a threshold, providing for output, by the one or more processors, an indication that the at least one frame of the intravascular imaging data is a false positive.

BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0025] Figure 1 is an example data collection system for use in respect of a blood vessel according to aspects of the disclosure. [0026] Figure 2 is an example system which can be used to provide determinations or predictions in respect of a blood vessel, for example in the context of the data collection system, according to aspects of the disclosure of figure 1.

[0027] Figures 3 to 7 illustrate example interface screens which may be implemented using the data collection system of Figure 1, according to aspects of the disclosure.

[0028] Figure 8 is a flow diagram illustrating a method of outputting a representation of a blood vessel which can be used in providing the interface screens of Figures 3 to 7, according to aspects of the disclosure. [0029] Figure 9A illustrates another example interface screen which may be implemented, for example, in the context of the data collection system of Figure 1, according to aspects of the disclosure.

[0030] Figure 9B illustrates another example interface screen which may be implemented for example in the context of the data collection system of Figure 1, and in which some frames of intraluminal image data are identified as false positives, according to aspects of the disclosure.

[0031] Figure 10 is a block diagram of an example detection system which may be used to identify false positives, such as those illustrated in Figure 9, and which may be implemented at least in part using the data collection system of Figure 1, according to aspects of the disclosure.

[0032] Figure 11 A illustrates another example interface screen in which the false positives of Figures 9B and 10 are used, according to aspects of the disclosure.

[0033] Figure 11B illustrates another example interface screen which may be implemented, for example, in the context of the data collection system of Figure 1, according to aspects of the disclosure.

[0034] Figures 12A and 12B illustrate further example interface screens before and after a lumen bridge module has been applied using frames identified as false positives, according to aspects of the disclosure.

[0035] Figure 13 is a flow diagram illustrating a method of identifying false positives in frames for example, using the detection system of Figure 10 and for use in respect of the interface screens of Figures 9B, 11, 12A and 12B, according to aspects of the disclosure.

[0036] Figure 14 is an example two-dimensional representation and pressure curve, according to aspects of the disclosure.

[0037] Figure 15 is another example two-dimensional representation and pressure curve, according to aspects of the disclosure.

[0038] Figure 16 is another example two-dimensional representation and pressure curve, according to aspects of the disclosure.

[0039] Figures 17A and 17B illustrate further example interface screens for providing a color map, which may be implemented, for example, in the context of the data collection system of Figure 1 , according to aspects of the disclosure.

[0040] Figures 18A-18D illustrate further example interface screens for providing an automatic update for regions of interest within an imaged vessel, which may be implemented, for example, in the context of the data collection system of Figure 1, according to aspects of the disclosure. [0041] Figures 19A-19E illustrate further example interface screens for performing an expansion analysis within an imaged vessel, which may be implemented, for example, in the context of the data collection system of Figure 1, according to aspects of the disclosure.

DETAILED DESCRIPTION

[0042] Some portions of the detailed description are presented in terms of methods such as algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations can be used by those skilled in the computer and software related fields. In one embodiment, an algorithm is here, and generally, conceived to be a self-consistent sequence of operations leading to a desired result. The operations performed as methods stops or otherwise described herein are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, transformed, compared, and otherwise manipulated.

[0043] The algorithms and displays presented herein are not inherently related to any particular computer or other apparatus. Various general purpose systems may be used with programs in accordance with the teachings herein, or it may prove convenient to construct more specialized apparatus to perform the required method steps. The required structure for a variety of these systems will appear from the description below.

[0044] The aspects, embodiments, features, and examples of the disclosure are to be considered illustrative in all respects and are not intended to limit the disclosure, the scope of which is defined only by the claims. Other embodiments, modifications, and usages will be apparent to those skilled in the art without departing from the spirit and scope of the claimed invention.

[0045] The use of headings and sections in the application is not meant to limit the invention; each section can apply to any aspect, embodiment, or feature of the invention.

[0046] Throughout the application, where compositions are described as having, including, or comprising specific components, or where processes are described as having, including or comprising specific process steps, it is contemplated that compositions of the present teachings also consist essentially of, or consist of, the recited components, and that the processes of the present teachings also consist essentially of, or consist of, the recited process steps.

[0047] In the application, where an element or component is said to be included in and/or selected from a list of recited elements or components, it should be understood that the element or component can be any one of the recited elements or components and can be selected from a group consisting of two or more of the recited elements or components. Further, it should be understood that elements and/or features of a composition, an apparatus, or a method described herein can be combined in a variety of ways without departing from the spirit and scope of the present teachings, whether explicit or implicit herein.

[0048] The use of the terms “include,” “includes,” “including,” “have,” “has,” or “having” should be generally understood as open-ended and non-limiting unless specifically stated otherwise.

[0049] The use of the singular herein includes the plural (and vice versa) unless specifically stated otherwise. Moreover, the singular forms “a,” “an,” and “the” include plural forms unless the context clearly dictates otherwise. In addition, where the use of the term “about” is before a quantitative value, the present teachings also include the specific quantitative value itself, unless specifically stated otherwise. As used herein, the term “about” refers to a ±10% variation from the nominal value. All numerical values and ranges disclosed herein are deemed to include “about” before each value.

[0050] It should be understood that the order of steps or order for performing certain actions is immaterial so long as the present teachings remain operable. Moreover, two or more steps or actions may be conducted simultaneously.

[0051] Where a range or list of values is provided, each intervening value between the upper and lower limits of that range or list of values is individually contemplated and is encompassed within the invention as if each value were specifically enumerated herein. In addition, smaller ranges between and including the upper and lower limits of a given range are contemplated and encompassed within the invention. The listing of exemplary values or ranges is not a disclaimer of other values or ranges between and including the upper and lower limits of a given range.

[0052] As disclosed above and herein, the present technology provides a method of predicting an impact of a treatment option for a blood vessel characteristic, the method comprising: receiving, by one or more processors, image data for a blood vessel; receiving, by the one or more processors, measurements of a characteristic of the blood vessel; outputting for display the measured characteristic; receiving a selection of a zone of for implementing a potential treatment option; and predicting, by the one or more processors responsive to receiving the selection, an updated characteristic based on the selected zone.

[0053] More particularly, the technology disclosed above and herein provides computer implemented methods of predicting and displaying the impact of a potential treatment option for a blood vessel on one or more values of a characteristic of the blood vessel. The technology disclosed above and herein may also provide corresponding graphical user interfaces enabling a user to interact with implementations of the method, corresponding computer readable media carrying computer program code arranged to implement such methods and graphical user interfaces when executed on suitable computer systems, and corresponding apparatus or systems arranged to implement such methods and graphical user interfaces for example in the form of suitably programmed computer systems.

[0054] Image data of a blood vessel, which has been acquired for example using intravascular imaging and/or non-invasive imaging, may be used to determine, for example to calculate, one or more values of a characteristic of the blood vessel, such as virtual flow reserve (VFR). Such values may be referred to as actual or current values, and may be determined for a single position or an extended region of the blood vessel, and/or for multiple positions and/or regions along the blood vessel. One or more such actual or current values of the characteristic may then be displayed to a user, for example in numerical form and/or in the form of a graph. If displayed in the form of a graph this may be in correlation with a graphical representation of the blood vessel which may for example be derived using the imaging data and in particular the intravascular imaging data, for example as a longitudinal cross section along the blood vessel. [0055] The technology disclosed above and herein enables the user to input a user selection of a zone of the blood vessel, with the selected zone indicating a spatial extent or position or other spatial characteristic of a potential treatment option such as the position and/or extent and/or type of a stent which may be used to implement such a treatment option. The technology then proceeds to automatically determine or calculate one or more predicted values of the characteristic which would occur if the potential treatment option was implemented as defined by the selected zone.

[0056] The technology then displays to the user the one or more predicted values, or other related predicted outcomes of the potential treatment option, for example in numerical and/or graphical form. The display of these values may be adapted, for example using a change in color or other graphical indicator, depending on whether they meet certain criteria, for example whether a value of one or more such predicted values exceeds or fails to meet a predefined threshold such as a particular level of virtual flow reserve.

[0057] Such technology, and other systems and methods described herein, may perform feature detection and alignment of relative imaging datasets from an intravascular imaging pullback, for example from an imaging pullback used to provide the above intravascular image data. For example, the intravascular imaging pullback may be an OCT or intravascular ultrasound (“IVUS”) pullback. The imaging data sets may be taken at one or more points in time corresponding to different arterial events or treatments. One or more representations of an artery may be displayed based on the imaging data set. The representations may include an indication of identification of virtual flow reserve (VFR). The one or more representations may be displayed to a user.

[0058] Image processing techniques and/or machine learning may compute VFR, for example as the values of the above mentioned characteristic. The frames of the pullback may be stretched and aligned using various windows or bins of alignment features. This image data can be presented using various graphical user interfaces.

[0059] The described technology can provide various workflows and options to facilitate a process of stent planning relative to a blood vessel such as an artery imaged during such a pullback. In general, deployment of shorter stents can result in less metal or other material being introduced in the artery. Using smaller stents can result in less trauma given the torturous nature of the arteries and their movement over time such as during various activities by a recipient of the stent. One or more shorter stents is sometimes desirable because they can be positioned to follow the bends of an artery rather one long stent which may apply stress to the artery when the artery bends or moves.

[0060] In one example, the above mentioned characteristic of the blood vessel, or more generally one or more cardiovascular or vascular system parameters suitable for evaluating potential stent placement, can include without limitation a Virtual Flow Reserve (VFR) values, flow velocity, a pressure value, a maximum flow, a minimum flow one or more fractional flow reserve (FFR) values, coronary flow reserve (CFR) values, coronary flow velocity reserve (CFVR) values, instantaneous flow reserve (IFR) values, one or more index of myocardial resistance (IMR) values and a vascular resistance value, a combination of the foregoing, a weighted average of one or more of the foregoing and another value, and values derived from the foregoing. In some embodiments, virtual flow reserve can also refer to virtual fractional flow reserve (VFR). In general, a VFR value can be determined by using an intravascular imaging probe to generate frames of imaging data that segment the artery through a pullback. [0061] In turn, this imaging data and lumen areas and diameters facilitate a volume-based analysis. Further, by using angiography and other parallel sources of data and coupling them, fluid dynamics, and the frames of imaging data vascular system parameters such as VFR can be used to obtain correlation similar to or better than FFR. These parameters can be used with virtual stents, landing zones, clusteringbased methods and others methods as described herein to perform stenting planning and other and analytic methods.

[0062] The technology disclosed above and herein also provides a method comprising: receiving, by one or more processors, intravascular imaging data of a vessel; determining, by the one or more processors, a lumen contour for at least one frame of the intravascular imaging data; determining, by the one or more processors executing an artificial intelligence model, a lumen mask for the at least one frame of the intravascular imaging data; comparing, by the one or more processors, the determined lumen contour and the lumen mask; determining, by the one or more processors based on the comparison, an overlap value; and when the overlap value for at least one frame of the intravascular imaging data is below a threshold, providing for output, by the one or more processors, an indication that the at least one frame of the intravascular imaging data is a false positive.

[0063] In particular, invention provides computer implemented methods of automatically checking or reviewing a prior classification of a segment of a blood vessel using one or more corresponding frames of intravascular imaging data, for example a classification that the segment is stenotic. The prior classification may have been made automatically, for example on the basis of the same or similar intravascular imaging data, and may apply to a single frame of such data or some other segment for example comprising multiple such frames. The method comprises receiving one or more frames of intravascular imaging data of the segment, for example such data that has been generated using an intravascular ultrasound or intravascular optical device. A lumen algorithm is then used to determine a lumen contour or boundary of the blood vessel within the segment from the one or more frames. The lumen algorithm may be a contour tracking algorithm or similar. A machine learning or Al model or algorithm, which is separate from the lumen algorithm, is then used to determine a lumen mask of the blood vessel within the segment. For example, the lumen mask may comprise pixels or regions each indicative of whether the pixel or region is inside or outside the lumen.

[0064] The lumen contour and the lumen mask then effectively provide two different determinations of the lumen boundary, and significant differences between the two may then be indicative of a likely false positive classification of the segment, for example as a stenotic segment.

[0065] The method therefore proceeds by determining a measure of similarity between the lumens defined by the lumen contour and the lumen mask, such as an overlap value. Based on the measure of similarity, for example, if the similarity fails to meet a predefined threshold, an output indication is provided that the classification of the segment as stenotic is, or is expected to be, a false positive.

[0066] The measure of similarity or overlap value may be a Jaccard index value, which could be calculated for example by taking a ratio of (for example by dividing) a cross sectional amount or area of the segment determined by both the lumen contour and the lumen mask to be within the lumen, and a cross sectional amount or area of the segment determined by either or both of the lumen contour and the lumen mask to be within the lumen.

[0067] The intravascular data may represent cross sectional images through the blood vessel, and the lumen contour and lumen mask may then be a contour and a mask within a corresponding cross section plane through the blood vessel.

