Automatic Lesion Detection

Abstract

The present disclosure provides systems and methods for determining external elastic lamina (EEL) values ​​for intravascular image frames and using predicted EEL to determine plaque burden for each image frame. The predicted EEL value can be determined based on a threshold number of image frames within a region of interest where the plaque burden is below a threshold plaque burden. In some examples, the predicted EEL value can be determined as a function of intimal and medial thicknesses and predicted lumen diameter derived from the HK model. The determined plaque burden and visible EEL arcs can be used to automatically identify lesions and suggest candidate treatment zones.

Description

[Technical Field]

[0001] [CROSS-REFERENCE TO RELATED APPLICATIONS] This application claims the benefit of the filing date of U.S. Provisional Patent Application No. 63 / 527,108, filed July 17, 2023, the disclosure of which is incorporated herein by reference. [Background technology]

[0002] Various measurements of a blood vessel can be determined automatically using computer software based on image data captured during pullback of an intravascular probe within the blood vessel. However, the presence of plaque or other deposits within the blood vessel can prevent or make the automatic determination of such measurements more difficult. This can result in incomplete data being obtained about the blood vessel from the pullback. The incomplete data can make further decisions using such data insufficient, such as the automatic selection of candidate treatment zones, or can prevent such further decisions from being made fully automatically. This can require a physician to manually input or adjust clinical parameters, such as candidate treatment zones. Summary of the Invention

[0003] The present disclosure generally relates to an apparatus, method, computer program code, etc. for automatically estimating external elastic lamina (EEL) values ​​for intravascular image frames. The estimated EEL can be used to determine plaque burden for the image frame from which the EEL value was estimated. The EEL value can be, for example, EEL diameter or EEL area. For example, plaque may overgrow within a vessel, preventing automatic determination of the EEL value during pullback, or the EEL value may not be visible to a user, such as an analyst or physician. For image frames in which EEL values ​​are not detected and / or have measurement errors, EEL values ​​can be predicted. EEL values ​​can be predicted for image frames within a region of interest. The region of interest can be, for example, a region of a vessel between two side branches.

[0004] In one example, the region of interest can include a first set of frames and a second set of frames. The first set of frames can be image frames in which an EEL value was detected during pullback. The second set of frames can be image frames in which no EEL value was detected during pullback. If the first set of frames includes a threshold number of image frames in which the plaque load is below a threshold plaque load, the EEL value can be predicted using at least the threshold number of frames. The threshold plaque load can be, for example, a healthy plaque load. The threshold plaque load can be a plaque load between about 0% and about 50%. A representative value of the EEL values ​​of the threshold number of image frames can be determined. The representative EEL value can be, for example, an average EEL value, a median EEL value, or a mean EEL value. The representative EEL value can be used as a predicted EEL value for the image frames in the second set of frames. According to some examples, the representative EEL value of the first set of frames can be used as the EEL value determined for the frames in the first subset of frames. For example, if some of the frames in the first subset of frames have EEL values ​​with measurement errors, the determined representative EEL value can be used as the EEL value for the frames in the first subset of frames with the erroneous EEL values. The erroneous EEL values ​​can be EEL values ​​that exceed the standard deviation of the determined representative EEL values, EEL values ​​that are outside a predetermined range of the determined representative EEL values, etc.

[0005] In another example, the EEL value for frames within the region of interest can be predicted as a function of the intimal and medial thicknesses. Continuing with the previous example, if the first set of frames does not include a threshold number of frames with a plaque burden below the threshold plaque burden, the EEL value can be predicted based on the intimal and medial thicknesses. In some examples, the EEL value can be further based on the HK model. According to some examples, the EEL value can be predicted based on the intimal and medial thicknesses and the HK model. According to some examples, if the EEL value in a segment of the vessel distal to the region of interest is greater than the predicted EEL value for the region of interest, the EEL value of the distal segment can be used as the predicted EEL value for the region of interest. The EEL value of the distal segment can be a representative EEL value for the distal segment. In some examples, if the EEL value in a segment of the vessel proximal to the region of interest is less than the predicted EEL value for the region of interest, the EEL value of the proximal segment can be used as the predicted EEL value for the region of interest. The EEL value of the proximal segment can be a representative EEL value for the proximal segment. The EEL values ​​for the proximal or distal segments can be used based on the natural taper of the vessel from the proximal to the distal vessel segment.

[0006] The predicted EEL value can be used to determine the plaque burden for each image frame. For example, the predicted EEL value and lumen area for a given frame can be used to determine the plaque burden. The determined plaque burden can be used to identify lesions within a blood vessel. According to some examples, if a first lesion is within a threshold distance from a second lesion, the first and second lesions can be identified as a composite lesion. The identified lesions can be used to identify candidate treatment zones. For example, candidate treatment zones can be identified as those corresponding to image frames that are not within the identified lesions, are a minimum distance away from the identified lesions, have a plaque burden less than the lesion load, a visible EEL equal to or greater than an arc threshold, etc.

[0007] One aspect of the present disclosure relates to a method comprising: one or more processors receiving vascular data including intravascular image data; based on the vascular data, the one or more processors identifying a region of interest in the vessel; based on the vascular data, the one or more processors identifying a first subset of frames in the region of interest having a plaque load below a threshold plaque load, the first subset of frames including a threshold number of frames; based on the EEL values ​​of the first subset of frames, the one or more processors determining an external elastic lamina (EEL) value for the first subset of frames; based on the EEL values ​​of the first subset of frames, the one or more processors determining a representative EEL value for the region of interest; and based on the representative EEL value, the one or more processors determining a plaque load for a second subset of frames in the region of interest, the second subset of frames being different from the first subset of frames.

[0008] The region of interest can be defined between the first and second vascular branches. The EEL value can be an EEL diameter or an EEL area, and the representative EEL value can be a representative EEL diameter or a representative EEL area. The threshold plaque burden can correspond to a plaque burden of 0% to 50% or about 0% to about 50%. The threshold number of frames can be at least three frames.

[0009] Determining the plaque burden can be further based on the lumen area for each frame. Determining the plaque burden for each frame can include calculating as 1 - (lumen area / representative EEL value).

[0010] The method may further include one or more processors identifying a lesion based on the determined plaque burden. The method may further include one or more processors identifying a candidate treatment zone based on the identified lesion. Identifying a candidate treatment zone may further include one or more processors identifying frames proximal and distal to the identified lesion, where at least one of the plaque burdens of the frames proximal and distal to the identified lesion is below a threshold or the visible EEL arc is equal to or greater than an arc threshold. If the first lesion is within a minimum gap distance from the second lesion, the method may further include one or more processors identifying the first and second lesions as a composite lesion.

[0011] Another aspect of the present disclosure relates to a system including one or more processors that can be configured to receive vascular data including intravascular image data, identify a region of interest in the vessel based on the vascular data, identify a first subset of frames within the region of interest having a plaque load below a threshold plaque load, the first subset of frames including a threshold number of frames, determine external elastic lamina (EEL) values ​​for the first subset of frames, determine a representative EEL value for the region of interest based on the EEL values ​​of the first subset of frames, and determine plaque load for a second subset of frames within the region of interest based on the representative EEL value, the second subset of frames being different from the first subset of frames.

[0012] Yet another aspect of the present disclosure relates to a non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to: receive vascular data including intravascular image data; identify a region of interest in a vessel based on the vascular data; identify a first subset of frames within the region of interest having a plaque load below a threshold plaque load, where the first subset of frames includes a threshold number of frames; determine external elastic lamina (EEL) values ​​for the first subset of frames; determine a representative EEL value for the region of interest based on the EEL values ​​of the first subset of frames; and determine a plaque load for a second subset of frames within the region of interest based on the representative EEL value, where the second subset of frames are different from the first subset of frames.

[0013] One aspect of the present disclosure relates to a method comprising: one or more processors receiving vascular data including intravascular image data; based on the vascular data, the one or more processors identifying a region of interest in the vessel; based on the vascular data, the one or more processors determining predicted external elastic lamina (EEL) values ​​for frames within the region of interest as a function of intimal and medial thickness; and based on the predicted EEL values, the one or more processors determining plaque burden for frames within the region of interest.

[0014] The region of interest can be defined between the first and second branch vessels. The predicted EEL value can be further determined as a function of at least one of the lumen area or lumen diameter derived from the HK model.

[0015] Determining the plaque burden can be further based on the lumen area for each frame. Determining the plaque burden for each frame can include calculating 1 - (lumen area / predicted EEL value).

[0016] The method may further include one or more processors identifying a lesion based on the determined plaque burden. The method may further include one or more processors identifying a candidate treatment zone based on the identified lesion. Identifying a candidate treatment zone may further include one or more processors identifying frames proximal and distal to the identified lesion, where at least one of the plaque burdens of the frames proximal and distal to the identified lesion is below a threshold or the visible EEL arc is equal to or greater than an arc threshold. If the first lesion is within a minimum gap distance from the second lesion, the method may further include one or more processors identifying the first and second lesions as a composite lesion.

[0017] The method may also include the one or more processors comparing the predicted EEL value for the region of interest to a representative EEL value for a proximal segment of the blood vessel, and if the representative EEL value for the proximal segment is less than the predicted EEL value for the region of interest, determining the plaque burden for the region of interest is based on the representative EEL value of the proximal segment.

[0018] The method may further include the one or more processors comparing the predicted EEL value for the region of interest to a representative EEL value for a distal segment of the blood vessel, and if the representative EEL value for the distal segment is greater than the predicted EEL value for the region of interest, determining the plaque burden for the region of interest is based on the representative EEL value of the distal segment.

[0019] Another aspect of the present disclosure relates to a system including one or more processors that can be configured to receive vascular data including intravascular image data, identify a region of interest in the vessel based on the vascular data, determine predicted external elastic lamina (EEL) values ​​for frames within the region of interest as a function of intimal and medial thickness based on the vascular data, and determine plaque burden for the frames within the region of interest based on the predicted EEL values.

[0020] Yet another aspect of the present disclosure relates to a non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to receive vascular data including intravascular image data; identify a region of interest in a vessel based on the vascular data; determine predicted external elastic lamina (EEL) values ​​for frames within the region of interest as a function of intimal and medial thickness based on the vascular data; and determine plaque burden for the frames within the region of interest based on the predicted EEL values.