[0068] Figure 1 illustrates a data collection system 100 for use in collecting intravascular data, calculating further data from the collected data, and for providing a graphical user interface enabling a user to view and interact with the collected and calculated intravascular data, as further described below. The system may include a data collection probe 104 that can be used to image a blood vessel 102. In some examples, the probe 104 may be an intravascular device, such as an optical coherence tomography (“OCT”) probe, an intravascular ultrasound (“IVUS”) catheter, micro-OCT probe, near infrared spectroscopy (NIRS) sensor, optical frequency domain imaging (“OFDI”), or any other device that can be used to image the blood vessel 102. In some examples, the probe 104 may be a pressure wire, a flow meter, etc. The probe 104 may include a device tip, one or more radiopaque markers, an optical fiber, a torque wire, or the like. Additionally, the device tip may include one or more data collecting subsystems such as an optical beam director, an acoustic beam director, a pressure detector sensor, other transducers or detectors, and combinations of the foregoing.

[0069] A guide wire, not shown, may be used to introduce the probe 104 into the blood vessel 102. The probe 104 may be introduced and pulled back along a length of a blood vessel while collecting data. As the probe 104 is pulled back, or retracted, a plurality of scans or OCT and/or IVUS data sets may be collected. The data sets, or frames of image data, may be used to identify features, such as vessel dimensions and pressure and flow characteristics.

[0070] The probe 104 may be connected to a subsystem 108 via an optical fiber 106. The subsystem 108 may include a light source, such as a laser, an interferometer having a sample arm and a reference arm, various optical paths, a clock generator, photodiodes, and other OCT and/or IVUS components.

[0071] The probe 104 may be connected to an optical receiver 110. According to some examples, the optical receiver 110 may be a balanced photodiode based system. The optical receiver 110 may be configured to receive light collected by the probe 104.

[0072] The data collection system 100 may further include, or be configured to receive data from, a non- invasive imaging system 120. The non-invasive imaging system 120 may be, for example, an imaging system based on angiography, fluoroscopy, x-ray, nuclear magnetic resonance, computer aided tomography, etc. non-invasive imaging system 120 may be configured to nonin vasively image the blood vessel 102. According to some examples, the non-invasive imaging system 120 may obtain one or more images before, during, and/or after a pullback of the data collection probe 104. Non-invasive imaging system 120 may be used to image a patient such that decisions can be made and various possible treatment options such as stent placement can be carried out. These and other imaging systems can be used to image a patient externally or internally to obtain raw data, which can include various types of image data. [0073] The non-invasive imaging system 120 may be in communication with subsystem 108. According to some examples, the non-invasive imaging system 120 may be wirelessly coupled to subsystem 108 via network. For example, the non-invasive imaging system 120 may be wirelessly coupled to subsystem 108 via a communications interface, such as Wi-Fi or Bluetooth. In some examples, the non-invasive imaging system 120 may be in communication with subsystem 108 via a wire, such as an optical fiber. In yet another example, external imaging device 120 may be indirectly communicatively coupled to subsystem 108 or computing device 112. For example, the non-invasive imaging device 120 may be coupled to a separate computing device (not shown) that is in communication with computing device 112. As another example, data from the imaging system 120 may be transferred to the computing device 112 using a computer- readable storage medium, from a storage device via a network, or the like.

[0074] The subsystem 108 comprises a computing device 112. The computing device may include one or more processors 113, memory 114, instructions 115, data 116, and one or more modules 117.

[0075] The one or more processors 113 may be any conventional processors, such as commercially available microprocessors. Alternatively, the one or more processors may be a dedicated device such as an application specific integrated circuit (ASIC) or other hardware-based processor. Although Figure IB functionally illustrates the processor, memory, and other elements of device 110 as being within the same block, it will be understood by those of ordinary skill in the art that the processor, computing device, or memory may actually include multiple processors, computing devices, or memories that may or may not be stored within the same physical housing. Similarly, the memory may be a hard drive or other storage media located in a housing different from that of device 112. Accordingly, references to a processor or computing device will be understood to include references to a collection of processors or computing devices or memories that may or may not operate in parallel.

[0076] Memory 114 may store information that is accessible by the processors, including instructions 115 that may be executed by the processors 113, and data 116. The memory 114 may be a type of memory operative to store information accessible by the processors 113, including a non-transitory computer- readable medium, or other medium that stores data that may be read with the aid of an electronic device, such as a hard-drive, memory card, read-only memory (“ROM”), random access memory (“RAM”), optical disks, as well as other write-capable and read-only memories. The subject matter disclosed herein may include different combinations of the foregoing, whereby different portions of the instructions 101 and data 119 are stored on different types of media.

[0077] Memory 114 may be retrieved, stored or modified by processors 113 in accordance with the instructions 115. For instance, although the present disclosure is not limited by a particular data structure, the data 116 may be stored in computer registers, in a relational database as a table having a plurality of different fields and records, XML documents, or flat files. The data 116 may also be formatted in a computer-readable format such as, but not limited to, binary values, ASCII or Unicode. By further way of example only, the data 116 may be stored as bitmaps comprised of pixels that are stored in compressed or uncompressed, or various image formats (e.g., JPEG), vector-based formats (e.g., SVG) or computer instructions for drawing graphics. Moreover, the data 116 may comprise information sufficient to identify the relevant information, such as numbers, descriptive text, proprietary codes, pointers, references to data stored in other memories (including other network locations) or information that is used by a function to calculate the relevant data.

[0078] The instructions 115 can be any set of instructions to be executed directly, such as machine code, or indirectly, such as scripts, by the processor 113. In that regard, the terms “instructions,” “application,” “steps,” and “programs” can be used interchangeably herein. The instructions can be stored in object code format for direct processing by the processor, or in any other computing device language including scripts or collections of independent source code modules that are interpreted on demand or compiled in advance. Functions, methods and routines of the instructions are explained in more detail below.

[0079] According to some examples, the computing device 112 may receive, either by a wired connection or via a wireless connection, data from the probe 104. The data may include, for example, intravascular data including intravascular imaging data, pressure data, temperature data, flow data, or the like. In some examples, where the probe 104 is an intravascular data collection device, the data received from the device may be used to determine plaque burden, fractional flow reserve (“FFR”) measurements at one or more locations along the vessel, calcium angles, external elastic lamina (“EEL”) detections, calcium detections, proximal frames, distal frames, EEL-based metrics, stent/no stent decisions, scores, recommendations for debulking and other procedures, evidence based recommendations informed by automatic detection of regions/features of interest, stent planning, etc.

[0080] The modules 117 may include various modules configured to implement various functions described in more detail later in this document. Such modules may include a VFR computation module and a display module. In some examples further types of modules may be included, such as modules for computing other vessel characteristics, stent detection modules, etc. According to some examples, the modules may include an image data processing pipeline or component modules thereof. The image processing pipeline may be used to transform collected OCT data into two-dimensional (“2D”) and/or three-dimensional (“3D”) views and/or representations of blood vessels, stents, and/or detected regions.

[0081] The modules 117 may further include a stent placement and prediction module. For example, the computing device 112 may determine an optimal placement for one or more stents, and predict how placement of the one or more stents at the selected locations would impact VFR values as compared to those computed for the vessel as is. By way of example only, the stent placement and prediction module may compare a first predicted impact of placing two smaller stents at neighboring lesions with a second predicted impact of placing one larger stent across both neighboring lesions. The stent placement and prediction module may include machine learning models trained to predict the impact of stent placement, size, or the like with respect to VFR.

[0082] In some examples, the modules 117 may include a detection module. The detection module may be configured to identify one or more frames that were wrongly identified as having a stenosis and correct the lumen area for said frames. In some examples, the false positive stenotic lumen segments may be caused by poor lumen flushing. In another example, the false positive stenotic lumen segments may be caused by other artifacts within the vessel, such as the probe, shadows, etc. [0083] The detection module may identify frames in the pullback that were wrongly identified of classified as having a stenosis. According to some examples, the detection module may identify the false positive stenotic segments using an artificial intelligence (“Al”) model, such as a machine learning model (“ML”). The Al model may be trained to output an indication as to whether a pixel within the image frame is within the lumen or outside the lumen. In some examples, the output of the Al model may be a lumen mask that indicates pixels or regions of the frame that are within the lumen of the vessel and pixels or regions of the frame that are outside the lumen of the vessel. The output of the Al model may be compared to a lumen contour determination from another algorithm or model. A Jaccard index, or Jaccard overlap, may be used to determine an overlap value of the output of the Al model and the lumen contour from the algorithm. The overlap value may, in some examples, be thresholded. Thresholding the overlap values may include, for example, comparing the overlap value to a threshold. When the overlap value is above a threshold, the lumen counter from the Al model and the algorithm may correspond, e.g., substantially match. When the overlap value is below the threshold, the lumen contour from the Al model and the algorithm may differ such that the frame may be identified as a false positive. The threshold can be a predetermined value or range of values. For example, the threshold could be 0.9025, a range of 0.9-0.91, or the like. In some examples, the threshold can be a percentage, such as 90%, or a range of percentages, such as 90-91%. While a value of 0.0925, a range of 0.9-0.91, a percentage of 90%, and a range of 90-91% are provided as possible threshold values, they are just some examples and are not intended to be limiting. For example, the threshold can be greater than 0.91, such as 0.94, less than 0.91, such as 0.87, etc.

[0084] In some examples, frames that are identified as being false positives, e.g., frames that were wrongly identified as being stenotic, may be color coded. In some examples, the identified frames maybe colored red, green, blue, etc., to draw a physician’s attention to the frames. In some examples, an indication or warning side may be provided at or near the identified frames.

[0085] In some examples, the modules 117 may include a lumen bridge module. The lumen area of the frames may be corrected by the lumen bridge by applying linear interpolation. In some examples, the user, or physician, may select the frames for the lumen bridge to correct. In another examples, the frames may be automatically selected such that the lumen bridge module automatically corrects the lumen contour of the identified frames.

[0086] According to some examples, the modules may additionally or alternatively include a video processing software module, a preprocessing software module, an image file size reduction software module, a catheter removal software module, a shadow removal software module, a vessel enhancement software module, a blob enhancement software module, a Laplacian of Gaussian filter or transform software module, a guide wire detection software module, an anatomic feature detection software module, stationary marker detection software module, a background subtraction module, a Frangi vesselness software module, an image intensity sampling module, a moving marker software detection module, iterative centerline testing software module, a morphological close operation software module, a feature tracking software module, a catheter detection software module, a bottom hat filter software module, a path detection software module, a Dijkstra software module, a Viterbi software module, fast marching method based software modules, a vessel centerline generation software module, a vessel centerline tracking module software module, a Hessian software module, an intensity sampling software module, a superposition of image intensity software module and other suitable software modules as described herein. According to some examples, the modules may include software such as preprocessing software, transforms, matrices, and other software-based components that are used to process image data or respond to patient triggers to facilitate co-registration of different types of image data by other software-based components or to otherwise perform annotation of image data to generate ground truths and other software, modules, and functions suitable for implementing various features of the disclosure. The modules can include lumen detection using a scan line based or image based approach, stent detection using a scan line based or image based approach, indicator generation, apposition bar generation for stent planning, guide wire shadow indicator to prevent confusion with dissention, side branches and missing data, and others.

[0087] In some examples, the modules may be configured to process the vessel data obtained by the probe 104 and/or imaging system 120 using machine learning algorithms, artificial intelligence, or the like.

[0088] The subsystem 108 may include a display 118 for outputting content to a user. As shown, the display 118 is separate from computing device 112 however, according to some examples, display 118 may be part of the computing device 112. The display 118 may output image data relating to one or more features detected in the blood vessel. For example, the output may include, without limitation, cross- sectional scan data, longitudinal scans, diameter graphs, image masks, etc. The output may further include lesions and visual indicators of vessel characteristics or lesion characteristics, such as computed pressure values, vessel size and shape, or the like. The output may further include visual indicia for candidate stent placement, such as an overlay highlighting selected vessel regions for potential stent placement. The display 118 may identify features with text, arrows, color coding, highlighting, contour lines, or other suitable human or machine readable indicia.

[0089] According to some examples the display 118 may be used to present a graphic user interface (“GUI”) to a user so that a user may interact with the computing device 112 and thereby cause particular content to be output on the display 118, typically using forms of input such as a mouse, keyboard, trackpad, microphone, gesture sensors, or any other type of user input device. One or more steps may be performed automatically or without user input to navigate images, input information, select and/or interact with an input, etc. The display 118 and input device, along with computing device 112, may allow for transition between different stages in a workflow, different viewing modes, etc. For example, the user may select a segment of vessel for analysis of VFR, enter data or commands in response to prompts when transitioning through different phases of a workflow, adjust candidate stent placements to generate an updated VFR computation, etc.

[0090] Using the data collection system 100 of Figure 1 , the predicted impact of placement of one or more stents of particular sizes at particular locations may be evaluated using VFR or another characteristic. According to some examples, the predicted impact may be computed or estimated using a trained Al model, typically executing on the computing device 112, such as a machine learning (ML) model trained on training examples. Each training example may be a case from a clinical trial and/or from the field. The ML model may compare pre-PCI (percutaneous cardiac intervention) information and post-PCI outcome for each case. The ML model estimate of predicted impact may be used to provide a physician or end-user a quantitative assessment of the stent placement impact on VFR.