[0021] 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 U.S. Patent and Trademark Office upon request and payment of the necessary fee. [Brief explanation of the drawings]

[0022] [Figure 1] FIG. 1 is a diagram of an exemplary system according to aspects of the present disclosure. [Figure 2A] 2 is a diagram of an exemplary intravascular image that can be captured using the system of FIG. 1, according to an embodiment of the present disclosure. [Figure 2B] FIG. 2 is a diagram of an example annotated intravascular image that may be captured and / or annotated using the system of FIG. 1, according to aspects of the present disclosure. [Figure 2C] 2 is a diagram of an example longitudinal representation that can be generated using the system of FIG. 1, according to an embodiment of the present disclosure. [Figure 3] 2 is a diagram of an example user interface that can be implemented using the system of FIG. 1 according to aspects of the present disclosure. [Figure 4] 1. FIG. 4 is a diagram of another example user interface that can be implemented using the system of FIG. 1, according to aspects of the present disclosure. [Figure 5]2 is an exemplary longitudinal representation for identifying a lesion that can be generated using the system of FIG. 1 according to an embodiment of the present disclosure. [Figure 6] 2 is a diagram of an example user interface for identifying candidate treatment zones that can be implemented using the system of FIG. 1 according to aspects of the present disclosure. [Figure 7] 2 is a diagram of an example user interface for stent planning that can be implemented using the system of FIG. 1 according to aspects of the present disclosure. [Figure 8] FIG. 8 is a flow diagram illustrating a method for determining plaque burden based on average EEL values ​​that can be used in providing the interface screens of FIGS. 3, 4, 6, and 7, according to an embodiment of the present disclosure. [Figure 9] FIG. 8 is a flow diagram illustrating a method for determining plaque burden based on predicted EEL values ​​that may be used in providing the interface screens of FIGS. 3, 4, 6, and 7, according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0023] The technology generally relates to automatically identifying lesions based on plaque burden within a blood vessel. The plaque burden can be determined as a function of a lumen value and an external elastic lamina (EEL) value. The EEL value can be, for example, an EEL diameter and / or an EEL area. The lumen value and the EEL value can be automatically determined based on intravascular image data acquired during pullback of the intravascular probe. In some instances, plaque overgrows within the blood vessel, preventing the EEL value from being determined based on the image data. In some instances, plaque may prevent the EEL from being visible to a user in the intravascular image data, or there may have been an error in determining the EEL value. The user can be, for example, an analyst, a physician, etc. In instances where the EEL value cannot be determined based on the image data acquired during pullback, the EEL value can be predicted.

[0024] For a region of interest, such as a region of a blood vessel between two side branches, an EEL value can be predicted as a representative EEL value. According to some examples, the representative EEL value can be predicted as an average, median, mean, or mode EEL value. For clarity throughout this disclosure, the representative EEL value is used. However, the representative EEL value can be interchangeable with the median EEL value, the average EEL value, and / or the mean EEL value, and thus the use of "representative" is not intended to be particularly limiting.

[0025] The representative EEL value can be determined based on a threshold number of frames within the region of interest. For example, the region of interest can have a first set of frames in which the plaque load of the image frames is below a threshold plaque load. The threshold plaque load can be between about 0% and about 50%. The first set of image frames can have a threshold number of image frames, such as at least three image frames. The region of interest can have a second set of image frames in which the plaque load is greater than the threshold plaque load, an EEL value could not be automatically determined, etc. The representative EEL value can be determined based on the EEL values ​​for the first set of frames. The representative EEL value can be used as a predicted EEL value for frames within the second set of frames. According to some examples, the representative EEL value can be used as an EEL value for at least one of the frames within the first set of frames. For example, the first set of frames can include one or more frames having EEL values ​​with measurement error. The representative EEL value can be used for frames within the first set of frames having EEL values ​​outside the standard deviation of the representative EEL value, outside a predetermined range of the representative EEL value, etc.

[0026] According to some examples, the EEL value can be predicted as a representative EEL value for a threshold number of frames based on the HK model. For example, when applying the HK model, the HK model operates under the assumption that the vessel size, e.g., diameter, area, etc., remains substantially constant between the two side branches.

[0027] In examples where the region of interest does not include a threshold number of frames with a plaque load less than the threshold plaque load or a visible EEL, the EEL value for the frames within the region of interest can be predicted as a function of the intima thickness and the media thickness. The intima thickness can be, for example, the minimum intima thickness, and the media thickness can be, for example, the minimum media thickness. The minimum intima thickness and the minimum media thickness can be determined based on histological observation of the blood vessel. For example, based on histological observation of the blood vessel, the intima thickness and the media thickness can be determined by the standard deviation. The minimum intima thickness and the minimum media thickness can be determined based on the lower limit of the standard deviation. For example, if the combined thickness of the intima and the media is determined to be 427 micrometers ± 190 micrometers, the minimum combined thickness of the intima and the media is 237 micrometers. According to some examples, the predicted EEL value can further be determined as a function of the lumen area / diameter derived from the HK model.

[0028] The HK model assumes that blood vessels naturally taper from a proximal segment of the blood vessel to a distal segment of the blood vessel. In such examples, the predicted EEL value for the region of interest can be compared to a representative EEL value for a segment proximal to the region of interest. According to some examples, the proximal segment can be adjacent to the region of interest or further proximal to the region of interest. If the representative EEL value for the proximal segment is less than the predicted EEL value, the representative EEL value can be used as the predicted EEL value for the region of interest. In some examples, the predicted EEL value for the region of interest can be compared to a representative EEL value for a segment of the blood vessel distal to the region of interest. In some examples, the distal segment can be adjacent to the region of interest or further distal to the region of interest. If the representative EEL value for the distal segment is greater than the predicted EEL value for the region of interest, the representative EEL value for the distal segment can be used as the predicted EEL value for the region of interest.

[0029] The representative EEL value and / or the predicted EEL value (the representative EEL value, the predicted EEL value, or both) can be used to determine the plaque burden for each frame in the region of interest. For example, the plaque burden can be determined based on the EEL value, which is either the median or the predicted EEL value, and the lumen area for each frame. The median or predicted EEL value can be, for example, the median or predicted EEL area. In some examples, the median or predicted EEL value can be the median or predicted EEL diameter. According to some examples, the plaque burden can be determined using the following formula: 1 - (lumen area / EEL value). The EEL value can be the average EEL value or the predicted EEL value.

[0030] Plaque burden can be used to identify lesions. For example, image frames with the greatest amount of plaque burden can be identified. According to some examples, the identified image frames can be those exceeding a disease threshold. The disease threshold can be, for example, a stenosis percentage, a plaque burden, etc. In some examples, the disease threshold can be, for example, a plaque burden having a range of about 60% to about 100%. According to some examples, the disease threshold can be a plaque burden of about 70%. In some examples, the disease threshold can be determined based on disease characteristics, such as plaque burden, stenosis percentage, etc., and the percentage can be greater than or less than 70%, or can range within or outside the range of 60% to 100%.

[0031] The identified image frame may be the initial image frame for the given lesion. To determine the size and location of the lesion, the plaque load of frames proximal and distal to the initial frame may be compared to the treatment zone threshold until the plaque load of the image frames proximal and distal to the initial frame is less than the treatment zone threshold. The treatment zone threshold may be approximately 50% plaque load. In some instances, the treatment zone threshold may be greater than or less than 50% plaque load. For example, the treatment zone threshold may correspond to approximately 45% plaque load. In some instances, the treatment zone threshold may range from approximately 0% to approximately 50% plaque load, etc. In some instances, the treatment zone threshold may be patient-specific, physician-determined, etc. In some instances, the treatment zone end frame may correspond to a visible EEL arc equal to or greater than the arc threshold. The arc threshold may be at least 180 degrees or approximately 180 degrees. Thus, the exemplary treatment zone thresholds described herein are not intended to be limiting. The image frames between the identified proximal and distal frames are then marked as corresponding to the lesion.

[0032] According to some examples, the process of identifying lesions is repeated until all lesions in the blood vessel are identified. For example, the image frame resulting from the identified lesion can be excluded, and the image frame with the next highest plaque burden can be identified. The lesion containing the image frame with the next highest plaque burden is determined by comparing the plaque burden of frames proximal and distal to the image frame with the next highest plaque burden with a treatment zone threshold. The plaque burden of frames proximal and distal to the image frame with the next highest plaque burden can be compared with the treatment zone threshold until the plaque burden of image frames proximal and distal to the initial frame is less than the treatment zone threshold. The image frames between the identified proximal and distal frames are then marked as corresponding to the lesion.

[0033] After each lesion is identified, a gap or distance between the lesions can be determined. In instances where the gap is substantially equal to or less than a threshold distance, the lesions can be combined and identified as a single lesion. For example, if a first lesion is within a minimum gap distance from a second lesion, e.g., if the gap between the first and second lesions is less than a threshold distance, the first and second lesions can be identified as a combined lesion. In instances where the gap is greater than the threshold distance, the lesions can be identified as separate lesions. According to some examples, a proposed or candidate treatment zone can be identified. The treatment zone can be, for example, a stent landing zone, a zone for balloon angioplasty or laser atherectomy, etc. In instances where the treatment zone is a stent landing zone, a stent landing zone for a stent can be identified based on the identified lesions.

[0034] By predicting EEL values ​​for image frames for which EEL values ​​could not be determined correctly during pullback, the plaque load for a given frame can be determined. For example, by predicting EEL values ​​for image frames, the plaque load for each image frame acquired during pullback can be determined. This allows for more accurate and complete vascular information compared to ignoring image frames for which EEL could not be determined during pullback. Furthermore, by determining plaque load for image frames along the pullback, the detected and predicted EEL values ​​can be used to more accurately identify the location, length, severity, etc. of a lesion. More accurate lesion identification may result in more accurate identification of a candidate treatment zone. For example, the plaque load and lesion location can be used to determine a candidate treatment zone end frame that does not contain a lesion and / or has a plaque load below a threshold load. This reduces the need for user intervention to identify and / or adjust the treatment zone to a location outside the lesion or a location where the plaque load is below a threshold. According to some examples, the treatment zone can be a stent landing zone. In some examples, the treatment zone can be for balloon angioplasty, etc.

[0035] 1 illustrates a data collection system 100 for use in collecting intravascular and extravascular data. The system may include an intravascular data collection probe 104, an external imaging device 120, and a subsystem 108. The subsystem 108 may include an optical receiver 110, a computing device 112, and a display 118.

[0036] The data collection probe 104 can be, for example, an optical coherence tomography (OCT) probe, an intravascular ultrasound (IVUS) catheter, a micro-OCT probe, a near infrared spectroscopy (NIRS) sensor, an optical frequency domain imaging (OFDI) sensor, or any other device that can be used to image the blood vessel 102. In some examples, the data collection probe 104 can be a pressure wire, a flow meter, or the like. The probe 104 can include a probe tip, one or more radiopaque markers, an optical fiber, and a torque wire. Additionally, the probe tip includes one or more data collection subsystems, such as an optical beam director, an acoustic beam director, a pressure detector sensor, other transducers or detectors, and combinations thereof.