[0091] In some examples, information input by the user using the GUI may comprise annotations. For example, the system may be configured to receive annotations to one or more representations of the vessel displayed using the GUI. The annotations may be, in some examples, an indication of plaque burden, fractional flow reserve (“FFR”) measurements at one or more locations along the vessel, calcium angles, EEL detections, calcium detections, proximal frames, distal frames, EEL-based metrics, stent/no stent decisions, scores, recommendations for debulking and other procedures, evidence based recommendations informed by automatic detection of regions/features of interest, stent planning, etc. In some examples, the annotations may be a treatment device landing zone, balloon device zone, vessel prep device zone, or lesion related zone. For example, the system may receive a user input corresponding to a proximal and distal location along the vessel corresponding to a proximal and distal location of a treatment device landing zone, balloon device zone, vessel prep device zone, or lesion related zone. According to some examples, the system may receive an input corresponding to a proximal and distal location along the vessel selecting frames for the detection module. The lumen area of the selected frames may be corrected by the detection module by applying linear interpolation.

[0092] According to some examples, annotations may also or instead be automatically determined by the data collection system 100. For example, the system may, based on vessel data, determine one or more of plaque burden, FFR measurements at one or more locations along the vessel, calcium angles, EEL detections, calcium detections, proximal frames, distal frames, EEL-based metrics, stent/no stent decisions, scores, recommendations for debulking and other procedures, evidence based recommendations informed by automatic detection of regions/features of interest, stent planning, a treatment device landing zone, balloon device landing zone, vessel prep device zone, lesion related zone, etc. The system may automatically provide the plaque burden, FFR measurements at one or more locations along the vessel, calcium angles, EEL detections, calcium detections, proximal frames, distal frames, EEL- based metrics, stent/no stent decisions, scores, recommendations for debulking and other procedures, evidence based recommendations informed by automatic detection of regions/features of interest, stent planning, a treatment device landing zone, balloon device landing zone, vessel prep device zone, lesion related zone, etc. for output as one or more annotations on at least one of the vessel representations.

[0093] The display 118 alone or in combination with computing device 112 may allow for toggling between one or more viewing modes in response to user inputs. For example, a user may be able to toggle between different intravascular data, images, etc. recorded during each of a plurality of pullbacks of the probe 104 within the blood vessel 102. In some examples, the user may be able to toggle between different representations, such as a longitudinal representation, a cross-sectional representation, a three-dimensional representation, intravascular images, color images, black and white images, live images, or the like.

[0094] In some examples, the display 118, alone or in combination with computing device 112, may present one or more menus to the user (such as a physician), and the user may provide input in response by selecting an item from the one or more menus. For example, the menu may allow the user to show or hide various features. As another example, there may be a menu for selecting blood vessel features to display. According to some examples, the display may output a menu including one or more inputs for analyzing and/or processing the information associated with the vessel data. For example, the inputs may include adding and/or removing a side branch, recalculating the VFR, measuring selected regions of interest, correcting false positives (e.g., lumen bridge module), or the like.

[0095] The content output on the display 118 may include one or more representations of the vessel 102. For example, the representations may include images data, longitudinal representations, three-dimensional representations, live representations, or the like. The longitudinal representation may include, for example, a representation of the blood vessel based on the lumen diameter that is symmetrical about the longest axis of the representation. In some examples, the representations may include graphical representations, such as a graphical representation of VFR, pressure values, flow values, or the like.

[0096] In some examples, the one or more visual representations of the images may include an indication of a lesion location, lesion severity, lesion length, or the like. Additionally or alternatively, the indication of the lesion may be color coded, where each color represents the severity, length, or other measurement related to the lesion.

[0097] In some examples, the content output on the display 118 may include candidate treatment zones. Such a candidate treatment zone may be, for example, a candidate stent landing zone. For example, the output may include an indication corresponding to a candidate proximal landing zone for a stent and a candidate distal landing zone for a stent. The candidate treatment zone may be determined based on the determined plaque burden, lesion locations, lesion length, or the like. The indications may be provided on any of the vessel representations, e.g., the three-dimensional representation, the longitudinal representation, the graphical representation, image data such as the external images, etc.

[0098] According to some examples, the display 118 and/or computing device 112 may be configured to receive one or more inputs from a user corresponding to a selection made on one or more representations such as representations of the blood vessel. For example, an input may be received from the user corresponding to a selection of an image frame on a longitudinal representation of the blood vessel. In response, other representations being output may be updated to display a corresponding indication or image frame. For example, a displayed extraluminal image may be updated to have an indication along the vessel corresponding to the location of the image frame selected in the longitudinal representation, a circumferential indication may be provided on a three-dimensional representation corresponding to the location of the image frame selected in the longitudinal representation, the cross-sectional image frame may be updated to correspond to the image frame selected in the longitudinal representation, etc. In some examples, the vessel data associated with the selected location may be updated and provided for display.

[0099] Figure 2 illustrates an example system 200 that uses data from past percutaneous cardiac interventions (PCIs) to determine stent size and placement and predict the impact thereof for future patients. Aspects of this system 200 may be used to provide such determinations and predictions within the data collection system 100 of figure 1. The system 200 may include a development environment 552 and a catheter lab 560. While shown as a catheter lab 560, the catheter lab 560 may be any location in which a physician inserts or implants a stent into a patient. For example, the catheter lab may be at a hospital, an outpatient surgical location, etc. Thus, identifying the location as catheter lab 560 is merely one example and is not intended to be limiting. However, in the context of figure 1 the illustrated functional elements of the catheter lab may be implemented within the computing device 112 of the subsystem 108, even though the development environment 552 may be implemented elsewhere.

[0100] The development center 552 may include a training database 5545, a machine learning system 556, and a trained predictive model 558A. The training database may contain PCI information at multiple levels. For example, the training database 554 may include coarse statistics from published clinical studies, records and imagery on individual PCIs from clinical trials, and data on PCIs collected in the field. These data may be in the form of input-output pairs, where the input for a case is all the information observable before the target vessel is prepared and stent deployed, and the output is the resulting stent expansion and other outcomes (complications, re -hospitalization, TVR, etc.). The input-output pairs may be one or more image frames. According to some examples, the input may be a plurality of images of the target vessel before the target vessel is prepared and the output may be a plurality of images of the target vessel after stent expansion, etc. The input plurality of images may correspond to the output plurality of images such that a first frame of the input plurality of images is from the same location within the target vessel as the first frame of the output plurality of images.

[0101] The machine learning system 556 may learn or model the relationship between these inputs and outputs. For example, the machine learning system 556 may detect different values from each of the plurality of input and output images. The values may include, but are not limited to, the VFR, vessel size, lesion size, stent size, stent numbers, the percentage of stent expansion, etc. for each of the plurality of input and output images. Each of these values may be used to later predict the impact of particular stent placement within the catheter lab 560. According to some examples, the machine learning system 556 may learn the relationship by adjusting internal parameters to minimize error in its output predictions.

[0102] According to some examples, the machine learning system may use a linear model such as a logistic regression to adjust internal parameters that are multiplicative weights placed on each predictor attribute. In some examples, the machine learning system may learn any number of decision trees such that its internal parameters may be the rules governing each tree. In some examples, when the machine learning system 556 adjust model parameters to minimize prediction error, the machine learning system 556 may re-run 557 the data to create an additional model.

[0103] The models created by the machine learning system 556 may comprise one or more trained predictive models 558. Such trained predictive models 558 may be configured to predict one or more values of a VFR or another characteristic that would be obtained if a stent of a particular size were placed at a particular lesion in a blood vessel. The trained predictive model 558 may be provided or sent to the catheter lab 560, for example to the computing device 112 of the subsystem 108 of figure 1 when implemented in such a catheter lab 560. According to some examples, the trained predictive model 558 may be shared via a network. In the catheter lab 560, the trained predictive model 558 may use and/or take information about a new target lesion 562, for example obtained using the data collection system 100, and generate VFR predictions 564 to support a user such as physician in refining their intervention strategy for example using the data collection system 100.

[0104] Figure 3 illustrates an example display 300 of a GUI which may be implemented by display 118 of the data collection system 100. On the GUI display 300, a dynamically adjustable calculation zone 332 lets a user such as a physician see the projected flow recovery if a stent is used in a specified region in a blood vessel 104. The projected flow recovery may be calculated for example using the trained predictive model 558 of figure 2 when implemented within the computing device 112 of figure 1. According to the example shown, the GUI screen 300 includes an angiography view 310, an information display 320, a longitudinal representation 330 of the blood vessel shown in angiography view 310, and a graphical display 340 corresponding to the information display 320. The angiography view may display an angiogram received from the non-invasive imaging system 120, while the longitudinal representation may be derived using data from the probe 104 and the graphical display and data display may show information calculated using appropriate modules 117 of the computing device 112.

[0105] A calculation zone 332 is shown as a colored overlay on top of the longitudinal representation 330. The extent of the region of the blood vessel included in the calculation zone 332 may be modified by a user. By way of example, the calculation zone 332 may have a selectable boundary 336, wherein the user can manipulate the boundary to position it at a different location along the vessel. The selectable boundary may be manipulated by selecting and dragging the boundary, by selecting the boundary and then selecting a location to which the boundary should be moved, or by any of a variety of other interactions with the GUI.

[0106] The calculation zone 332 shown in the longitudinal representation 330 may correspond to a region of interest 312 in the angiography view 310. For example, as the calculation zone 332 is manipulated in the longitudinal representation 330 to change its boundary, the region of interest 312 is automatically updated to correspondingly change its boundaries. Thus, for example, shrinking or enlarging or moving the calculation zone 332 in the longitudinal representation 330 will shrink or enlarge or move the region of interest 312 shown in the angiography view 310. The region of interest 312 may be represented using a colored overlay, similar to the representation of the calculation zone 332.

[0107] The longitudinal representation 330 and the angiography view 310 may include additional indications that correspond to one another. For example, frame marker 334 may represent a position along the vessel, corresponding to angiography frame marker 314.

[0108] Vessel characteristics may be computed based on the calculation zone 332, for example by suitable modules 117 of the computing device 112. For example, such characteristics may include a determined or calculated actual VFR of the vessel, and a predicted VFR that would be obtained if a treatment option was implemented in the vessel at a location corresponding to the calculation zone 332. Such treatment options may include placement of a stent or other options. The calculated actual and predicted VFRs or other characteristics may be determined for example using a trained predictive model 558 as illustrated in, and discussed above in respect of, figure 2. [0109] The computed and predicted vessel characteristics may be displayed in any of a variety of formats. For example, as shown in Figure 3, information display 320 includes a numerical representation of the determined actual VFR 325 and a numerical representation of the predicted VFR 328. Similarly, graphical display 340 includes a graphical representation of the actual VFR 345 and a graphical representation of the predicted VFR 348. While the actual VFR values, however represented, should not change, the predicted VFR values may be dynamically updated as the calculation zone 332 is manipulated. In this regard, a user such as a physician can see in real time how treatment options will impact VFR. For example, the user can manipulate a size and placement of the calculation zone 332 to see the corresponding predicted VFR values, thereby identifying how deploying stents of different sizes and/or at different locations would impact VFR. [0110] Figure 4 provides a further example display 400 of the GUI of figure 3, showing how multiple stented regions can be re-sized to calculate the cumulative effect on the vessel of interest. For example, zones 431 and 432 can be selected by a user. The zones 431, 432 may be selected based on, for example, narrowings shown in longitudinal representation 430 of the vessel and/or vessel characteristics shown in graphical view 440. For example, as shown in Figure 4, graphical view 440 includes a graphical representation of determined actual VFR 445. The graphical representation 445 shows a first drop 441 in VFR shortly after the 10 mm demarcation in the longitudinal representation, and a second drop 442 in VFR around the 30 mm demarcation. The user may define the zones 431 , 432 based on the start and ends of these drops 441, 442 in VFR. According to some examples, the computing device 112 may suggest these zones 431, 432 based on the actual VFR or other vessel characteristics. In the example shown, alignment indicators 452 are depicted to show a correlation between the graphical view 440 and the zones 431 , 432 in the longitudinal representation 430.

[0111] The zones 431 , 432 as defined in the graphical view 440 and/or longitudinal representation 430 may also be depicted in angiography view 410. For example, as shown in Figure 4, these zones 431, 432 correspond to zones 411, 412, respectively. As zones are defined and/or updated in one view, they may be dynamically updated in corresponding views as well.