[0037] Although the examples provided herein relate to intravascular imaging devices such as OCT probes, the use of OCT probes is not intended to be limiting. For example, an IVUS catheter, pressure wire, or another intravascular data collection device can be used in conjunction with or in place of an OCT probe. Generally, the present disclosure can apply to any intravascular data collection device that can be used to generate and receive signals containing measurement information, e.g., image data, about the blood vessel in which they are used. These devices can include, without limitation, imaging devices, e.g., optical or ultrasound probes, pressure sensor devices, and other devices suitable for collecting data about blood vessels or other components of the cardiovascular system.

[0038] A guidewire, not shown, can be used to introduce the probe 104 into the blood vessel 102. The probe 104 can be introduced and pulled back along the length of the blood vessel while collecting data.

[0039] The probe 104 can be connected to a subsystem 108 via an optical fiber 106. The subsystem 108 can include a light source such as a laser, an interferometer having a sample arm and a reference arm, various optical paths, a clock generator, a photodiode, and other OCT, IVUS, micro-OCT, NIRS, and / or pressure wire components.

[0040] The probe 104 can be connected to a light receiver 110. According to some examples, the light receiver 110 can be a balanced photodiode-based system. The light receiver 110 can be configured to receive light collected by the probe 104. The probe 104 can be coupled to the light receiver 110 via a wired or wireless connection.

[0041] The system 100 may further include or be configured to receive data from an external imaging device 120. The external imaging device may be, for example, an imaging system based on angiography, fluoroscopy, x-ray, nuclear magnetic resonance, computer-aided tomography, etc. The external imaging device 120 may be configured to non-invasively image the blood vessel 102. According to some examples, the external imaging device 120 may acquire one or more images before, during, or after pullback of the data collection probe 104. The external imaging device 120 may be used to image the patient to make clinical decisions and enable various possible treatment options, such as stent placement. These and other imaging systems may be used to image the exterior or interior of a patient and acquire raw data, which may include various types of image data.

[0042] The external imaging device 120 can communicate with the subsystem 108. According to some examples, the external imaging device 120 can be wirelessly coupled to the subsystem 108 via a communication interface, such as Wi-Fi or Bluetooth. In some examples, the external imaging device 120 can communicate with the subsystem 108 via a wire, such as an optical fiber. In yet another example, the external imaging device 120 can be indirectly communicatively coupled to the subsystem 108 or the computing device 112. For example, the external imaging device 120 can be coupled to a separate computing device (not shown) that communicates with the computing device 112. As another example, image data from the external imaging device 120 can be transferred to the computing device 112 using a computer-readable storage medium.

[0043] Subsystem 108 may include a computing device 112. One or more steps may be performed automatically or without user input, such as navigating an image, entering information, selecting an input and / or interacting with an input. In some examples, one or more steps may be performed based on receiving user input via a mouse click, keyboard, touch screen, verbal command, etc. Computing device 112 may include one or more processors 113, memory 114, instructions 115, data 116, and one or more modules 117.

[0044] The one or more processors 113 may be any conventional processor, such as a commercially available microprocessor. Alternatively, the one or more processors may be dedicated devices, such as application specific integrated circuits (ASICs) or other hardware-based processors. While FIG. 1 functionally depicts the processor, memory, and other elements of device 112 as being within the same block, those skilled in the art will understand that a 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 enclosure. Similarly, memory may be a hard drive or other storage medium located in a different enclosure than that of device 112. Thus, reference to a processor or computing device will be understood to include reference to a collection of processors or computing devices or memories that may or may not operate in parallel.

[0045] The memory 114 may store information accessible by the processor 113, including instructions 115 and data 116 that may be executed by the processor 113. The memory 114 may be any type of memory that operates to store information accessible by the processor 113, including a non-transitory computer-readable medium or other medium that stores data that is readable by an electronic device, such as a hard drive, a memory card, a read-only memory (ROM), a random access memory (RAM), an optical disk, and other writable and read-only memory. The subject matter disclosed herein may include different combinations of the above, whereby different portions of the instructions 115 and data 116 are stored on different types of media.

[0046] Memory 114 may be retrieved, stored, or modified by processor 113 in accordance with instructions 115. For example, although this disclosure is not limited by a particular data structure, data 116 may be stored in a computer register, a relational database as a table with multiple distinct fields and records, an XML document, or a flat file. Data 116 may also be formatted in a computer-readable format, such as, but not limited to, binary values, ASCII, or Unicode. Further by way of example only, data 116 may be stored as a bitmap comprised of pixels stored compressed or uncompressed, or various image formats (e.g., JPEG), vector-based formats (e.g., SVG), or computer instructions for drawing graphics. Furthermore, data 116 may include associated information, such as numbers, descriptive text, proprietary codes, pointers, references to data stored in other memory (including other network locations), or information sufficient to identify information used by a function that computes the associated data.

[0047] The instructions 115 may be any set of instructions executed by the processor 113 directly, e.g., in machine code, or indirectly, e.g., in script. In that regard, the terms "instructions," "application," "steps," and "program" are used interchangeably herein. The instructions may be stored in object code format for direct processing by the processor, or in any other computing device language, including a script or collection of independent source code modules that are interpreted on demand or pre-compiled. The functions, methods, and routines of the instructions are described in more detail below.

[0048] According to some examples, the computing device 112 can receive data from the probe 104 via either a wired or wireless connection. The data can include, for example, intravascular data, such as intravascular image data, pressure data, temperature data, flow data, etc. The data received by the computing device 112 from the probe 104 can be used to determine plaque burden, fractional flow reserve (FFR) measurements at one or more locations along the vessel, calcium angles, EEL detection values, calcium detection values, proximal frames, distal frames, EEL-based metrics, stent or no stent decisions, scores, recommendations for debulking and other procedures, evidence-based recommendations informed by automated detection of target areas or features, stent planning, etc.

[0049] Data acquired by probe 104 and / or external imaging device 120 can be processed by one or more modules to obtain information about the vessel, such as lumen contour, vessel diameter, vessel cross-sectional area, lumen area, EEL value, EEL diameter, EEL arc, lesion location, lesion size, plaque burden, VFR, FFR, landing zone, treatment zone, virtual stent bounded by the landing zone, etc. According to some examples, modules 117 can include an EEL detection module 122, a lumen detection module 123, a lesion detection module 124, and a co-registration module.

[0050] The EEL detection module 122 can automatically detect and measure the EEL diameter in a given intravascular image frame taken during pullback of the vessel. The lumen detection module 123 can automatically detect and measure the lumen diameter in a given intravascular image frame taken during pullback of the vessel. The lesion detection module 124 can automatically detect lesions in a vessel based on vessel data obtained from the probe 104 and / or the external imaging device 120. The side branch detection module 125 can process vessel data obtained from the probe 104 and / or the external imaging device 120 to detect one or more side branches of a vessel.

[0051] The computing device 112 can be adapted to co-register vascular data acquired during pullback of the probe 104 with intravascular and / or extraluminal images. For example, the computing device 112 can be configured to receive and store extraluminal image data, e.g., image data generated by the external imaging device 120 and acquired by a frame grabber. The computing device 112 can be configured to receive and store endovascular image data, e.g., image data generated by the probe 104 and acquired by a frame grabber. In some examples, the computing device 112 can access a co-registration module 121 to co-register the vascular data with the luminal image. The luminal image can be an extraluminal image, e.g., an angiogram, an X-ray, etc. The co-registration module 121 can co-register endovascular data, such as endovascular images, plaque burden, EEL measurements, lumen diameter measurements, pressure readings, virtual flow reserve (VFR), fractional flow reserve (FFR), resting full-cycle ratio (RFR), flow rates, etc., with extraluminal images. In some examples, the co-registration module 121 can co-register endovascular data with endoluminal images, such as endoluminal images captured by an OCT probe, an IVUS probe, a micro-OCT probe, etc.

[0052] In one example, the co-registration module 121 can co-register endoluminal data captured during pullback with one or more extraluminal images. For example, the extraluminal image frames can be pre-processed. Various matrices, such as convolution matrices, Hessian matrices, etc., can be applied on a pixel-by-pixel basis to modify brightness, remove, or otherwise alter a given angiographic image frame. As discussed herein, the pre-processing steps can enhance, modify, and / or remove features of the extraluminal image to improve the accuracy, processing speed, success rate, and other characteristics of subsequent processing steps. Vascular centerlines can be determined and / or calculated. In some examples, the vascular centerlines can be superimposed or otherwise displayed on the pre-processed extraluminal image. According to some examples, the vascular centerlines can represent the trajectory of the collection probe 104 through the vessel during pullback. In some examples, the centerlines may be referred to as traces. Additionally or alternatively, marker bands or radiopaque markers can be detected in the extraluminal image frames. According to some examples, the extraluminal image frames and the data received by the collection probe 104 can be co-registered based on the determined positions of the marker bands.

[0053] According to some examples, the modules may additionally or alternatively include a video processing software module, a pre-processing 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 nodule enhancement software module, a Laplacian of Gaussian filter or transform software module, a guidewire detection software module, an anatomical feature detection software module, a stationary marker detection software module, a background subtraction module, a flange vascularity software module, an image intensity sampling module, a moving marker software detection module, an iterative centerline testing software module, a background subtraction software module, a morphological proximity 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, a fast marching algorithm based software module, a vascular centerline generation software module, a vascular centerline tracking software module, a Hessian software module, an intensity sampling software module, an image intensity overlay software module, and other suitable software modules described herein.According to some examples, the modules can include software, such as pre-processing software, transformations, matrices, and other software-based components used to process image data or to facilitate co-registration of different types of image data by other software-based components in response to a patient trigger, or to otherwise perform annotation of the image data to generate ground truth and other software, modules, and functionality suitable for implementing various features of the present disclosure. Modules can include lumen detection using scanline-based or image-based techniques, stent detection using scanline-based or image-based techniques, indicator generation, apposition bar generation for stent planning, guidewire shadow indicators to prevent confusion due to dissection, side branch and defect data, etc.

[0054] In some examples, the module may be configured to process vascular data acquired by the probe 104 and / or the external imaging device 120 using machine learning algorithms, artificial intelligence, or the like.

[0055] According to some examples, the computing device 112 may include a module, e.g., a classification module, for classifying different features within a blood vessel. The module for classifying different features may be a machine learning system (MLS) implemented by training a classifier to segment or manipulate an image so that its constituent tissues, tissue types, and other regions of interest are detected and characterized based on type or another parameter. In one example, the lumen, intima, media, and plaque are detected and identified as having boundaries corresponding to these different tissues. Training a given MLS / neural network requires teaching the network with known inputs and known outputs.

[0056] Training data can include, for example, ground truth data, along with image data elements that can include various image formats, for example, raw grayscale images, provided as input to a convolutional neural network (CNN) for MLS. MLS is run for a period of time of epochs until the training error is reduced, minimized, or otherwise falls below a threshold. In one embodiment, the period ranges from about 100 epochs to about 1000 epochs.