[0112] Figure 5 provides a further example display of the GUI of figures 3 and 4, showing a different way in which the two treatment zones defined in Figure 4 may be represented. For example, Figure 5 shows user adjustable calculation zones 531, 532. The predicted VFR values, shown as numerical representation 528 in information display 520, may be computed based on both calculation zones 531, 532. For example, the predicted VFR value 528 indicates a predicted impact of treatment at both calculation zones 531, 532. [0113] Information in the GUI displays of figures 3 to 5 may indicate to the user whether treatment at the defined treatment areas of the blood vessel 104 would result in a sufficient improvement. For example, the computing device 112 may determine whether the estimated VFR would meet or exceed a predetermined threshold. The predetermined threshold may be set by the user, such as based on patient risk or other factors, or it may be set based on industry standard or other information. When the estimated VFR value is below the threshold, an indication in the GUI may alert the user. For example, as shown in Figure 5, the text of the numerical VFR representation 528 changes color, from white to orange, to alert the user. It should be understood that any of a variety of other indicia are possible, such as the appearance of icons, flashing the screen, different colors, audible alerts, etc.

[0114] Figure 6 illustrates a further example display 600 of the GUI comprising an enlarged angiography view 610 that fills most or all of the screen, for example of display 118. A pressure drop graph 660 is aligned with the vessel of interest, which would help the user plan possible stent locations on the angiography view 610. For example, the pressure drop graph 660 is illustrated as a colored bar having a vertical length that corresponds to a length of the vessel of interest. The colored bar shows two darkened regions 661, 662 where there is a measured drop in pressure in the vessel. As shown, the pressure drop graph 660 further includes numerical indications corresponding to each drop in pressure.

[0115] The angiography view 610 includes markers 571, 572 corresponding to a location at which each drop in pressure occurs. The markers 571, 572 may be inserted by the user or automatically generated by the computing device.

[0116] Figure 7 illustrates a further example display the GUI of figures 3 to 6 which presents to the user pressure loss in the blood vessel 104 at discrete locations. For example, as shown, the angiography view includes discrete markers 711-714 at each location of pressure drop. The markers 711-714 may be, for example, colored overlays. The colored overlays may be any of a variety of shapes or colors. As shown, the size of the overlay may vary in correlation to a degree of pressure drop. For example, markers 711 and 714 are larger as compared to markers 712, 713, thereby indicating a relatively larger drop in pressure.

[0117] The graphical representation may also include markers 741-744 corresponding to the angiography markers 711-714. For example, the graphical display markers 741-744 may identify each portion of the graphical representation of the actual VFR where drop in pressure exists or exceeds a predetermined threshold. The markers 741-744 are shown as encircling the locations where the pressure drop occurs, with a numerical value indicating the amount of pressure drop at each location. However, it should be understood that other representations are possible. The graphical markers 741-744 are also shown as correlated with longitudinal representation 730 of the vessel. For example, alignment marks 751-754 indicate how each location of pressure drop along the graphical representation aligns with the longitudinal representation.

[0118] Such indications of lost pressure at each discrete lesion enable the user to determine the most effective approach for treating the vessel. For example, the indications may inform the user as to the exact locations of pressure drop and the degree of pressure drop at each discrete location. The user may consider the number and degree of pressure drops, the distance between discrete locations, and other information to determine the best treatment option. The user may then simulate the determined treatment option to see the predicted VFR for that treatment option, as discussed above.

[0119] While the foregoing description provides several examples of the information that may be displayed in GUI, it should be understood that additional displays, views, or information may also be included in the GUI, or that some information may be replaced with other information or omitted. By way of example only, the GUI may additionally or alternatively present a comparison view to allow the user to visually compare the predicted impact of different stent options. [0120] Figure 8 illustrates an example method of outputting a representation of a blood vessel which may be implemented within the data collection system 100 of figure 1, for example using the computing device 112, and using the GUI presented on the display 118 as illustrated in figures 3 to 7. The following operations do not have to be performed in the precise order described below. Rather, various operations can be handled in a different order or simultaneously, and operations may be added or omitted.

[0121] In block 810 one or more processors receive one or more frames including image data of a blood vessel segment, for example from the probe 104 and/or the non-invasive imaging system 120 of figure 1. The image data may be used, for example, to display a representation of the vessel, such as an angiographic image. The frames may be obtained during one or more imaging pullbacks. For example, the pullbacks may be taken pre-treatment, post-treatment, pre-stenting, post-stenting, pre-atherectomy, postatherectomy, pre-angioplasty, post-angioplasty, post-optimization, etc. According to some examples, the pullbacks may be taken after stenting and/or after a physician has further ballooned the stent with various balloon diameters and pressures.

[0122] In block 820, one or more values of a characteristic of the vessel are measured or determined, for example by calculation. The characteristic may be, for example, pressure, flow, lumen diameter, and/or other characteristics. The characteristic may be measured or determined using techniques such as VFR or other measurement techniques, for example using image data received in block 810. The measurements or determinations may be made at multiple points along the blood vessel. These measured or determined values of the characteristic may be referred to as current or actual values of the characteristic, to distinguish them from predicted values as discussed below.

[0123] In block 830, the measured or determined values of the characteristic are displayed in correlation with the multiple points along the blood vessel. By way of example, measured pressure may be displayed by a graph, gradient, numerical values, or any of a variety of other indicia. According to some examples, the display of the measured characteristic may be responsive to interaction with portions of a representation of the blood vessel on the display. For example, the user may click on a portion of the depicted blood vessel to trigger depiction of the measured characteristic at that portion.

[0124] In block 840, selection of a zone is received, typically through selection by a user of the GUI. The zone may be selected for implementation of a potential treatment option, such as placement of a stent. Selection of the zone may include manipulation of boundaries of a colored overlay defining the zone to make the zone larger or smaller, and/or to adjust a positioning of the zone. According to one example, the zone may be adjusted relative to one representation of the vessel and in response another representation of the blood vessel may be automatically updated. For example, as boundaries of the colored overlay are adjusted with respect to a longitudinal representation of the vessel, boundaries of a corresponding overlay in an angiographic view may automatically be updated. According to some examples, an initial zone may be suggested by the computing device, wherein the suggested initial zone may be modified by the user. According to further examples, selecting the zone may include selecting multiple zones, for example, corresponding to multiple candidate treatment sites. [0125] In block 850, updated values of the characteristic are predicted based on the selected zone. For example, one or more predicted VFR values may be calculated to predict how the VFR for the vessel is expected to change if a stent having a size and location corresponding to the selected zone deployed in the blood vessel. According to some examples, the updated characteristic may be displayed in conjunction with the measured characteristic, such as adjacent. According to further examples, other indicia may be provided to indicate whether the updated characteristic meets a predetermined threshold. For example, the characteristic may be displayed in a particular color, font, size, location, etc., or other indicia may be provided such as icons, text displays, audible alerts, or the like.

[0126] Figure 9A illustrates a further interface aspect which may be implemented within the GUI of the data collection system of Figure 1 , in this case as example GUI screen 900A, and which may be used in combination with aspects of the GUI shown in figures 3 to 7 and 9B. This interface aspect provides an action tray 901. The action tray 901, or menu, may be a pop-up, overlay, drop-down, etc., and in Figure 9 A, the action tray 901 is an overlay on representations of the blood vessel. The action tray 901 includes one or more inputs or controls of the GUI for a user to instruct the system to carry out analysis and/or processing of the information associated with the vessel data. For example, the controls may include one or more of a recalibrate control 903, a switch to guided mode control 905, a compare pullbacks control 907, a VFR control 909, a 3D view control 911, a redo coregistration control 913, a pullback notes control

915, an instructions for use control 917, a case setting control 919, an anonymize screen control 921 (for hiding personally identifiable information), a screen capture control 923 (still image or video), and a reset pullback control 925. In some examples, the inputs or controls may cause the system to carry out any of adding and/or removing a side branch, recalculating the VFR, measuring selected regions of interest, correcting false positives (e.g., lumen bridge module), or the like.

[0127] Figure 9B illustrates a further example screen 900B of the GUI which may be implemented in the data collection system of figure 1, in which classifications of one or more frames in the longitudinal representation 914 have been identified as incorrect, or more particularly as false positives. False positives may be, for example, image frames captured during a pullback of probe 104 that have been wrongly classified as being stenotic, for example by analysis carried out automatically by the computing device 112. The GUI screen 900B may include, for example, an extraluminal image 910, an intraluminal image

916, a graphical representation 912, and the longitudinal representation 914.

[0128] The extraluminal image 910 may include one or more indications. The indications may identify a region of interest 906c, a current location of probe 104, an indication of the location corresponding to the selected frame 904a on the longitudinal representation 914, or the like. The region of interest 906c may be defined by a proximal and distal point. The region of interest 906c identified on the extraluminal image 910 may correspond to the region of interest 906a, 906b identified on the longitudinal representation 914 and graphical representation 912, respectively. According to some examples, the region of interest 906a-c (collectively “region of interest 906”) may correspond to a stent zone, treatment zone, or the like.

[0129] The GUI may be arranged to update the region of interest 906a-c based on changes made to the geometry of the vessel for the purposes of calculating a characteristic of the blood vessel such as VFR. For example, the system may be receive one or more user instructions using the GUI, or other inputs, to change the geometry of the vessel for such purposes. As an example, referring to Figure 11B, the system may receive a user instruction to add and/or remove a side branch 1109 for the purposes of calculating a characteristic such as VFR. Based on one or more subsequent user inputs received by the system, the geometry of the vessel may be changed for such purposes by adding and/or removing a side branch(es). The system may then determine, based on the updated geometry of the vessel, that the region of interest 906a-c should also be updated. The system may provide for output an indication of the updated region of interest. In some examples, in addition to the indication of the updated region of interest, the system may provide for output one or more inputs for accepting the updated region of interest or rejecting the updated region of interest. Accepting the updated region of interest may, in some examples, cause the indications identifying the region of interest 906a-c to disappear such that only the indication of the updated region of interest is provided for output. Rejecting the updated region of interest may cause the indications identifying the updated region of interest to disappear such that only the indication identifying the region of interest 906a-c remains.

[0130] The graphical representation 912 may be a graphical representation of the VFR values determined for the pullback. The graphical representation 912 may include a baseline indication 920 and a maximum indication 918. The baseline indication 920 may be the VFR value without intervention, e.g., percutaneous coronary intervention (“PCI”), stenting, or the like. The maximum indication 918 may be, for example, the maximum VFR value that can be obtained with ideal intervention, e.g., maximum stent expansion. The graphical representation 912 may include an indication 904b of the VFR value for the selected frame 904. As the selected frame 904a changes along the longitudinal representation 914, the indication 904b on the graphical representation 912 may be updated to correspond to the selected frame 904a. In some examples, an input may be received on and/or within the graphical representation 912 to update or change the location of indication 904b. In such an example, the location of the selected frame 904a on the longitudinal representation 914 and/or an indication on the extraluminal image 910 may be updated to correspond to the selection on the graphical representation 912.

[0131] According to some examples, the graphical representation may be a representation of the pressure values. For example, the graphical representation may be a pressure drop curve, such as those shown in Figures 14-16.

[0132] The intraluminal image 916 may correspond to the image taken during the pullback at the location selected on the extraluminal image 910, graphical representation 912, and/or longitudinal representation 914. For example, intraluminal image 916 may correspond to image of selected frame 904a-b (collectively “selected frame 904”). As the selected frame 904 changes based on received inputs, the intraluminal image 916 may update to correspond to the selected frame 904. According to some examples, the intraluminal image 916 may include one or more indications corresponding to an amount of calcium, the EEL, or other relevant information. For example, as shown in screen 1700B in Figure 17B, the intraluminal image 916 may include a calcium arc 1708. The calcium arc may be an arc concentric to the center of the intraluminal image 916. The calcium arc 1708 may extend around the intraluminal image 1708 a number of degrees corresponding to the detected calcium. An indication 1714 of the calcium may also be provided for output on the longitudinal representation 914. In some examples, the intraluminal image 916 may include an indication of the lumen 1710, the EEL 1712, or the like.

[0133] The longitudinal representation 914 may be a representation of the vessel generated based on the diameter values of the vessel. In some examples, the longitudinal representation 914 may be symmetrical about the longest axis of the representation. As shown, the longitudinal representation 914 extends horizontally across the GUI screen 900. However, in some examples, the longitudinal representation 914 may extend vertically on the GUI screen 900.

[0134] The longitudinal representation 914 may include an indication 902 of one or more frames or segments of the depicted blood vessel that have been wrongly classified, for example as stenotic, for example automatically by operation of the computing device 112, or optionally manually for example by manual input by a user of the GUI. These wrongly classified frames or segments may be referred to as false positives. The indication 902 may be, for example, a color coded indication thereby differentiating the indicated frames from the remaining frames in the longitudinal representation 914.

[0135] The false positives may be identified using a detection system, such as the detection system 1000 shown in Figure 10, which may be implemented at least in part by the computing device 112. Figure 10 depicts a block diagram of an example detection system 1000. The detection system 1000 may be configured to identify frames that were wrongly classified as being stenotic. The detection system 1000 may be executed in two stages, including a lumen algorithm execution stage 1002 which may be implemented using a lumen algorithm, and a modeling stage 1004 which may be implemented using an artificial intelligence model which is distinct from and/or separate to the lumen algorithm. The stages can be performed in sequence or at least partially in parallel.