[0057] After a given MLS / CNN is trained, it can process raw images and output image segmentations containing classified or characterized tissues. In one example, polar coordinate images from the pullback of an intravascular imaging probe are manipulated by the trained MLS to generate classified images. According to some examples, various pre-processing steps / operations can be performed to further improve either training or processing time, or both.

[0058] The preprocessing step can include lumen detection using a previously trained MLS using a training set with annotated lumen regions or segments. The preprocessing step can also be selected to train the network during backpropagation and / or to increase the prediction speed of the trained network. Thus, the preprocessing step can include image data smoothing, circular shifting, cyclic shifting, excluding portions of the image data, e.g., depth data below the noise floor (data elimination can be performed alternately, such as removing every other scanline of the image or every other column), pixel filtering to remove noise and improve region uniformity, and other preprocessing steps.

[0059] In some examples, the preprocessing step can include a normalization step, which can include normalizing the intensity of one or more or all of the image data, ground truth annotations, masks, frames, scanlines, neural network outputs, and other intensity-based data so that the intensities are normalized within a range of about 0 to about 1, or another applicable range. In some examples, the MLS training and prediction is performed on polar coordinate images, and after classification, the polar coordinate view is converted to a Cartesian view, and annotations generated by the trained MLS are displayed on the image using color coding or other suitable indicia or visualization.

[0060] A given cost function can provide a metric for evaluating the output of a machine learning system by comparing a ground truth input / training set with the expected output when operating on patient data. The goodness of fit between the training data and the output data can be measured by the cost function. The output of the cost function can be a value corresponding to an error metric. This output is a comparison metric, such as a difference or distance, that summarizes how successful the machine learning system or neural network or other operating component thereof is in terms of accurate predictions given the predicted results and the ground truth used to train the system. If the output result of the cost function is zero, the system is effectively operating perfectly. In this way, iterative modifications to the system can be used to reduce the system's cost or error function and improve its prediction accuracy.

[0061] Additionally, a pixel-wise cost function is specified to measure the distance between the prediction and the ground truth or another suitable metric or score. In some examples, backpropagation involves updating each of the weights in the network base based on values ​​derived from the cost function. In one embodiment, partial derivatives of the cost function are used to update the weights during backpropagation. This weight update process has the advantage that the actual prediction results are closer to the ground truth. This has the advantage of reducing or minimizing the error for each output neuron or node of the neural network.

[0062] In some examples, annotated mask regions corresponding to sets or groups of pixels define a ground truth mask used to train one or more neural networks disclosed herein. Once the neural network is trained, a predicted or detected mask is generated. This mask includes a set of pixels corresponding to regions of the user data as well as an identifier of the region's feature or class, such as luminal, calcium, EEL, or another class or feature disclosed herein. In one example, predicted results are generated for each class, and then all of the predicted results for a given input image data frame, such as OCT, IVUS, or other image data, are compared pixel-by-pixel to generate a final predicted result for all classes. In one example, the predicted results are displayed as an output image mask with regions corresponding to specific classes indicated by a color or other indicator and one or more legends summarizing which indicators map to which classes.

[0063] According to some examples, the classification module can classify the type of plaque present. For example, the classification module can classify the plaque as calcified. Additionally, given that the presence of plaque and other detectable features in a given portion of an artery may indicate the presence of a narrowing resulting from a stenosis or the like, another feature of the present disclosure can be to quickly and automatically obtain one or more scores associated with a given plaque or stenosis to help facilitate decision-making by an end user. For example, a given score determined using image data and its machine learning-based analysis can help determine whether immediate action is not recommended, whether a stent should be placed for the stenosis, or whether other procedures, such as atherectomy or bypass, are warranted.

[0064] In healthy patients, arteries have various layers, including the intima, media, and adventitia, arranged in a consistent structure. As a result of the atherosclerosis process, the intima becomes pathologically thickened and can contain plaques composed of different tissue types, including fibers, proteoglycans, lipids, and calcium, as well as macrophages and other inflammatory cells. These tissue types have different characteristics when imaged using various imaging systems, which can be used to establish training data sets for one or more of the machine learning systems disclosed herein. The plaques considered to be most pathologically significant are so-called vulnerable plaques, which consist of a fibrous cap with an underlying lipid pool. Different atherosclerotic plaques have different geometric shapes. For example, foam cells typically form ribbon-like features on the shoulders of larger lipid pools. The tunica media can appear as a ring around the vessel. Shape information is currently used in the qualitative assessment of OCT images. In one example, a neural net is trained to identify fibrous caps and / or fibrous caps with underlying lipid pools. In various instances, references to calcium herein include, without limitation, calcified plaque and other calcium-containing tissues.

[0065] The ability to quickly perform an imaging procedure on a patient to obtain arterial images and process the images using a machine learning system while the patient is still catheterized and ready to receive a stent or other treatment option results in significant time savings and improvements in patient outcomes. The classification module can be similar to the classification method described in U.S. Patent No. 11,250,294, entitled "SYSTEMS AND METHODS FOR CLASSIFICATION OF ARTERIAL IMAGE REGIONS AND FEATURES THEREOF," the contents of which are incorporated herein by reference in their entirety.

[0066] The subsystem 108 may include a display 118 that outputs content to a user. The display 118 may be integrated with the computing device 112 or may be a stand-alone unit electronically coupled to the computing device 112. The display 118 may output intravascular data related to one or more features detected in the vessel and / or acquired during pullback. For example, the output may include, but is not limited to, a three-dimensional representation generated based on cross-sectional scan data, long-axis scans, endoluminal and / or extraluminal images, diameter graphs, image masks, lumen boundaries, plaque size, plaque circumference, visual indicia of plaque location, visual indicia of risk posed to stent expansion, flow rate, recommended treatment zones, etc. The display 118 may identify features using text, arrows, color coding, highlighting, outlines, or other suitable human- or machine-readable indicia.

[0067] According to some examples, the display 118 can include a graphic user interface (GUI). The display 118 can be a touchscreen display that can provide input for a user to navigate through images, enter information, select inputs, and / or interact with inputs, etc. In some examples, the display 118 and / or the computing device 112 can include input devices, such as a trackpad, mouse, keyboard, etc., that can enable a user to navigate through images, enter information, select inputs, and / or interact with inputs, etc. The display 118, alone or in combination with the computing device 112, can enable toggling between one or more viewing modes in response to user input. For example, a user can toggle between different endovascular data, images, etc., recorded during each of the pullbacks.

[0068] In some examples, the display 118, alone or in combination with the computing device 112, can present one or more menus as output to the physician, and the physician can responsively provide input by selecting items from the one or more menus. For example, the menus can allow the user to show or hide various features. As another example, there can be a menu for selecting which vascular features to display.

[0069] The output can include image data, such as cross-sectional images acquired by the probe 104 during pullback and / or external images acquired by the external imaging device 120. In some examples, the output can include a longitudinal representation of the vessel, a graphical representation of the vessel data, a three-dimensional representation of the vessel and / or vessel data, etc. The vessel representation can provide various viewing angles and cross-sectional views. In some examples, the position of the EEL, its diameter, or other EEL-based parameters can be output for display in relation to angle measurements, detected calcium arcs, plaque burden, etc.

[0070] In some examples, one or more visual representations of the image may include an indication of a lesion location, lesion severity, lesion length, etc. Additionally or alternatively, the lesion indications may be color-coded, with each color representing a severity, length, or other measurement associated with the lesion.

[0071] In some examples, the output can include candidate treatment zones. The candidate treatment zones can be, for example, candidate stent landing zones. For example, the output can include a representation corresponding to a candidate proximal landing zone for the stent and a candidate distal landing zone for the stent. The candidate treatment zones can be determined based on determined plaque burden, lesion location, lesion length, etc. The representation can be provided on any of the vessel representations, e.g., a three-dimensional representation, a longitudinal representation, a graphical representation, image data, e.g., an external image, etc.

[0072] According to some examples, the display 118 and / or the computing device 112 can be configured to receive one or more inputs corresponding to a selection on one or more representations. For example, an input corresponding to a selection of an image frame on a longitudinal representation can be received. In response, other representations provided for output can be updated to display corresponding views or image frames. For example, the extraluminal image can be updated to have a view along the vessel corresponding to the position of the selected image frame on the longitudinal representation, a circumferential view can be provided on the three-dimensional representation corresponding to the position of the selected image frame on the longitudinal representation, and a cross-sectional image frame can be updated to correspond to the selected image frame on the longitudinal representation. In some examples, vascular data associated with the selected position can be updated and provided for display.

[0073] Figure 2A is an example image frame acquired during pullback of the probe 104 of Figure 1. For example, multiple image frames can be acquired during pullback of the probe 104. The image frame 200A can be an OCT image frame, an IVUS image frame, a NIRS image frame, an OFDI image frame, etc.

[0074] Image frame 200A can be processed using one or more modules 117 to determine vascular data. Vascular data can include, for example, vascular measurements, such as EEL measurements, lumen measurements, plaque burden, calcium burden, intimal thickness, medial thickness, etc. For example, EEL detection module 122 can automatically determine EEL values ​​for image frames, including image frame 200A. Lumen detection module 123 can automatically determine lumen area for image frames, including image frame 200A. Lesion detection module 124 can automatically determine plaque burden for image frames, including image frame 200A. The plaque burden for a given image frame can, in some examples, be determined based on the EEL value and lumen area determined for the given image frame.

[0075] 2B is an example annotated image frame. Image frame 200B can be similar to image frame 200A in that image frame 200B could also be acquired during pullback of probe 104. Thus, image frame 200B can be an OCT image frame, an IVUS image frame, a NIRS image frame, an OFDI image frame, etc.

[0076] Image frame 200B can be processed similarly to image frame 200A so that vascular measurements can be determined. In some examples, an image frame, such as image frame 200B, can be processed to identify various features within image frame 200B. For example, image frame 200B can be processed to identify probe 104, vessel wall W, adventitia AD, lumen L, etc. In some examples, image frame 200B can be processed to identify EEL, plaque, calcium plaque, calcium angle, etc. According to some examples, output by display 118 can include an indication of the identified vessel wall W, adventitia AD, lumen L, calcium angle, plaque burden, etc.

[0077] The lumen L can be identified by the lumen detection module 123. In some examples, the lumen L can be automatically identified and used to determine the lumen area of ​​a blood vessel in a given image frame. The lumen area can be used to determine the plaque burden in the image frame. For example, an EEL value for a given frame can be determined by the EEL detection module 122, and the plaque burden for the frame can be determined as plaque load = 1 - (lumen area / EEL value), where the EEL value can be, for example, the EEL area or the EEL diameter.

[0078] In some examples, an EEL value cannot be automatically determined for a frame, such as when the plaque load reduces the visibility of the EEL. In other examples, the plaque load may be greater than a threshold. For example, if the plaque load is greater than a threshold, the plaque load may be considered unhealthy. In some examples, there may be measurement error in the EEL value, such as when the EEL is not sufficiently visible to be measured. In examples where an EEL value cannot be determined or where the plaque load is greater than a threshold, an EEL value can be predicted to determine the plaque load for a given frame.

[0079] FIG. 2C shows an exemplary longitudinal representation of a blood vessel. The longitudinal representation 200C can be generated based on image frames acquired by the probe 104 during pullback, e.g., by the computing device 112 of FIG. 1 . The longitudinal representation 200C can include a designation "P" corresponding to the proximal end of the pullback and a designation "D" corresponding to the distal end of the pullback. According to some examples, the longitudinal representation 200C can include a designation corresponding to the location of a side branch "SB" along the blood vessel. The longitudinal representation 200C can be color-coded, shaded, or include some type of designation between different tissue types within the blood vessel. For example, areas of calcium "Ca" can be color-coded with a first color, and areas without calcium can be a different color.

[0080] The longitudinal representation 200C may include an indication of the image frame or region where the image frame contains an EEL value. For example, the longitudinal representation 200C may include dashed lines along the length of the representation indicating the frame or region where an EEL value was measured or detected. According to some examples, the longitudinal representation 200C may include regions where an EEL value was not measured and / or regions where the plaque burden was above a threshold. If the EEL was not visible enough to be measured, the EEL value may not have been measured or may have been inaccurate. For example, regions "ME1" and "ME2" may have a plaque burden above a threshold. The absence of dotted lines within regions ME1 and ME2 may indicate that plaque may have overgrown in these regions ME1 and ME2, such that EEL is not detected and / or is not detected at a level that meets a threshold indicating an EEL value. The EEL value may be, for example, EEL area and / or EEL diameter. In some examples, the absence of dotted lines representing EEL regions may indicate that an EEL value for these image frames could not be determined.

[0081] According to some examples, regions of interest can be identified. The regions of interest 230, 232, 234 can be regions of the vessel between the side branches SB. The regions of interest 230, 232, 234 can include subregions 236, 238, 240 where the EEL value was not determined and / or was below a threshold. In such regions 236, 238, 240, the EEL value can be predicted.

[0082] According to some examples, detected or determined EEL values ​​within regions of interest 230, 232, 234 can be used to predict EEL values ​​in the respective sub-regions 236, 238, 240. For clarity herein, the disclosure is directed to region of interest 232 and sub-region 238. However, the methods described herein can also be applied to regions of interest 230, 234 and their respective sub-regions 236, 240. Thus, the use of region of interest 232 and sub-region 238 is merely exemplary and not intended to be limiting.

[0083] The region of interest 232 can be located between two side branches SB. One or more modules 117 can be used to process image frames within the region of interest 232 acquired during pullback. For example, EEL values, lumen area values, and plaque burden can be automatically determined. As previously mentioned, EEL values ​​may not be determined within the subregion 238 and / or plaque burden may reduce the visibility of the EEL, making the EEL values ​​unsuitable for use in determining plaque burden. In such instances, EEL values ​​for image frames within the subregion 238 can be predicted.

[0084] In some examples, the predicted EEL value can be determined based on EEL values ​​determined for the remainder of the image frames within region 232. For example, region 232 can include multiple image frames. The frames of a first subset can have a plaque load less than a threshold plaque load. EEL values ​​can be determined for at least the image frames of the first subset. The image frames of a second subset, e.g., those within subregion 238, have a plaque load greater than the threshold plaque load. In some examples, the EEL values ​​for the frames of the second subset can be EEL values ​​that are erroneously measured due to plaque load or that are undetectable due to plaque load. In examples where the image frames of the first subset include a threshold number of frames with plaque loads below the threshold plaque load, EEL values ​​for the threshold number of frames can be determined. A representative EEL value can be determined based on the EEL values ​​of the image frames in the first subset with plaque loads below the threshold plaque load. The threshold plaque load can be, for example, between about 0% and about 50%. For example, the threshold plaque burden can be about 40%, about 45%, etc. In some instances, the threshold plaque burden can be greater than or less than about 0% to about 50%. In other instances, the threshold plaque burden can be a range, patient-specific, physician-determined, etc.

[0085] According to some examples, a representative EEL value for the image frames in the first subset of image frames can be used as the EEL value for image frames in the first subset of image frames that have EEL values ​​with measurement errors. For example, some of the image frames in the first set of image frames can have EEL values ​​that are outside a predetermined range of the representative EEL value, such as more than a standard deviation from the representative EEL value. In such examples, e.g., examples where there is measurement error in the EEL values, the representative EEL value for the first set of frames can be used as the EEL value for the frames with measurement errors.

[0086] The threshold number of frames can be determined based on the number of frames in the region of interest. For example, for a larger region of interest, the threshold number of frames may be greater than the threshold number of frames for a smaller region of interest. In some examples, the threshold number of frames can be three image frames, regardless of the size of the region of interest.

[0087] The predicted EEL value and lumen area can be used to determine plaque burden for frames where an EEL value was not automatically determined, frames where the plaque burden was greater than a threshold plaque load, frames where the EEL value was measured inaccurately, etc. In such instances, the plaque burden for a given frame can be determined as 1 - (lumen area / predicted EEL value). In this formula, the predicted EEL value can be the predicted EEL area or the predicted EEL diameter, or both.

[0088] According to some examples, the predicted EEL value can be determined as a function of the intimal and medial thicknesses. The intimal thickness can be, for example, the minimum intimal thickness, and the medial thickness can be, for example, the minimum medial thickness. The minimum intimal and medial thicknesses can be determined based on representative blood vessels. For example, representative blood vessels with plaque burden values ​​less than a threshold value can be used to determine average values ​​for the minimum intimal and medial thicknesses. In one example, the combined minimum intimal and medial thickness can be 0.237 mm. In some examples, the respective minimum intimal and medial thicknesses can be 0.198 mm. The combined minimum intimal and medial thickness of 0.237 mm and the respective minimum intimal and medial thicknesses of 0.198 mm are merely some examples and are not intended to be limiting. For example, the combined thickness can be greater than or less than about 0.237 mm, and the respective thicknesses can be greater than or less than about 0.198 mm. In some examples, the respective minimum intimal and medial thicknesses can be the same value, or they can be different values. According to some examples, the minimum intimal and medial thicknesses can be based on histological observation of the blood vessel. For example, based on histological observation of the blood vessel, the intimal and medial thicknesses can be determined by standard deviation. The minimum intimal and medial thicknesses can be determined based on the lower limit of the standard deviation. For example, if the combined intima and media thickness is determined to be 427 micrometers ± 190 micrometers, the minimum combined intima and media thickness is 237 micrometers. The combined intima and media thickness may vary based on the study dataset size, the patients participating in the study, etc. Therefore, a combined minimum intimal and medial thickness of 0.237 mm is used, although the combined minimum intimal and medial thickness can be any value determined by histological observation.

[0089] The predicted EEL value can further be determined as a function of an HK model, which predicts an expected lumen diameter. An assumption of the HK model is that the size of the vessel within the region of interest 232, e.g., the size of the vessel between side branches, is substantially constant. For example, the HK model predicts the lumen diameter based on the assumption that the diameter, circumference, and / or area of ​​the vessel are substantially constant along the region of interest. According to some examples, the HK model can provide a prediction of what the lumen diameter should be for a vessel in a given image frame if the vessel were healthy. A healthy vessel can be one with a plaque load below a threshold plaque load, a lumen diameter above a threshold diameter value, etc.

[0090] According to some examples, the predicted EEL can be determined using the following formula: Predicted EEL value = (expected lumen diameter) + 2 * (minimum intimal thickness + minimum medial thickness)

[0091] where the predicted EEL value can be the predicted EEL area and / or the predicted EEL diameter, and the predicted lumen diameter can be, for example, the lumen diameter determined using the HK model.

[0092] Because the EEL is outside the lumen, additional distance can be added in the form of minimum intimal and medial thickness to predict the EEL value. For example, the minimum intimal and medial thickness can compensate for the additional width between the lumen and the EEL. The minimum intimal and medial thickness can be added together and multiplied by two to compensate for the additional width on either side of the vessel. The predicted EEL value can be determined by adding twice the minimum intimal and medial thickness to the expected lumen diameter predicted by the HK model.

[0093] The predicted EEL value and lumen area can be used to determine the plaque burden for frames where the EEL value was not automatically determined and / or was determined inaccurately. In such instances, the plaque burden for a given frame can be determined as 1 - (lumen area / predicted EEL value).

[0094] Figure 3 illustrates an exemplary user interface for use on the display 118 of Figure 1. The user interface 300 may include an upper interface portion "UP" and a lower interface portion "LP." The upper interface portion UP may include image frames, such as cross-sectional images of the vessel taken during pullback, and the lower interface portion LP may include a longitudinal representation 350 of the vessel. The longitudinal representation 350 may be generated using image frames captured by the probe 104 during pullback.

[0095] In the upper part UP of the interface 300, there may be three image frames corresponding to an image frame or a particular view or slice of the vessel: a view in the proximal reference frame 151, a selected frame 153 corresponding to the image frame having the diamond 165 in the longitudinal representation 350, and a view in the distal reference frame 155.

[0096] The proximal and distal reference frames 151 and 155 indicate the lumen L of the blood vessel in the image frames by one or more dotted lines passing through the lumen L. These values ​​obtained from measuring these lines can be applied to the measured or detected EEL locations, points, or pixels. According to some examples, these values ​​can be used to generate a measured EEL value, e.g., an EEL diameter, or an average EEL diameter. Some exemplary EEL diameter measurements are shown as approximately 3.8 mm (proximal) and approximately 3.4 mm (distal). According to some examples, the measured EEL value can be used to predict an EEL value for one or more frames for which an EEL value cannot be determined.

[0097] The lower portion LP of the interface 300 may show a combination of calcium "Ca" and EEL as a longitudinal representation 350. The longitudinal representation 350 may be similar to the longitudinal representation 200C in that the longitudinal representation 350 may include an indication of one or more side branches SB, EEL values, calcium area, etc. According to some examples, the interface 300 may include an indication of one or more regions of interest 332, e.g., a region of a vessel located between two side branches. In some examples, the interface 300 may include an indication of a subregion 338 in which EEL values ​​for image frames within the subregion could not be determined. The EEL value for image frames within the subregion 338 may be predicted as the median EEL value within the region 332 if there is a threshold number of image frames with a plaque load less than a threshold plaque load and / or if a visible EEL arc is equal to or greater than an arc threshold. The arc threshold may be, for example, at least 180 degrees or approximately 180 degrees. The threshold plaque load may be, for example, a healthy plaque load. A healthy plaque burden, in some instances, can correspond to a plaque burden of 0% to 50%. In some instances, a healthy plaque burden can correspond to a plaque burden of 35% to 50%. In other instances, a healthy plaque burden can correspond to a range of plaque burden greater than 50%. Thus, 0% to 50% and 35% to 50% plaque burden are merely some examples and are not intended to be limiting.