[0136] The lumen algorithm execution stage 1002 may be configured to identify the lumen contour of the lumen within one or more frames of intravascular data using the lumen algorithm, which may for example be a contour tracking algorithm. In this way, the contour of the lumen may be automatically identified. The contour of the lumen may be identified by segmenting the tissue from the intravascular image. The lumen contour is then reconstructed by tracking the proximal border of the detected tissue and fitting a smooth curve. In this example, proximal refers to the region closest to the intravascular catheter. Neighboring frames of intravascular image data can be used to correct low-quality lumen contours.

[0137] The modeling stage 1004 may be configured to predict lumen regions in intravascular images. The modeling stage 1004 may, in some examples, be an Al model. The modeling stage 1004 may be configured to receive inference data 1006 and/or training data 1008 for use in predicting the lumen regions, e.g., regions of the intravascular image that are within the lumen and regions of the intravascular image that are outside of the lumen. The modeling stage 1004 may receive at least a portion of the subsequent image frames as input and provide, as output, probability maps for the portion of image frames. The probability map indicates the probability of the pixels being within the lumen or outside the lumen of the vessel. According to some examples, the higher the probability the more likely the pixel is from the lumen. The probability may, in some examples, be compared to a threshold to determine whether the pixel is within the lumen or outside the lumen of the vessel. The threshold may be a predetermined value, such as a value between 0-1. In some examples, the threshold may be 0.5, 0.6, etc. In some examples, the threshold may be a percentage, such as between 0-100%. In such an example, the threshold may be 50%, 60%, etc. While examples of between 0-1, 0.5, 0.6, 1-100%, 50%, and 60% are provided herein, they are just some examples of what the threshold could be and, therefore, are not intended to be limiting. The probability masks in various forms may be referred to as lumen masks. Each such lumen mask may then comprise pixels or regions which have values indicative of whether the pixel is inside or outside the lumen. The pixels of the lumen masks can, but need not, correspond directly to the pixels of the intravascular image frames.

[0138] According to some examples, various post-processing techniques may be applied to the segmented lumen mask. The post-processing techniques may include, for example, cross-frame processing. Crossframe processing can improve the spatial consistency between lumen masks.

[0139] The inference data 1006 can include data associated with identifying lumen regions within an intravascular image. The inference data 1006 may include, for example, intraluminal images of a patient, extraluminal images of the patient, other health related factors associated with the patient, or the like.

[0140] The training data 1008 can correspond to an Al learning task for identifying lumen areas. The training data 1008 may include consecutive intravascular images, e.g., consecutive image frames captured during a pullback of probe 104. According to some examples, the intravascular images may be stacked. In some examples, the training data 1008 may be three-dimensional (“3D”) data. The 3D data may be three- dimensional images. In some examples, the 3D images are generated based on intravascular imaging data, such as intravascular images. For example, intravascular images captured by probe 104 during a pullback may be used to generated 3D images. In some examples, the intravascular images captured during the pullback may be chunked, or grouped, into sections of frames. The 3D images may be generated based on the sections of frames.

[0141] The training data 1008 may include, in some examples, ground truth data from the lumen. The ground truth data from the lumen may include, for example, intravascular images in which the identified lumen contour has been confirmed to correspond to the lumen contour in the intravascular image.

[0142] The modeling stage 1004 may therefore be configured to output a prediction of pixels or regions of the intravascular image that are within the lumen of the vessel and regions of the intravascular image that are outside the lumen of the vessel. For example, the modeling stage 1004 may, based on the inference data 1006 and/or training data 1008, generate a lumen mask. The lumen mask may identify, or differentiate, the regions of the intravascular frame that are within the lumen of the vessel and the regions of the intravascular frame that are outside of the lumen of the vessel.

[0143] The detection system 1000 may compare the output of the lumen algorithm execution stage 1002 and the output of the modeling stage 1004. For example, the detection system 1000 may compare the identified lumen contour by the lumen algorithm for intravascular frame “n” to the lumen mask output by the modeling stage 1004. The comparison may provide an indication as to the degree, or amount, of agreement between the identified lumen contour and/or region. In some examples, the detection system 1000 may use a Jaccard index, or overlap, to determine the degree, or amount, of agreement between the output of the lumen algorithm execution stage 1002 and the output of the modeling stage 1004. The amount of this agreement may be referred to as measure of similarity. In some examples, the amount of agreement or measure of similarity may correspond to an amount of overlap between the lumen contour and the lumen mask. In such an example, both the lumen contour and the lumen mask may be converted such that pixels from the interior of the lumen contained positive detections and from the exterior contained negative detections. The overlap value may, in some examples, be compared to a threshold. Overlap values above the threshold may indicate that there is a high degree of consistency between the lumen contour identified by the lumen algorithm execution stage 1002 and lumen mask output the modeling stage 1004. Overlap values below the threshold may indicate that there is a low degree of consistency between the lumen contour identified by the lumen algorithm execution stage 1002 and lumen mask output from the modeling stage 1004. In some examples, when the overlap value is below the threshold, the frames may be identified as false positives. The output 1010 of the detection system may be an indication on the longitudinal representation 914 that the frames are false positives. If the measure of similarity or overlap is a Jaccard index value, then this may be calculated by dividing a cross sectional amount or area of the segment determined by both the lumen contour and the lumen mask to be within the lumen by a cross sectional amount or area of the segment determined by either or both of the lumen contour and the lumen mask to be within the lumen. In this way, if both the lumen contour and lumen mask agreed exactly as to the boundary of the lumen, the Jaccard index value would be 1.0, and if there was no overlap the index value would be zero.

[0144] Figure 11 A illustrates an example screen 1100A of the GUI which may be implemented within the data collection system of Figure 1 , in which only a portion of the longitudinal representation is selected. As shown in Figure 1 IB, GUI screen 1100B may include one or more menus 1101, 1113 including one or more controls or inputs for a user to instruct analysis and/or processing of the information associated with the vessel data. For example, the controls or inputs may include controls to instruct the system to add and/or remove a side branch 1109 for example for the purposes of calculating a characteristic of the blood vessel such as VFR, to actually recalculate the VFR 1107, to measure selected regions of interest 1119, to correct false positives (e.g., lumen bridge module) 1111, or the like. The controls or inputs may also instruct the system to include adjust exclusion zones 1103, add/delete bookmark 1105, hide/show augmentations 1115, zoom and pan image 1117, 3D view 1121, and edit lumen contour 1123. For example, based on the identification of frames automatically identified as wrongly classified as being stenotic, a portion of the longitudinal representation 914 may be excluded from consideration when determining VFR, FFR, or other information related to the vessel.

[0145] In some examples, the portion 1102 of the longitudinal representation 914 to be considered may be automatically identified. For example, the system may determine that a grouping of frames 902a identified as false positives at the start or end of a pullback should be excluded. In such an example, the frames at the start or end of the pullback may not be properly flushed, may include other artifacts, or the like, thereby causes the false positives. In such an example, when there is a threshold number of frames at the start or end of the pullback that have been identified as false positives, the system may automatically exclude that portion 1104 of the longitudinal representation 914 from consideration when determining other information, e.g., VFR values, pressure values, flow rates, or the like.

[0146] According to some examples, the portion 1102 of the longitudinal representation 914 to be considered may be manually identified. For example, the system may receive one or more inputs identifying a distal end of the portion 1102 and a proximal end of the portion 1102. In some examples, the input may be a click and drag input. In such an example, the system may receive an input, e.g., a selection, identifying a first end of the portion 1102 and a second input, e.g., a release of the selection, identifying a second end of the portion 1102. The portion 1104 not included in the selection may be removed from consideration when determining other information associated with the vessel.

[0147] After the portion 1102 of the longitudinal representation is selected, the graphical representation 912 may be updated based on the selected portion 1102. While the example provided above is with respect to selection the portion 1102 on the longitudinal representation 912, the portion to be considered may be identified based on inputs received in connection with the graphical representation 912. Further, while the examples provided herein are with respect to selecting the portion 1102 to be considered, the inputs may be received in connection with the portion 1104 to be removed from consideration.

[0148] Figures 12A and 12B illustrate example screens 1200A, 1200B of the GUI before and after the lumen bridge module mentioned above in respect of Figure 1 has been applied. GUI screens 1200 A, 1200B may, similar to GUI screen 900, include a menu (not shown) including one or more inputs for analyzing and/or processing the information associated with the vessel data. For example, the inputs may include adding and/or removing a side branch, recalculating the VFR, measuring selected regions of interest, correcting false positives (e.g., lumen bridge module), or the like. The lumen bridge module may be applied to one or more consecutive frames captured during the pullback of the probe 104. Typically, the lumen bridge module is applied to frames, such as frames 902b, that have been identified as false positives, e.g., frames that were wrongly identified as being stenotic.

[0149] The lumen bridge module may select a plurality of frames 1202, including the identified frames 902b. In some examples, the system may automatically select the frames 1202. In another example, the system may receive an input corresponding to a selection of a distal frame of the set of frames 1202 and a proximal frame of the set of frames 1202. The selection may, in some examples, be highlighted, as shown in Figure 12B. The set of frames 1202 may include the identified frames 902b.

[0150] The frames 1202 may be interpolated by the lumen bridge module. The interpolation may, in some examples, be a linear interpolation. Interpolation of the frames 1202 may include, for example, tapering the lumen contour and/or diameter from between the proximal and distal points of the selected frames 1202. For example, comparing the lumen profile of the longitudinal representation in Figure 12A to Figure 12B, the lumen profile in Figure 12B has a tapered contour for the frames 1202 inclusive of the identified frames 902b. In contrast, in Figure 12A, the lumen profile narrows at and/or near the location of identified frames 902b.

[0151] The graphical representation 912 may be updated based on the interpolation of frames 1202. As shown in Figure 12B, after the lumen bridge module interpolates frames 1202, the baseline 920 VFR is 0.80 and the maximum 918 VFR is 0.85. In contrast, as shown in Figure 12A, before the lumen bridge module interpolated frames 1202, the baseline 920 VFR is 0.79 and the maximum 918 VFR is 0.85. According to some examples, the maximum VFR, identified as “VFRmax” on example screens 1200A, 1200B (and example screens 1700A, 1700B in Figures 17A and 17B) may, alternatively, be referred to as a target VFR or “VFRtarget.” The maximum VFR, or target VFR, may correspond to a predicted VFR that could be achieved via PCI.

[0152] Figure 13 illustrates an example method of outputting a representation of a blood vessel which incorporates the methods described above for automatically reviewing whether a prior classification of one or more frames or a segment of a blood vessel as stenotic is correct, or whether it is a false positive. The method may be performed by, for example, the computing device of Figure 1. The following operations do not have to be performed in the precise order described below. Rather, various operations can be handled in a different order or simultaneously, and operations may be added or omitted.

[0153] In block 1310, intravascular imaging data of a vessel is. The intravascular imaging data may include, for example, intravascular image frames. The intravascular imaging data may be captured during a pullback of a probe 104 within the vessel using a system such as that of Figure 1.

[0154] In block 1320, a lumen contour for at least one frame of the intravascular imaging data is determined. The lumen contour may be automatically determined using the lumen algorithm execution stage 1002 illustrated in Figure 10, for example, using the above mentioned lumen algorithm, which may be a contour tracking algorithm. The lumen algorithm execution stage 1002 may utilize a model of the vessel tissue to detect the lumen contour of the vessel. For example, as the vessel is scanned by probe 104, features of the lumen may be detected and extracted from the intravascular data obtained by the probe 104. The intravascular data and/or extracted features may be used to generate a model of the vessel. The lumen algorithm execution stage 1002 segments the tissue from the intravascular image and/or model. The lumen contour is then reconstructed by tracking the proximal border of the detected tissue and fitting a smooth curve. In this example, proximal refers to the region closest to the intravascular catheter. Neighboring frames of intravascular image data can be used to correct low-quality lumen contours.

[0155] In block 1330, a lumen mask for the at least one frame of the intravascular imaging data is determined using an Al model which is separate to or distinct from the lumen algorithm used in block 1320. The Al model may be, for example, the modeling stage 1004 of Figure 10. The modeling stage 1004 may be configured to predict regions of the frame that are within the lumen of the vessel and regions of the frame that are outside the lumen of the vessel. The identification results in a lumen mask, for example comprising pixels or regions each indicative of whether the pixel or region is inside or outside the lumen. In contrast to the lumen algorithm execution stage 1002 which may typically be implemented using a geometric graphical analysis, the modeling stage 1004 may use an Al model trained using a large number of intraluminal images and annotations to extract features of the lumen.