[0098] In some examples, the EEL values ​​for the image frames in subregion 338 can be predicted as a function of the HK model, the minimum intimal thickness, and the minimum medial thickness. For example, if there are not enough image frames with plaque burden below the threshold plaque burden, the EEL values ​​for the image frames can be predicted as a function of the HK model, the minimum intimal thickness, and the minimum medial thickness.

[0099] According to some examples, the predicted EEL value determined as a function of the HK model can be compared to the EEL value of a proximal segment of the blood vessel and / or a distal segment of the blood vessel. The proximal segment and / or distal segment of the blood vessel can be a segment of the blood vessel adjacent to the region of interest or further proximal and / or distal to the region of interest. For example, there may be a distance between the proximal segment and the region of interest and / or between the distal segment and the region of interest. The EEL value of the proximal segment and / or distal segment of the blood vessel can, in some examples, be a representative EEL value of the proximal segment and / or distal segment of the blood vessel. If the predicted EEL value is less than the representative EEL value of the distal segment, the representative EEL value of the distal segment can be used to determine the plaque burden of the frame within the region of interest. Because the HK model assumes that the blood vessel has a natural taper from the proximal to the distal vessel segment, if the predicted EEL value is less than the representative EEL value of the distal segment, the representative EEL value of the distal segment can be used. If the predicted EEL value is greater than the representative EEL value of the proximal segment, the representative EEL value of the proximal segment can be used to determine the plaque burden of the frame within the region of interest. Because the HK model assumes that the vessel has a natural taper from the proximal to the distal vessel segment, if the predicted EEL value is greater than the representative EEL value of the proximal segment, the representative EEL value of the proximal segment can be used.

[0100] 4 illustrates another exemplary user interface for use on display 118 of FIG. 1. User interface 400 may be similar to user interface 300, having an upper portion "UP" and a lower portion "LP." The difference between interface 300 and interface 400 may be the representation output for display. For example, longitudinal representation 350 provided for display in interface 300 may be a lumen profile, while longitudinal representation 450 provided for display in interface 400 may be a transverse representation of the vessel, e.g., an L-view representation.

[0101] As shown, the upper portion UP of the interface 400 includes three frames, with a user-selectable frame in the center. Upon receiving input corresponding to the selection and / or movement of the "US" marker, the user-selected frame 153 may be updated. Additionally, the vascular data associated with the user-selected frame 153 may be updated to correspond to the selected frame. This may allow for different EEL and lumen diameters to be considered for the proximal and distal fiducials 151 and 155. Bookmarks BKM are also displayed that the user can set using the GUI to allow for quick navigation between frames. Another marker is displayed indicating the frame with the minimum lumen diameter MLD; in other cases, MLA or minimum lumen area may be displayed.

[0102] FIG. 5 illustrates an exemplary longitudinal representation of a blood vessel that may be generated using computing device 112. According to some examples, lesions may be automatically identified using longitudinal representation 500 and / or the data used to generate longitudinal representation 500. For example, lesions may be identified based on the plaque load for each frame. The plaque load may be determined as a function of the EEL value and lumen area for a given frame. In some examples, a determined EEL value may be used, and in some examples, a predicted EEL value may be used to determine the plaque load. The plaque load may be determined as 1-(lumen area / EEL value), where the EEL value is the determined or predicted EEL value for the image frame. Frames with plaque loads above a threshold plaque load may be marked or displayed as part of a lesion. The threshold plaque load may be, for example, 70% plaque load. In some examples, the threshold plaque load may be, for example, 25%, 50%, or the like. The threshold plaque load may be patient-specific. In some instances, the threshold plaque burden can be adjusted by a physician to correspond to any value or percentage of plaque burden, and thus the 70% example is merely illustrative and not intended to be limiting.

[0103] Two distinct lesions L1, L2 can be identified as shown in longitudinal representation 500. According to some examples, based on the proximity of each lesion L1, L2 to one another, lesions L1, L2 can be identified as a single lesion LC.

[0104] To identify lesion L1, an image frame M1 in which the plaque load is greater than a threshold plaque load can be identified. According to some examples, an image frame with the greatest plaque load can be identified. In some examples, an image frame in which the plaque load is greater than a disease threshold can be identified. The plaque load of each frame proximal and distal to the identified image frame can be determined and compared to a treatment zone threshold until the plaque load of the proximal image frame P1 and the distal image frame D1 are identified as being less than the treatment zone threshold. The treatment zone threshold can be, for example, about 0% to about 50% plaque load. A treatment zone end frame can also be defined as a frame P1, D1 in which the plaque load is less than the treatment zone threshold and / or the visible arc of the EEL is greater than or equal to the arc threshold. The arc threshold can be, for example, at least 180 degrees or about 180 degrees. Lesion L1 can be identified as extending between the identified proximal image frame P1 and the distal image frame D1.

[0105] This process can be repeated until all lesions, e.g., lesion L2, have been identified for the pullback. For example, the image frame with the next highest plaque burden can be identified from the remaining image frames, e.g., from the image frame other than the image frame identified as lesion L1. For example, the next highest plaque burden can be M2. According to some examples, an image frame with a plaque burden greater than a disease threshold can be identified from the remaining image frames. Similar to identifying lesion L1, lesion L2 can be identified by comparing the plaque burdens of the proximal and distal image frames P2 and D2 until a proximal image frame P2 and a distal image frame D2 are identified in which the plaque burden is less than the treatment zone threshold and / or the visible EEL arc is greater than or equal to the arc threshold. The arc threshold can be, for example, at least 180 degrees or approximately 180 degrees. Although only two lesions L1 and L2 are identified on the longitudinal representation 500, more or fewer lesions can be identified for a given pullback.

[0106] According to some examples, after one or more separate lesions L1, L2 are identified, the lesions can be combined into a single lesion LC. For example, if the distance between the separate lesions is less than a threshold, the one or more separate lesions can be combined into a single lesion. The distance between the lesions can be measured based on the number of image frames between the distal end of a first lesion and the proximal end of another lesion. In some examples, the distance between the lesions can be measured in millimeters or any other form of distance measurement. The threshold distance can be patient-specific, updated by a physician, preset, etc. In some examples, the threshold distance between the lesions can be about 5 mm to about 10 mm. According to some examples, the threshold distance can be about 8 mm to about 10 mm. In other examples, the threshold distance can be less than about 5 mm or greater than about 10 mm. In some examples, the threshold distance can be patient-specific, physician preference, etc. Therefore, it should be noted that the threshold distance of about 5 mm to about 10 mm is merely an example and is not intended to be limiting.

[0107] Using lesions L1 and L2 in FIG. 5 as an example, the distance between the distal image frame D1 of lesion L1 and the proximal image frame P2 of lesion L2 can be determined. If the distance between the distal image frame D1 of lesion L1 and the proximal image frame P2 of lesion L2 is less than a threshold distance, lesions L1 and L2 can be combined. As shown, the distance between lesion L1 and lesion L2 is less than the threshold. Therefore, lesions L1 and L2 are combined into a single lesion LC.

[0108] Combining lesions with gaps or distances less than a threshold distance can prevent the treatment zone end frame from being located in the gap between the lesions. For example, by combining separate lesions into a single composite lesion, the system can automatically identify treatment zone end frames at locations proximal and distal to the composite lesion, rather than in the gap between the two lesions.

[0109] FIG. 6 illustrates another exemplary user interface that can be provided for output on the display 118. The user interface 600 illustrates candidate treatment zone end frames "TZ." The candidate treatment zone end frames TZ can be located proximal and distal to the identified lesion LC. For example, the candidate treatment zone end frames TZ for a stent can be determined based on a comparison of the plaque burden of image frames proximal and distal to the lesion LC with a treatment zone threshold. The treatment zone threshold can be, for example, a plaque burden of about 0% to about 50%. If image frames proximal and distal to the lesion LC are identified as having a plaque burden substantially corresponding to or less than the treatment zone threshold and / or having a visible EEL arc equal to or greater than the arc threshold, the image frames can be identified as candidate treatment zone end frames for the stent. The arc threshold can be, for example, at least 180 degrees or about 180 degrees. The candidate treatment zones can be provided for output on a longitudinal representation 650 of the vessel or any other co-registered representation. For example, a representation of the candidate treatment zone end frame TZ can be provided for output onto a co-registered external image frame 660, for example an angiography image frame.

[0110] The interface 600 may include a cross-sectional representation of a blood vessel, e.g., a cross-sectional image frame of a blood vessel. The cross-sectional image 662 may include indicia identifying the location of the EEL, e.g., a dashed line, and indicia identifying the lumen "L," e.g., a solid line. Corresponding vessel data, e.g., an EEL value and a lumen diameter value, may be provided for output along with the cross-sectional image 662. For example, the cross-sectional image 662 may correspond to an image frame 662 selected on the longitudinal representation 650. As shown, the selected image frame 662 has an EEL diameter of 2.61 mm and a lumen diameter of 2.50 mm. According to some examples, the interface 600 may include a representation of the EEL arc. For example, the interface 600 may output a representation of the EEL arc on the intravascular image, a longitudinal representation, etc. In some examples, the interface 600 may provide an output of the EEL arc as a value, such as a percentage.

[0111] These EEL and lumen values ​​can be used to help inform stent selection or other treatment devices. For example, based on the EEL and lumen values, a physician may then be able to assess candidate treatment zones TZ that have been automatically suggested by the system. According to some examples, co-registering the longitudinal representation with the extraluminal image so that the candidate treatment zones are displayed on both the longitudinal representation and the extraluminal image can aid in assessing and evaluating the candidate treatment zones.

[0112] According to some examples, the cross-sectional representation may include an indication of plaque burden, calcium burden, etc. For example, the cross-sectional representation may include arcs identifying regions of plaque burden, calcium burden, etc. In some examples, the representation may be color-coded to identify different classes identified in the cross-sectional representation. Class types can include, for example, intima, media, adventitia, lumen, external elastic lamina (EEL), internal elastic lamina (IEL), plaque, calcium, calcium plaque, side branch, guidewire, stent strut, stent, jailed stent, bioresorbable vascular scaffold (BVS), drug eluting stent (DES), fibrous, blooming artifact, pressure wire, lipid, atherosclerotic plaque, stenosis, calcium, calcified plaque, calcium-containing tissue, lesion, fat, malapposed stent, underexpanded stent, overexpanded stent, radiopaque marker, arterial tree branching angle, probe calibration element, doped film, light scattering particle, sheath, doped sheath, fiducial registration point, diameter measurement, radial measurement, guide catheter, shadow area, guidewire segment, length, or thickness.