[0156] In block 1340, the lumen contour and the lumen mask are compared. According to some examples, the lumen contour and the lumen mask may be compared to determine how much agreement there is between the lumen contour and the lumen mask. [0157] In block 1350, a measure of similarity results from the comparison between the lumen contour and the lumen mask. This measure of similarity may be an overlap value providing an indication of how much agreement there is between the lumen contour and the lumen mask in respect of the boundary of the lumen within the segment or one or more frames of the blood vessel classified as stenotic . For example, the more similar the lumen contour and the lumen mask are, the greater the agreement may be and, therefore, the greater the overlap value may be. The overlap value may, in some examples, be a Jaccard index value. In some examples, the overlap value may be thresholded. In such an example, when the overlap value for the frame is below the threshold, the frame may be identified as a false positive.

[0158] In block 1360, when the measure of similarity or overlap value for the at least one frame of the intravascular imaging data is below a threshold, an indication that classification of the at least one frame of the intravascular imaging data as stenotic is a false positive may be provided for output. The indication may be, for example, a color coding of the at least one frame or a segment of the blood vessel that has been identified as a false positive. The false positive may be, for example, a frame of intravascular imaging data that was wrongly identified as being stenotic. Frames may be wrongly identified as being stenotic when the frames are located at or near a side branch, when the vessel is not properly and/or fully flushed, when there are other artifacts in the vessel captured in the frame, or the like.

[0159] According to some examples, a two-dimensional representation of the vessel may be generated based on the intravascular imaging data. The two-dimensional representation may be symmetrical about a longest axis of the two-dimensional. The two-dimensional representation may be generated based on diameter values of the vessel.

[0160] An input may be received relative to the two-dimensional representation. The input may be, for example, a user input. The input may correspond to a selection of a plurality of frames of that intravascular imaging data. The plurality of frames may include, for example, at least one frame that was identified as a false positive. The selected plurality of frames may be interpolated to taper the lumen profile. By interpolating the selected frames, the lumen profile may be adjusted, or corrected, to better represent the actual lumen profile as compared to the lumen profile generated based on the frames identified as false positives.

[0161] In some examples, additional values associated with the vessel may be updated based on the interpolation of the plurality of frames. For example, as the lumen profile of the interpolated frames changes, e.g., updated, calculations made based on the lumen profile may be updated. The values may include, for example, a VFR value for a selected frame, a baseline VFR value, a maximum VFR value, or the like. In some examples, the values may include an EEL value, diameter value, flow rate, pressure, etc. According to some examples, a graphical representation may be generated based on the intravascular imaging data. The graphical representation may be updated based on the interpolated plurality of frames. For example, as values associated with the vessel are updated, the graphical representation of the values may be updated. The graphical representation may , in some examples, be a representation of VFR values or pressure values. [0162] Figure 14 illustrates an example pressure curve that is generated based on intravascular imaging data, for example by the computing device 112 of figure 1, and which may be presented to a user using the above GUI. The pressure curve 1400 may be a graphical representation of the pressure measurements along a pullback of probe 104. The pressure curve 1400 may substantially correspond to a two-dimensional representation 1402 of the intravascular imaging data. For example, the pressure curve 1400 may graphically represent the pressure measurements of the vessel shown in the two-dimensional representation 1402. The pressure measurements may be derived from information obtained during the pullback of probe 104. For example, the vessel lumen dimensions may be obtained and/or determined based on the intravascular data obtained during the pullback. Using a model, such as a resistance model, the pressure along the pullback region can be determined based on the vessel lumen dimensions and/or other intravascular data obtained during the pullback. As the pressure along the pullback is determined on a frame by frame basis, the determined pressure measurements can be transformed into a pressure curve 1400. According to some examples, the longitudinal axis of the of the pressure curve 1400 may correspond to a location of the pullback in the two-dimensional representation 1402.

[0163] The pressure curve 1400 may, in some examples, be used to determine VFR values. For example, as VFR is the ratio of the distal pressure to aortic pressure, the VFR value for a location within the vessel may be determined. The aortic pressure may, in some examples, correspond to the proximal location of the vessel. As shown in Figure 14, the aortic pressure used when determining VFR is 90mmHg, as it is the most proximal pressure measurement of the vessel. A VFR value for any point along the pullback may be determine using the following equation: VFR = Pd/Pa, where Pd is the distal pressure value and Pa is the aortic pressure value.

[0164] According to some examples, a flow resistance model may be used to determine the VFR of the blood vessel. The flow resistance of the vessel may be determined using the intravascular imaging data of the vessel. The intravascular imaging data may be used to identify the length of the vessel, or the region of interest, as well as the cross-sectional diameter or area of the vessel. The location and the cross-sectional diameter or area of side branches within the vessel may also be identified using the intravascular imaging data. The vessel may include a proximal and distal segment that were not included in the region of interest, but still have an effect on the pressure drops that occur between the aortic pressure (“Pa”) and the distal- end pressure (“Pd”). VFR may be determined based on the intravascular imaging data and the determined vessel size. In some examples, the vessel size may be determined from the intravascular imaging data and/or the extraluminal images.

[0165] The pressure curve 1400 may provide an indication of a lesion region 1406. The indication may correspond to an area along the vessel in which there is a pressure droppage above a threshold, indicating that the pressure droppage is significant. The pressure drop can correspond to lumen reduction, caused by legion region 1406. The indication of the lesion region 1406 can be used as an initial reference for the user. For example, the indication of the legion region 1406 can allow the use to make one or more additional treatment decisions, e.g., coronary interventions. In some examples, the indication of the legion region 1406 can draw the user’s attention to that portion of the vessel to determine whether it is a false positive, such as due to anatomical features (e.g., side branches) or imaging artifacts. The lesion region 1404 identified on the pressure curve 1400 may, in some examples, be identified on as a lesion region 1406 on the two-dimensional representation 1402. In some examples, based on the continuous pressure measurements, a user, such as a physician, can assess the lesion significance. In some examples, the lesion significance may be based on the pressure drop, or changes, within a given lesion region 1404.

[0166] The lesion region 1404 and/or lesion significance may be used to determine one or more interventions. The interventions may include, for example, stenting, removing calcium build up, or the like. [0167] Figure 15 illustrates an example pressure curve that is generated based on intravascular imaging data captured after an intervention, for example by the computing device 112 of Figure 1, and which may be presented to a user using the above GUI . Example types of interventions include stent implantation, balloon angioplasty, laser angioplasty, atherectomy, rotational atherectomy, chronic total occlusions, brachytherapy, or the like. The pressure curve 1500 is generated based intravascular data obtained after stent deployment. The pressure curve 1500 may substantially correspond to a two-dimensional representation 1502 of the post intervention intravascular imaging data. For example, the pressure curve 1500 may graphically represent the post intervention pressure measurements of the vessel shown in the two-dimensional representation 1500. The post intervention pressure curve 1500 may, in some examples, be used to determine post intervention VFR values. In some examples, the post intervention pressure curve 1500 may be used to generate a VFR curve.

[0168] One or both of the pressure curve 1500 and the two-dimensional representation 1502 may include an indication of the stented region 1504, 1506, respectively. The areas of the pressure curve 1500 outside of the stented region 1504 may be used to determine the efficacy of the intervention. In some examples, the areas of the pressure curve 1500 outside of the stented region 1504 may be used to determine the success of the intervention. For examples, if areas 1508, 1510 outside of the stented region 1504 include pressure drops that indicate additional lesions, the efficacy and/or success of the intervention may be determined to be poor. In some examples, if areas 1508, 1510 outside of the stented region 1504 indicate additional lesions, a different and/or additional intervention may be required. The different and/or additional intervention may, in some examples, include selecting a longer stent.

[0169] According to some examples, the system may automatically identify the regions with the most significant pressure drop within the pressure curve 1500 for each pullback. The system may identify, postPCI, the new next most significant pressure drop within the curve 1500. The user can then access the newly identified most significant pressure drop to determine whether one or more additional treatments, e.g., coronary interventions, are necessary. In some examples, the system may automatically determine that additional treatments are necessary. In another example, the system may automatically provide a suggested additional treatment option.

[0170] Figure 16 illustrates another example pressure curve that is generated based on intravascular imaging data captured after an intervention, such as stent deployment, for example by the computing device 112 of Figure 1 , and which may be presented to a user using the above GUI. The pressure curve 1600 may substantially correspond to a two-dimensional representation 1602 of the post intervention intravascular imaging data. For example, the pressure curve 1600 may graphically represent the post intervention pressure measurements of the vessel shown in the two-dimensional representation 1602. The post intervention pressure curve 1600 may, in some examples, be used to determine post intervention VFR values. In some examples, the post intervention pressure curve 1600 may be used to generate a VFR curve. [0171] One or both of the pressure curve 1600 and the two-dimensional representation 1602 may include an indication of the stented region 1604, 1606, respectively. The adequacy of the expansion of the deployed stent may be evaluated based on the pressure curve 1600. According to some examples, the adequacy of the expansion may be determined automatically based on a comparison of actual stent expansion to a target stent expansion profile. In another example, the adequacy of the expansion may be determined automatically based on a comparison of post-percutaneous intervention (“PCI”) pressure values to the prePCI pressure values and/or the predicted pressure values. The predicted pressure values correspond to pressure values in the lumen where the stent was fully expanded, e.g., 100% expansion. The comparison of pre-PCI and post-PCI and/or predicted pressure values can provide an indication as to the adequacy of the expansion. In some examples, rather than comparing pre-PCI and post-PCI and/or predicted pressure values, VFR values may be compared to determine adequate expansion. According to some examples, adequate expansion may include an indication of the post-PCI VFR value being above approximately 0.8. [0172] Adequate expansion of deployed stent may be determined based on an increase in the pressure values within the stented area. For example, the pressure value for a location along the pullback before stent may be compared to the pressure value for the same location after the stent is deployed. If the change in pressure is greater than a threshold amount, the stent may be adequately deployed. For example, an inadequately expanded stent may insignificantly reduce the pressure drop within the stented region 1604, resulting in an ineffective clinical outcome.

[0173] Figures 17A and 17B illustrates aspects of the GUI, in the form of example screens 1700A and 1700B, which may be implemented within the data collection system of Figure 1 , for example in the context of any of figures 3 to 7, 9A, 9B, and 11 A to 12B. In this aspect of the GUI, the extraluminal image 910 of the vessel, and/or graphical representation 912 of one or more characteristics of the vessel, includes color coding. The color coding may be provided, for example, as a “color map” 1702, 1704 where the color used in the image or graph for a particular location within the vessel corresponds to a pressure value or other characteristic of the vessel at that location, as depicted in the extraluminal image 910, on the graphical representation 912, and/or longitudinal presentation 914. In some examples, the color map may correspond to the pressure value at a given point on the curve, e.g., the curve of the graphical representation 912, the pressure curve shown in Figures 14-16, or the like. According to some examples, such as the example shown on screen 1700, at 1.0 of the graphical representation 912, the baseline indication 920, e.g., curve, is green while at 0.7 the baseline indication 920 is red. The baseline indication 920 transitions from green to red, e.g., by going from green to yellow, yellow to orange, and orange to red, as the color map transitions between the two anchor values. The anchor values may correspond to the values at the proximal and distal end of the vessel or region of interest. [0174] In some examples, the colors of the color map may correspond to a range of values on the graphical representation 910. For example, graphical representation 910 shown on screen 1700 is a graphical representation of VFR values. In such an example, green on the color map may correspond to VFR values between 1 and 0.95, yellow may correspond to VFR values between 0.95-0.85, orange may correspond to VFR values between 0.85-0.75, and red may correspond to VFR values 0.75 and below. According to some examples, such as shown by the scale 1706 in screen 1700B, green may correspond to values around 1.0, yellow-green may correspond to values around 0.9, yellow may correspond to values around 0.8, yellow- orange may correspond to values around 0.7, orange may correspond to values around 0.6, and red may correspond to values around 0.5. The ranges and/or values provided for the colors of the color map are just some examples of what the ranges and/or values could be an are not intended to be limiting.

[0175] In examples where the graphical representation 910 is a pressure curve, such as the pressure curve shown in Figures 12-14, the green on the color map may correspond to a pressure value of approximately 1.00, the yellow on the color map may correspond to a pressure value of approximately 0.85, , and the red on the color map may correspond to a pressure value of approximately 0.70 or less. As the green transitions to yellow, the colors may correspond to pressure values between, approximately, 1 (green) to 0.85 (yellow). As the yellow transitions to red, the colors may correspond to pressure values between, approximately, 0.85 (yellow) and 0.70 (red).

[0176] Further, while only green, yellow, orange, and red are shown, a variety of other colors may be included in the color map. For example, the colors may be any of red, orange, yellow, green, blue, indigo, violet, pink, etc. Moreover, while the examples provided herein have green corresponding to the best, or healthiest, pressure value, the colors may be provided in any order such that red may indicate the healthiest pressure value. Accordingly, the examples provided herein are just some combinations of ranges, colors, and orders and are not intended to be limiting.

[0177] The GUI screens 1700A, 1700B may include a color map on the extraluminal image 910 in addition to, or as an alternative of, the color map on the graphical representation 912. The color map may be overlaid and/or layered on the vessel in the extraluminal image 910. In some examples, the color map may be layered over the vessel trace and/or centerline of the vessel in the extraluminal image 910, as shown. In some examples, the color map may be provided next to and/or relative to the vessel in the extraluminal image 910 (not shown).