[0113] Additionally or alternatively, the display may be color coded to correspond to the severity of plaque burden, lesions, calcium accumulation, etc. Color coding provided for output on cross-sectional representations may, in some instances, be provided for output on longitudinal representations and / or extraluminal images.

[0114] According to some examples, interface 600 can include a three-dimensional (3D) representation of a blood vessel including the region of interest. The 3D representation can be in addition to or in place of one of the other representations of the blood vessel. For example, interface 600 can include one or more of an extraluminal image, a cross-sectional image frame, a longitudinal representation, a 3D representation, etc. In some examples, the longitudinal representation can be symmetric about a longitudinal axis of the blood vessel and / or representation. In another example, the longitudinal representation can be a carpet view of the blood vessel. Each representation of the blood vessel, e.g., the extraluminal image, the cross-sectional image, the 3D representation, the longitudinal representation, etc., can be configured to be color-coded, include annotations and / or indicators, etc.

[0115] FIG. 7 illustrates another exemplary user interface that can be provided for output on the display 118. The user interface 700 can be used to support stent planning through virtual stent placement. In some examples, the virtual stent placement can correspond to an automatically identified treatment zone end frame TZ based on a detected lesion. A depiction or display of the virtual stent can be provided on one or more of the representations of the blood vessel. For example, similar to the interface 600, the interface 700 can include one or more representations of the blood vessel, including an extraluminal image 770, a cross-sectional image 772, and a longitudinal representation 750. A display of the virtual stent can be provided for output on the extraluminal image 770 and / or the longitudinal representation 750.

[0116] According to some examples, interface 700 can output vascular data, such as EEL values, lumen values, VFR values, MLA values, percent diameter stenosis, etc. Interface 700 can include pre-stent values ​​and predicted post-stent values. For example, the pre-stent VFR value can be 0.75, while the predicted post-stent VFR value can be 0.90. Similarly, although not shown, interface 700 can provide for output pre-stent EEL values, lumen diameter values, etc., and corresponding predicted post-stent values. The predicted values ​​can be determined using machine learning or neural networks trained for predictive analytics.

[0117] 8 illustrates a flow diagram of an example process for automatically determining plaque burden for frames in a region of interest that may be implemented using the computing device 112. The following operations do not have to be performed in the exact order described below. Rather, various operations may be processed in a different order or simultaneously, and operations may be added or omitted.

[0118] At block 810, vascular data including intravascular image data may be received. The intravascular image data may be, for example, one or more image frames acquired by the probe during pullback through a blood vessel. The one or more image frames may be OCT image frames, IVUS image frames, etc. The vascular data, in some examples, may be values ​​or measurements associated with a blood vessel in a given image frame. For example, the vascular data may be an EEL value, e.g., EEL diameter, a lumen value, e.g., lumen area or diameter, etc. In some examples, the vascular data may include a pressure value, a VFR value, an IMR value, etc.

[0119] At block 820, a region of interest in the vessel may be identified based on the vessel data. For example, the vessel data may include a representation of one or more side branches. The region of interest may be defined as the area between a first side branch and a second side branch.

[0120] At block 830, a first subset of frames within the region of interest having a plaque burden below a threshold plaque burden may be identified. The threshold plaque burden may be, for example, between 0% and 45%. The first subset of frames may include a threshold number of frames. The threshold number of frames may be at least three frames.

[0121] At block 840, an EEL value may be determined for the first subset of frames. According to some examples, the EEL value may be determined for each frame of the first subset of frames. In another example, the EEL value may be determined for a threshold number of frames of the first subset of frames. The EEL value for the first subset of frames may be determined automatically based on the vascular data. The EEL value may be, for example, an EEL diameter and / or an EEL area.

[0122] In block 850, a representative EEL value for the region of interest may be determined based on the EEL values ​​of the frames of the first subset. The representative EEL value for the region of interest may be, for example, any or all of a median EEL value, an average EEL value, a mean EEL value, or a mode EEL value. For example, a median EEL value of the frames of the first subset may be determined. The determined representative EEL value of the frames of the first subset may be used as a predicted EEL value for a second subset of frames within the region of interest. According to some examples, the determined representative EEL value of the first set of frames may be used as a predicted EEL value for some of the frames in the first subset of frames. For example, if some of the frames in the first subset of frames have EEL values ​​with measurement errors, the determined representative EEL value may be used as a predicted EEL value for the frames in the first subset of frames with the erroneous EEL values. The erroneous EEL value may be an EEL value that exceeds the standard deviation of the determined representative EEL value, an EEL value that falls outside a predetermined range of the determined representative EEL value, etc.

[0123] The second subset of frames can be different from the first subset of frames. The second subset of frames can be frames within the region of interest for which EEL values ​​were not automatically determined during pullback. In some examples, the second subset of frames can be frames within the region of interest for which the EEL values ​​have measurement error, are not visible, etc.

[0124] At block 860, the plaque burden for the frames within the region of interest can be determined based on the representative EEL value. The plaque burden can further be based on the lumen area for each frame. For example, the plaque burden can be determined for each frame using the formula 1-(lumen area / representative EEL value). In this formula, the representative EEL value can be the representative EEL area and / or the representative EEL diameter.

[0125] According to some examples, the plaque burden for each frame can be used to identify lesions in a blood vessel. For example, a lesion can be identified by identifying an image frame with a maximum plaque burden. According to some examples, a lesion can be identified by identifying an image frame in which the plaque burden is greater than a disease threshold. The disease threshold can be, for example, a plaque burden disease threshold. The plaque burden disease threshold can be, for example, approximately 70%. In some examples, the plaque burden disease threshold can be greater or less than 70%. For example, the plaque burden disease threshold can be patient-specific, physician-determined, etc.

[0126] Using the identified image frame as an initial frame, the plaque loads of frames proximal and distal to the initial frame can be compared to the treatment zone threshold until the plaque loads of the image frames proximal and distal to the initial frame are less than the treatment zone threshold. The treatment zone threshold, in some examples, can be greater than or less than a 50% plaque load. For example, the treatment zone threshold can correspond to a plaque load of approximately 45%. In some examples, the treatment zone threshold can range from approximately 0% to approximately 50% plaque load, etc. In some examples, the treatment zone threshold can be patient-specific, physician-determined, etc. Additionally or alternatively, the treatment zone end frame can correspond to a frame in which the visible EEL arc is equal to or greater than the arc threshold. In some examples, the arc threshold can be at least 180 degrees or approximately 180 degrees. In some examples, the arc threshold can be greater than or less than 180 degrees. For example, the arc threshold can be patient-specific, physician-determined, etc., and the arc threshold can be any predetermined value.

[0127] The image frame between the identified proximal and distal frames is then marked as corresponding to the lesion. The process is repeated for the image frame with the next highest plaque burden, without considering the image frame identified as corresponding to the lesion. After each lesion is identified, the gap or distance between the lesions is determined. In instances where the gap is substantially equal to or less than a distance threshold, the lesions can be combined and identified as a single lesion. For example, if a first lesion is within a minimum gap distance from a second lesion, e.g., if the gap between the first and second lesions is less than a distance threshold, the first and second lesions can be identified as a composite lesion. In instances where the gap is greater than the distance threshold, the lesions can be identified as separate lesions. According to some examples, proposed or candidate treatment zones for a stent can be identified based on the identified lesions.

[0128] 9 illustrates a flow diagram of an example process for automatically determining plaque burden for frames in a region of interest that may be implemented using the computing device 112. The following operations do not have to be performed in the exact order described below. Rather, various operations may be processed in a different order or simultaneously, and operations may be added or omitted.

[0129] At block 910, vascular data including intravascular image data may be received. The intravascular image data may be, for example, one or more image frames acquired by the probe during pullback through a blood vessel. The one or more image frames may be OCT image frames, IVUS image frames, etc. The vascular data, in some examples, may be values ​​or measurements associated with a blood vessel in a given image frame. For example, the vascular data may be an EEL value, e.g., EEL diameter, a lumen value, e.g., lumen area or diameter, etc. In some examples, the vascular data may include a pressure value, a VFR value, an IMR value, etc.

[0130] At block 920, a region of interest in the vessel may be identified based on the vessel data. For example, the vessel data may include a representation of one or more side branches. The region of interest may be defined as the region between a first side branch and a second side branch.

[0131] In block 930, predicted EEL values ​​for frames within the region of interest may be determined as a function of the intimal and medial thicknesses. According to some examples, the predicted EEL value may be a representative EEL value for frames within the region of interest. In another example, the predicted EEL value may be determined frame-by-frame for frames within the region of interest. For example, a predicted EEL value may be determined for each frame within the region of interest. In some examples, respective predicted EEL values ​​may be determined for at least some of the frames within the region of interest. According to some examples, the intimal thickness may be a minimum intimal thickness, and the medial thickness may be a minimum medial thickness. The minimum intimal and medial thicknesses may be determined based on histological observation of the blood vessel. For example, the intimal and medial thicknesses may be determined by standard deviation based on histological observation of the blood vessel. The minimum intimal and medial thicknesses may be determined based on a lower limit of the standard deviation.

[0132] According to some examples, the predicted EEL can be further determined as a function of the HK model. The HK model predicts lumen diameter based on the assumption that the diameter, circumference, and / or area of ​​a blood vessel are substantially constant along a region of interest. For example, the predicted EEL can be determined using the following equation: Predicted EEL value = (expected lumen diameter) + 2 * (minimum intimal thickness + minimum medial thickness)

[0133] where the predicted EEL value can be the predicted EEL area and / or the predicted EEL diameter, and the predicted lumen diameter can be, for example, the lumen diameter determined using the HK model.

[0134] According to some examples, the predicted EEL value determined as a function of the HK model can be compared to the EEL value of a proximal segment of the blood vessel and / or a distal segment of the blood vessel. The EEL value of the proximal segment and / or distal segment of the blood vessel can, in some examples, be a representative EEL value for the proximal segment and / or distal segment of the blood vessel. The representative EEL value for the region of interest can be, for example, any or all of the median EEL value, mean EEL value, average EEL value, or mode EEL value. If the predicted EEL value is less than the representative EEL value of the distal segment, the representative EEL value of the distal segment can be used to determine the plaque burden of the frame within the region of interest. If the predicted EEL value is less than the representative EEL value of the distal segment due to the HK model assuming that the blood vessel has a natural taper from the proximal to the distal vessel segment, the representative EEL value of the distal segment can be used. If the predicted EEL value is greater than the representative EEL value of the proximal segment, the representative EEL value of the proximal segment can be used to determine the plaque burden of the frame within the region of interest. Because the HK model assumes that blood vessels have a natural taper from the proximal to the distal vessel segment, if the predicted EEL value is greater than the representative EEL value of the proximal segment, the representative EEL value of the proximal segment can be used.