[0178] The color map may allow a user, e.g., a physician, to easily identify at what point along the vessel the pressure corresponds to a particular value. In some examples, rather than determine a particular value, the color may provide an indication as to the relative values at a given location, thereby assisting in diagnosing and/or identifying regions of interest. For example, based on the color on the color map, prophylactic treatment locations, such as balloons, stents, or the likes, may be easily detectable.

[0179] The described color map aspect of the GUI interface as depicted in figures 17A and 17B may for example be used to depict actual values of a characteristic of the blood vessel determined using image data of the blood vessel, for example before receiving user selection of a zone of the blood vessel for implementing a potential treatment option and then automatically determining predicted values of the characteristic which would occur if the potential treatment option was implemented as defined by the selected zone, for which see for example figures 3 to 78 and the related description text . However, the color map aspect may also or instead be used to depict such predicted values.

[0180] The described color map aspect of the GUI may also be used to depict values or ranges of values of VFR or pressure for the vessel which have been updated following interpolation of a plurality of image frames forming a vessel segment that has incorrectly be identified as stenotic, so as to taper the lumen profile for the plurality of frames, for which see figures 9B to 13 and the associated description text.

[0181] The disclosed system can be configured to automatically update a region of interest in response to a change in vessel geometry. For example, Figures 18A-D of GUI screens 1800A-D allow the user to alter the vessel and present the user with an option to accept or reject an updated region of interest that is automatically determined by the system in response to the alterations. In GUI screen 1800A of Figure 18 A, an extraluminal image 910, an intraluminal image 916, a graphical representation 912, and the longitudinal representation 914 is displayed. Within longitudinal representation 914 a proximal bracket 1808a and a distal bracket 1808b define a region of interest 1806a within the represented vessel. In addition, longitudinal representation 912 contains a plurality of side-branch indicia 1810a-d, which indicate the location of identified side branches within the represented vessel. The graphical representation 912 displays region of interest 1806b and extraluminal image 910 displays region of interest 1806c, each of which corresponds to the region of interest 1806a provided in longitudinal representation 914.

[0182] In accordance with aspects of the disclosure, a user may interact with GUI screen 1800A, so as to alter the geometry of the vessel that is represented on the display. For example, Figure 18B is GUI screen 1800B in which the user has performed a bridge edit for a selected region 1820 within the represented vessel. In particular, region 1820 within longitudinal representation 914 has been selected by the user and the user has provided a command for the system to alter the geometry of region 1820, so that it is smoothly tapered between proximal end 1822a and distal end 1822b of region 1820. In some instances the user may perform an alteration, such as a bridge edit, for frames within longitudinal representation 914 that have been marked with identification 902, indicating that the frames have been wrongly classified. Upon receiving the command to change geometry of the represented vessel, the system may automatically update the VFR analysis, including updating the designated region of interest 1806a, which can correspond to the region for which the planned stenting is to occur. The user may be made aware of the availability of this automatic update via an update indicia on the display. For example, GUI screen 1800B displays a checkmark 1830 on the right-hand side of the screen, which indicates that the received user input has resulted in a potential update.

[0183] In GUI screen 1800C of Figure 18C the user has selected checkmark 1830, so as to bring up menu 1840, which contains an updated-ROI acceptance icon 1842, as well as an updated-ROI rejection icon 1844. In GUI screen 1800C, the user has pressed or hovered a cursor over acceptance icon 1842, so as to display the available updated region of interest that was automatically determined based on the change in vessel geometry that was previously made by the user. In response to selection of acceptance icon 1842, longitudinal representation 914 will be altered to display updated region of interest 1816a. Corresponding alterations are also made to graphical representation 912, which now displays updated region of interest 1816b, and to extraluminal image 910, which now displays updated region of interest 1816c. However, user may instead select updated-ROI rejection icon 1844, which will return the GUI to displaying the previous regions of interest 1806a-c, as shown in GUI screen 1800D of Figure 18D.

[0184] After user has accepted or rejected the updated-ROI, the checkmark 1830 may be removed from the GUI screen. However, if the user later removes the bridge edit for region 1820, the GUI may again display checkmark 1830, so as to indicate that the alteration to the vessel has made another updated-ROI available. Other vessel alterations may also result in the automatic determination of an updated region of interest that can be either accepted or rejected by the user. For example, returning to Figure 18 A, the user may determine that one or more of the identified side branches, as designated by side-branch indicia 1810a- d, are incorrect, and may remove or otherwise alter one or more of the side-branch indicia 1810a-d. An updated region of interest may then be made available to the user based on the user’ s alteration of side branches, and the user may then accept or reject the updated region of interest.

[0185] The disclosed system may also be configured to automatically identify regions within a vessel for which a stent has already been placed. These stented regions may be identified within the longitudinal representation 914. In such an instance, the system may automatically provide the user with an expansion analysis of the stent, including identification of stent malapposition and under-expansion. However, if no stent is detected within the imaged vessel, then the user will not be presented with an expansion analysis.

[0186] For example, Figure 19A is a GUI screen 1900A that provides a review of an OCT pullback in which no stent has been detected. GUI screen 1900A contains a longitudinal representation 914, an extraluminal image 910, and a cross-sectional image 916. Longitudinal representation 914 contains a lumen boundary 1912 that identifies the mean diameter of the vessel lumen along its length. In addition, longitudinal representation 914 contains an EEL boundary 1914 that identifies the mean diameter of the EEL along a length of the vessel. Extraluminal image 910 includes an overlay of VFR data, and a cross- sectional image 916 includes a lumen diameter value 1932, an EEL diameter value 1934, and a lumen area value 1936 for the location shown in the cross-sectional image 916.

[0187] While no stent has been detected within the vessel, the user may still perform a manual expansion analysis for the imaged vessel. For example, Figure 19B is a portion of GUI screen 1900B in which a menu 1940 is displayed with a number of icons 1941-1946. Included within menu 1940 is an icon 1944 that allows for an expansion analysis to be performed. Upon selecting icon 1944, the GUI may allow the user to select the region for which the expansion analysis is to be performed. For example, Figure 19C is a GUI screen 1900C in which the system has entered a manual expansion analysis mode in which the user selects a proximal location 1922 and a distal location 1924 within longitudinal representation 914 that defines region 1920a.

[0188] Upon receiving the user input that defines region 1920a, the system can provide the user with an expansion analysis for region 1920a. This manual expansion analysis can be used to assess the expansion of a stent that has not been detected by the system, such as stents that consist of bioresorbable polymer scaffolds that are not detected within the image data. Accordingly, region 1920a may be manually identified by the user as corresponding to the region at which a bioresorbable stent has been placed. Similarly, areas that have previously been treated by a drug-coated balloon will not be identified within the image data, but may instead be manually identified by the user by implementing the manual expansion analysis.

[0189] Figure 19D is a GUI screen 1900D in which under-expansion areas 1950 are shown within region 1920a. These under-expansion areas 1950 may be based on predetermined settings or based on user input. For example, the user may identify the diameter of the stent or balloon that was used, as well as the desired expansion percentage for the stent. The desired expansion may be based on what would be possible for maximum expansion within the vessel, given the EEL diameters within the vessel. As shown in GUI screen 1900D, the expansion setting 1954 has been set to 100%, meaning that orange under-expansion areas 1950 correspond to the amount of under-expansion relative to what the lumen diameter would be if the stent was expanded to 100% of what would be possible within the vessel. Alternatively, the expansion setting 1954 could be set to other levels of expansion, such as 80% or 90% expansion. The GUI for the manual expansion analysis also allows for an apposition analysis in which a region of potential apposition is shown with respect to an apposition threshold 1956, which may also be set by the user.

[0190] Longitudinal representation 914 may also provide the EEL diameter, lumen diameter, and lumen area values for one or more locations. In GUI 1900D, these values are displayed for the proximal location 1922 and the distal location 1924 within longitudinal representation 914. In addition, longitudinal representation 914 identifies the location of maximum under-expansion 1952, as well as providing the percentage of expansion for that location. As seen in GUI screen 1900D, location 1952 is at 25% expansion. The GUI also provides an expansion percentage 1960 for the location that is being shown as cross-sectional image 916. In addition, extraluminal image 910 identifies the region 1920b that has been selected for the manual expansion analysis.

[0191] The manual expansion analysis may also be used to assess locations for which vessel treatment should occur, including cases in which there is diffuse stenosis within the vessel. For example, Figure 19E is a GUI 1900E in which an expansion analysis has been performed for a non-stented vessel, and a number of potentially flow-limiting areas 1950 are shown within region 1920 of longitudinal representation 914. The length and size of these flow-limiting areas 1950 can be used to identify potential regions of interest within the vessel that will correspond to locations for treatment via angioplasty, stenting, drug-coated balloon, or other intervention.

[0192] The systems and methods described above are advantageous in that they provide a user-friendly, real-time mechanism for reliably assessing potential treatment options. The physician may quickly and easily assess multiple potential stent placements and sizes, and compare various viable options. The physician may also receive real-time feedback indicating the expected impact of the treatment options.

[0193] Unless otherwise stated, the foregoing alternative examples are not mutually exclusive, but may be implemented in various combinations to achieve unique advantages. As these and other variations and combinations of the features discussed above can be utilized without departing from the subject matter defined by the claims, the foregoing description of the embodiments should be taken by way of illustration rather than by way of limitation of the subject matter defined by the claims. In addition, the provision of the examples described herein, as well as clauses phrased as “such as,” “including” and the like, should not be interpreted as limiting the subject matter of the claims to the specific examples; rather, the examples are intended to illustrate only one of many possible embodiments. Further, the same reference numbers in different drawings can identify the same or similar elements.

Claims

1. A method of predicting impact of a potential treatment option for a blood vessel based on one or more values of a characteristic of the blood vessel, the method comprising: receiving, by one or more processors, image data of the blood vessel; determining using the image data, by the one or more processors, one or more actual values of the characteristic; outputting for display one or more of the actual values of the characteristic; receiving a user selection of a zone of the blood vessel for implementing a potential treatment option; and determining, by the one or more processors, one or more predicted values of the characteristic which would occur if the potential treatment option was implemented as defined by the selected zone.

2. The method of claim 1 , wherein the actual and/or predicted values of the characteristic are determined at multiple positions along the blood vessel.

3. The method of claim 1 or 2, wherein outputting for display the one or more of the actual values of the characteristic comprises outputting for display a graphical representation of the actual values in spatial correlation with a graphical representation of the blood vessel.

4. The method of claim 1 or 2, wherein outputting for display the one or more of the actual values of the characteristic comprises outputting at least one of a numerical depiction of an actual value of the characteristic or a graphical representation of actual values of the characteristic.

5. The method of any preceding claim, wherein the potential treatment option includes placement of a stent having a size and placement location corresponding to the selected zone.

6. The method of any preceding claim, wherein receiving the user selection of the zone comprises receiving from the user adjustments to a boundary of an overlay depicted in relation to a graphical representation of the blood vessel.

7. The method of any preceding claim, further comprising outputting for display a first depiction of the user selected zone relative to a first representation of the blood vessel and a second depiction of the user selected zone relative to a second representation of the blood vessel.

8. The method of claim 7, further comprising: receiving user input modifying a boundary of the first depiction of the zone; and automatically updating the second depiction of the zone based on the user input.

9. The method of any preceding claim, further comprising determining whether one or more of the predicted values of the characteristic meet a predetermined threshold.

10. The method of claim 9, further comprising outputting a representation of the predicted values of the characteristic in a first mode if the predicted updated values characteristic meet the predetermined threshold, and outputting the representation of the predicted values of the characteristic in a second mode if the predicted updated measures do not meet the predetermined threshold.

11. The method of any preceding claim, wherein the image data used to determine the one or more actual values of the characteristic is intravascular image data of the blood vessel.

12. The method of any preceding claim, wherein the characteristic is virtual flow reserve.

13. The method of any preceding claim, wherein outputting for display the one or more of the actual values of the characteristic comprises providing for output, by the one or more processors, a color map on at least one of an extraluminal image or a graphical representation of the actual values.

14. The method of claim 13, wherein the color map comprises one or more colors, each of the one or more colors corresponding to a respective value or range of values of the characteristic.

15. A system for predicting impact of a potential treatment option for a blood vessel on one or more values of a characteristic of the blood vessel, the system comprising: one or more processors, the one or more processors configured to: receive image data of the blood vessel; determine using the image data one or more actual values of the characteristic; output for display one or more of the actual values of the characteristic; receive a user selection of a zone of the blood vessel for implementing a potential treatment option; determine one or more predicted values of the characteristic which would occur if the potential treatment option was implemented as defined by the selected zone.

16. The system of claim 15, wherein the actual and/or predicted values of the characteristic are determined at multiple positions along the blood vessel

17. The system of claim 15 or 16, wherein when outputting for display the one or more of the actual values of the characteristic the one or more processors are further configured to output for display a graphical representation of the actual values in spatial correlation with a graphical representation of the blood vessel.

18. The system of claim 15 or 16, wherein outputting for display the one or more actual values of the characteristic comprises outputting at least one of a numerical depiction of an actual value of the characteristic or a graphical representation of actual values of the characteristic.

19. The system of any of claims 15 to 18, wherein the potential treatment option includes placement of a stent having a size and placement location corresponding to the selected zone.

20. The system of any of claims 15 to 19, wherein when receiving the user selection of the zone the one or more processors are further configured to receive from the user adjustments to a boundary of an overlay depicted in relation to a graphical representation of the blood vessel.

21. The system of any of claims 15 to 20, further comprising outputting for display a first depiction of the user selected zone relative to a first representation of the blood vessel and a second depiction of the user selected zone relative to a second representation of the blood vessel.

22. The system of claim 21, wherein the one or more processors are further configured to: receive user input modifying a boundary of the first depiction of the zone; and automatically update the second depiction of the zone based on the user input.

23. The system of any of claims 15 to 22, wherein the one or more processors are further configured to determine whether one or more of the predicted values of the characteristic meet a predetermined threshold.

24. The system of claim 23, wherein the one or more processors are further configured to output a representation of the predicted values of the characteristic in a first mode if the predicted updated values characteristic meet the predetermined threshold, and outputting the representation of the predicted values of the characteristic in a second mode if the predicted updated measures do not meet the predetermined threshold.

25. The system of any of claims 15 to 24, wherein the image data used to determine the one or more actual values of the characteristic is intravascular image data of the blood vessel.

26. The system of any of claims 15 to 25, wherein the characteristic is virtual flow reserve.

27. The system of any of claims 15 to 26, wherein when outputting for display the one or more of the actual values of the characteristic the one or more processors are further configured to provide for output, by the one or more processors, a color map on at least one of an extraluminal image or a graphical representation of the actual values.

28. The system of claim 27, wherein the color map comprises one or more colors, each of the one or more colors corresponding to a respective value or range of values of the characteristic.

29. One or more non-transitory computer-readable storage media encoding instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising: receiving image data of the blood vessel; determining using the image data one or more actual values of the characteristic; outputting for display one or more of the actual values of the characteristic; receiving a user selection of a zone of the blood vessel for implementing a potential treatment option; determining one or more predicted values of the characteristic which would occur if the potential treatment option was implemented as defined by the selected zone.

30. The one or more non-transitory computer-readable storage media of claim 29, wherein the actual and/or predicted values of the characteristic are determined at multiple positions along the blood vessel

31. The one or more non-transitory computer-readable storage media of claim 29 or 30, wherein outputting for display the one or more of the actual values of the characteristic comprises outputting for display a graphical representation of the actual values in spatial correlation with a graphical representation of the blood vessel.

32. The one or more non-transitory computer-readable storage media of any of claims 29 to

31, wherein outputting for display the one or more of the actual values of the characteristic comprises outputting at least one of a numerical depiction of an actual value of the characteristic or a graphical representation of actual values of the characteristic.

33. The one or more non-transitory computer-readable storage media of any of claims 29 to

32, wherein the potential treatment option includes placement of a stent having a size and placement location corresponding to the selected zone.

34. The one or more non-transitory computer-readable storage media of any of claims 29 to

33, wherein receiving the user selection of the zone comprises receiving from the user adjustments to a boundary of an overlay depicted in relation to a graphical representation of the blood vessel.

35. The one or more non-transitory computer-readable storage media of any of claims 29 to 34, wherein the operations further comprise outputting for display a first depiction of the user selected zone relative to a first representation of the blood vessel and a second depiction of the user selected zone relative to a second representation of the blood vessel.

36. The one or more non-transitory computer-readable storage media of claim 35, wherein the operations further comprise: receiving user input modifying a boundary of the first depiction of the zone; and automatically updating the second depiction of the zone based on the user input.

37. The one or more non-transitory computer-readable storage media of any of claims 29 to 36, wherein the operations further comprise determining whether one or more of the predicted values of the characteristic meet a predetermined threshold.

38. The one or more non-transitory computer-readable storage media of claim 37, wherein the operations further comprise outputting a representation of the predicted values of the characteristic in a first mode if the predicted updated values characteristic meet the predetermined threshold, and outputting the representation of the predicted values of the characteristic in a second mode if the predicted updated measures do not meet the predetermined threshold.

39. The one or more non-transitory computer-readable storage media of any of claims 29 to

38, wherein the image data used to determine the one or more actual values of the characteristic is intravascular image data of the blood vessel.

40. The one or more non-transitory computer-readable storage media of any of claims 29 to

39, wherein the characteristic is virtual flow reserve.

41. The one or more non-transitory computer-readable storage media of any of claims 29 to

40, wherein outputting for display the one or more of the actual values of the characteristic comprises providing for output, by the one or more processors, a color map on at least one of an extraluminal image or a graphical representation of the actual values.

42. The one or more non-transitory computer-readable storage media of claim 41 , wherein the color map comprises one or more colors, each of the one or more colors corresponding to a respective value or range of values of the characteristic.

43. A method of automatically reviewing a classification of a segment of a blood vessel as stenotic, comprising: receiving, by one or more processors, one or more frames of intravascular imaging data of the segment; determining from at least one of the frames, by the one or more processors using a lumen algorithm, a lumen contour of the blood vessel within the segment; determining from at least one of the frames, by the one or more processors using an artificial intelligence model separate to the lumen algorithm, a lumen mask of the blood vessel within the segment; determining, by the one or more processors, a measure of similarity between the lumens defined by the lumen contour and the lumen mask; and if the measure of similarity is below a threshold, providing for output, by the one or more processors, an indication that the classification of the segment as stenotic is a false positive.

44. The method of claim 43, wherein the lumen algorithm is a contour tracking algorithm and the determined lumen contour is a determined boundary of the lumen within the segment.

45. The method of claim 43 or 44, wherein the lumen mask comprises pixels each indicative of whether the pixel is inside or outside the lumen.

46. The method of any of claims 43 to 45, wherein the measure of similarity is a Jaccard index value.

47. The method of claim 46, wherein the Jaccard index value is calculated by dividing a cross sectional amount of the segment determined by both the lumen contour and the lumen mask to be within the lumen by a cross sectional amount of the segment determined by either or both of the lumen contour and the lumen mask to be within the lumen.

48. The method of any of claims 43 to 47, further comprising generating, by the one or more processors based on the intravascular imaging data, a two-dimensional representation of the vessel, wherein the two-dimensional representation is symmetrical about a longest axis of the two-dimensional representation.

49. The method of claim 48, further comprising: receiving, by the one or more processors, an input relative to the two-dimensional representation, wherein: the input corresponds to a selection of a plurality of frames of the intravascular imaging data, and the selection of the plurality of frames includes the at least one frame of the intravascular imaging data that is the false positive; and interpolating, by the one or more processors, the plurality of frames to taper the lumen profile for the selected plurality of frames.

50. The method of claim 49, further comprising updating, by the one or more processors, additional values associated with the vessel based on the interpolation of the plurality of frames.

51. The method of claim 50, wherein the additional values includes at least one of virtual flow reserve (VFR) value for a selected frame, a baseline VFR value, or a maximum VFR value.

52. The method of claim 50, further comprising: generating, by the one or more processors based on the intravascular imaging data, a graphical representation; and updating, by the one or more processors based on the interpolated plurality of frames, the graphical representation.

53. The method of claim 52, wherein the graphical representation is a representation of virtual flow reserve (VFR) values or pressure values.

54. The method of claim 53, wherein the graphical representation is a color map comprising one or more colors, each of the one or more colors corresponding to a respective value or range of values of the VFR values or pressure values.

55. A system for automatically reviewing a classification of a segment of a blood vessel as stenotic, the system comprising: one or more processors, the one or more processors configured to: receive one or more frames of intravascular imaging data of the segment; determine from at least one of the frames, using a lumen algorithm, a lumen contour of the blood vessel within the segment; determine from at least one of the frames, using an artificial intelligence model separate to the lumen algorithm, a lumen mask of the blood vessel within the segment, determine a measure of similarity between the lumens defined by the lumen contour and the lumen mask; and if the measure of similarity is below a threshold, providing for output an indication that the classification of the segment as stenotic is a false positive.

56. The system of claim 55, wherein the lumen algorithm is a contour tracking algorithm and the determined lumen contour is a determined boundary of the lumen within the segment.

57. The system of claim 55 or 576, wherein the lumen mask comprises pixels each indicative of whether the pixel is inside or outside the lumen.

58. The system of any of claims 55 to 57, wherein the measure of similarity is a Jaccard index value.

59. The system of claim 58, wherein the Jaccard index value is calculated by dividing a cross sectional amount of the segment determined by both the lumen contour and the lumen mask to be within the lumen by a cross sectional amount of the segment determined by either or both of the lumen contour and the lumen mask to be within the lumen.

60. The system of any of claims 55 to 59, wherein the one or more processors are further configured to generate, based on the intravascular imaging data, a two-dimensional representation of the vessel, wherein the two-dimensional representation is symmetrical about a longest axis of the two- dimensional representation.

61. The system of claim 60, wherein the one or more processors are further configured to: receive an input relative to the two-dimensional representation, wherein: the input corresponds to a selection of a plurality of frames of the intravascular imaging data, and the selection of the plurality of frames includes the at least one frame of the intravascular imaging data that is the false positive; and interpolate the plurality of frames to taper the lumen profile for the selected plurality of frames.

62. The system of claim 61, further comprising updating, by the one or more processors, additional values associated with the vessel based on the interpolation of the plurality of frames.

63. The system of claim 62, wherein the additional values includes at least one of virtual flow reserve (VFR) value for a selected frame, a baseline VFR value, or a maximum VFR value.

64. The system of claim 62, wherein the one or more processors are further configured to: generate, based on the intravascular imaging data, a graphical representation; and update, based on the interpolated plurality of frames, the graphical representation.

65. The system of claim 64, wherein the graphical representation is a representation of virtual flow reserve (VFR) values or pressure values.

66. The system of claim 65, wherein the graphical representation is a color map comprising one or more colors, each of the one or more colors corresponding to a respective value or range of values of the VFR values or pressure values.

67. One or more non-transitory computer-readable storage media encoding instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising: receiving one or more frames of intravascular imaging data of the segment; determining from at least one of the frames, using a lumen algorithm, a lumen contour of the blood vessel within the segment; determining from at least one of the frames, using an artificial intelligence model separate to the lumen algorithm, a lumen mask of the blood vessel within the segment, determining a measure of similarity between the lumens defined by the lumen contour and the lumen mask; and if the measure of similarity is below a threshold, providing for output, by the one or more processors, an indication that the classification of the segment as stenotic is a false positive.

68. The one or more non-transitory computer-readable storage media of claim 67, wherein the lumen algorithm is a contour tracking algorithm and the determined lumen contour is a determined boundary of the lumen within the segment.

69. The one or more non-transitory computer-readable storage media of claim 67 or 7108, wherein the lumen mask comprises pixels each indicative of whether the pixel is inside or outside the lumen.

70. The one or more non-transitory computer-readable storage media of any of claims 67 to 69, wherein the measure of similarity is a Jaccard index value.

71. The one or more non-transitory computer-readable storage media of claim 70, wherein the Jaccard index value is calculated by dividing a cross sectional amount of the segment determined by both the lumen contour and the lumen mask to be within the lumen by a cross sectional amount of the segment determined by either or both of the lumen contour and the lumen mask to be within the lumen.

72. The one or more non-transitory computer-readable storage media of any of claims 67 to 71, wherein the operations further comprise generating, based on the intravascular imaging data, a two- dimensional representation of the vessel, wherein the two-dimensional representation is symmetrical about a longest axis of the two-dimensional representation.

73. The one or more non-transitory computer-readable storage media of claim 72, wherein the operations further comprise: receiving an input relative to the two-dimensional representation, wherein: the input corresponds to a selection of a plurality of frames of the intravascular imaging data, and the selection of the plurality of frames includes the at least one frame of the intravascular imaging data that is the false positive; and interpolating the plurality of frames to taper the lumen profile for the selected plurality of frames.

74. The one or more non-transitory computer-readable storage media of claim 73, wherein the operations further comprise updating additional values associated with the vessel based on the interpolation of the plurality of frames.

75. The one or more non-transitory computer-readable storage media of claim 74, wherein the additional values includes at least one of virtual flow reserve (VFR) value for a selected frame, a baseline VFR value, or a maximum VFR value.

76. The one or more non-transitory computer-readable storage media of claim 74, wherein the operations further comprise: generating, based on the intravascular imaging data, a graphical representation; and updating, based on the interpolated plurality of frames, the graphical representation.

77. The one or more non-transitory computer-readable storage media of claim 76, wherein the graphical representation is a representation of virtual flow reserve (VFR) values or pressure values.

78. The one or more non-transitory computer-readable storage media of claim 77, wherein the graphical representation is a color map comprising one or more colors, each of the one or more colors corresponding to a respective value or range of values of the VFR values or pressure values.