[0135] At block 940, the plaque burden for the frames within the region of interest may be determined based on the predicted EEL values. The plaque burden may further be based on the lumen area for each frame. For example, the plaque burden may be determined for each frame using the formula 1 - (lumen area / predicted EEL value).

[0136] Similar to the example process described with respect to FIG. 8 , the plaque burden for each frame can be used to identify lesions within a blood vessel. For example, a lesion can be identified by identifying an image frame with a maximum plaque burden. According to some examples, a lesion can be identified by identifying an image frame in which the plaque burden is greater than a disease threshold. The disease threshold can be, for example, a plaque burden disease threshold. The plaque burden disease threshold can be, for example, approximately 70%. In some examples, the plaque burden disease threshold can be greater or less than 70%. For example, the plaque burden disease threshold can be patient-specific, physician-determined, etc.

[0137] Using that image frame as the initial frame, the plaque burden of frames proximal and distal to the initial frame can be compared to the treatment zone threshold until the plaque burden of the image frames proximal and distal to the initial frame is less than the treatment zone threshold and / or until a visible EEL arc is greater than the arc threshold. The image frame between the identified proximal and distal frames is then marked as corresponding to the lesion. The process is repeated for the image frame with the next highest plaque burden, without considering the image frame identified as corresponding to the lesion. After each lesion is identified, the gap or distance between the lesions is determined. In instances where the gap is substantially equal to or less than the distance threshold, the lesions can be combined and identified as a single lesion. For example, if a first lesion is within a minimum gap distance from a second lesion, e.g., if the gap between the first and second lesions is less than the distance threshold, the first and second lesions can be identified as a composite lesion. In instances where the gap is greater than the distance threshold, the lesions can be identified as separate lesions. According to some examples, the proposed or candidate treatment zone can be a stent landing zone. A stent landing zone for the stent can be identified based on the identified lesion. In some examples, treatment zones can be for balloon angioplasty, laser atherectomy, etc.

[0138] According to some examples, the EEL value for a given frame during pullback may not be determinable due to plaque overgrowth or other obstructions during pullback. In such examples, more vascular information can be determined by predicting EEL values ​​for these image frames. For example, plaque burden can be determined for each frame by averaging EEL values ​​or by predicting EEL values ​​as a function of intimal and medial thicknesses and the HK model. The determined plaque burden can be used to more accurately identify lesion location, length, severity, and the like. Furthermore, the predicted EEL value can be used to determine plaque burden for frames where EEL was not detected or was inaccurately measured, thereby more accurately identifying candidate treatment zones. For example, candidate treatment zones are typically adjusted by a physician during the stent planning stage. In contrast, automatically determining or predicting EEL values ​​and using the determined EEL values ​​to determine plaque burden enables automatic identification of candidate treatment zone end frames by identifying where the plaque burden is below a threshold, where a visible EEL arc is above an arc threshold, or both. This reduces the need for user intervention and allows for greater accuracy as the predicted EEL values ​​allow for a more informed decision-making process.

[0139] Overall, the systems and methods disclosed herein provide a variety of automated clinical tools to help physicians decide whether to treat a given patient, and if so, which lesions or stenoses to treat. The system can provide guidance for treating the most critical lesions based on physiology. Additionally, details about plaque type and other features detected by MLS can be used to select the shortest stent that will provide the greatest blood flow restoration. Additionally, virtual stenting can be implemented, providing interactive planning that allows the stent to be tailored for placement within the artery informed by the tissue classification and other measurements disclosed herein.

[0140] The aspects, features, and examples of the present disclosure are to be considered in all respects illustrative and are not intended to limit the disclosure, the scope of which is defined solely by the claims. Other examples, modifications, and uses will be apparent to those skilled in the art without departing from the spirit and scope of the claimed invention.

[0141] Throughout this application, when a composition is described as having, including, or comprising particular components, or a process is described as having, including, or comprising particular process steps, it is intended that the 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.

[0142] When an element or component is referred to herein as being included in and / or selected from a list of described elements or components, it should be understood that the element or component can be any one of the described elements or components, or can be selected from a group consisting of two or more of the described elements or components. Furthermore, it should be understood that the elements and / or features of the compositions, devices, or methods described herein, whether expressly or implied herein, can be combined in various ways without departing from the spirit and scope of the present teachings.

[0143] The use of the words "include," "includes," "including," or "have," "has," "having" should generally be understood to be open-ended and non-limiting, unless otherwise specified.

[0144] The use of the singular herein includes the plural (and vice versa) unless specifically stated otherwise. Furthermore, the singular forms "a," "an," and "the" include the plural unless the context clearly dictates otherwise. Furthermore, when the word "about" is used before a quantitative value, the present teachings also include that specific quantitative value itself unless specifically stated otherwise. As used herein, the word "about" refers to a ±10% variation from the nominal value. All numerical values ​​and numerical ranges disclosed herein are considered to include "about" before each value.

[0145] 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.

[0146] When a range or list of values ​​is provided, each intervening value between the upper and lower limits of the range or list of values ​​is considered individually and is encompassed by the invention as if each value were specifically recited herein. Additionally, smaller ranges between (and including) the upper and lower limits of a given range are also considered and encompassed by the invention. The listing of exemplary values ​​or ranges does not waive any other values ​​or ranges between (and including) the upper and lower limits of a given range.

Claims

1. receiving, by one or more processors, vascular data including intravascular image data; identifying, by the one or more processors, a vascular region of interest based on the vascular data; the one or more processors identifying a first subset of frames within the region of interest having a plaque burden below a threshold plaque burden, the first subset of frames including at least a threshold number of frames; the one or more processors determining external elastic lamina (EEL) values ​​for the first subset of frames; determining, by the one or more processors, a representative EEL value for the region of interest based on the EEL values ​​of the first subset of frames; determining, by the one or more processors, the plaque burden for a second subset of frames within the region of interest based on the representative EEL value, the second subset of frames being different from the first subset of frames; The method comprising:

2. The method of claim 1 , wherein the representative EEL value is at least one of a mean EEL value, an average EEL value, a median EEL value, or a mode EEL value.

3. The method of claim 1 or 2, wherein the region of interest is defined between a first branch vessel and a second branch vessel.

4. The EEL value is the EEL diameter or EEL area, The method according to any one of claims 1 to 3, wherein the representative EEL value is a representative EEL diameter or a representative EEL area.

5. The method of any one of claims 1 to 4, wherein the threshold plaque burden corresponds to a plaque burden of 0% to 50%.

6. The method of any one of claims 1 to 5, wherein the threshold number of frames is at least three frames.

7. The method of any one of claims 1 to 6, wherein determining the plaque burden is further based on the lumen area for each frame.

8. 7. The method of claim 6, wherein determining the plaque burden for each frame comprises calculating as 1-(the lumen area / the representative EEL value).

9. The method of any one of claims 1 to 8, further comprising the step of the one or more processors identifying lesions based on the determined plaque burden.

10. The method of claim 9 , further comprising the one or more processors identifying candidate treatment zones based on the identified lesions.

11. 11. The method of claim 10, wherein identifying the candidate treatment zone further comprises the one or more processors identifying frames proximal and distal to the identified lesion, wherein at least one of the plaque burdens of the frames proximal and distal to the identified lesion are below a threshold or a visible EEL arc is equal to or greater than an arc threshold.

12. 10. The method of claim 9, further comprising the one or more processors identifying the first lesion and the second lesion as a composite lesion if the first lesion is within a minimum gap distance from the second lesion.

13. 1. A system comprising one or more processors, The one or more processors: receiving vascular data including intravascular image data; identifying a target region of a blood vessel based on the blood vessel data; identifying a first subset of frames within the region of interest having a plaque burden below a threshold plaque burden, the first subset of frames including at least a threshold number of frames; determining external elastic lamina (EEL) values ​​for the first subset of frames; determining a representative EEL value for the region of interest based on the EEL values ​​of the first subset of frames; determining a plaque burden for a second subset of frames within the region of interest based on the representative EEL value, the second subset of frames being different from the first subset of frames; A system that is configured to:

14. The system of claim 13 , wherein the representative EEL value is at least one of a mean EEL value, an average EEL value, a median EEL value, or a mode EEL value.

15. 15. The system of claim 13 or 14, wherein the region of interest is defined between a first branch vessel and a second branch vessel.

16. The EEL value is the EEL diameter or EEL area, The system according to any one of claims 13 to 15, wherein the representative EEL value is a representative EEL diameter or a representative EEL area.

17. The system of any one of claims 13 to 16, wherein the threshold plaque burden corresponds to a plaque burden between 0% and 50%.

18. The system of any one of claims 13 to 17, wherein the threshold number of frames is at least three frames.

19. The system of any one of claims 13 to 18, wherein determining the plaque burden is further based on a lumen area for each frame.

20. 20. The system of claim 19, wherein determining the plaque burden for each frame comprises calculating as 1 - (the lumen area / the representative EEL value).

21. The system of any one of claims 13 to 20, wherein the one or more processors are further configured to identify lesions based on the determined plaque burden.

22. 22. The system of claim 21, wherein the one or more processors are further configured to identify candidate treatment zones based on the identified lesion.

23. 23. The system of claim 22, wherein when identifying the candidate treatment zone, the one or more processors are further configured to identify frames proximal and distal to the identified lesion, wherein at least one of the plaque burdens of the frames proximal and distal to the identified lesion are below a threshold or a visible EEL arc is equal to or greater than an arc threshold.

24. 22. The system of claim 21, wherein the one or more processors are further configured to identify the first lesion and the second lesion as a composite lesion if the first lesion is within a minimum gap distance from the second lesion.

25. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to: receiving vascular data including intravascular image data; Identifying a target region of a blood vessel based on the blood vessel data; identifying a first subset of frames within the region of interest having a plaque burden below a threshold plaque burden, wherein the first subset of frames includes at least a threshold number of frames; determining external elastic lamina (EEL) values ​​for the first subset of frames; determining a representative EEL value for the region of interest based on the EEL values ​​of the frames of the first subset; determining a plaque burden for a second subset of frames within the region of interest based on the representative EEL value, wherein the second subset of frames is different from the first subset of frames; A non-transitory computer-readable medium that causes

26. 26. The non-transitory computer-readable medium of claim 25, wherein the representative EEL value is at least one of a mean EEL value, an average EEL value, a median EEL value, or a mode EEL value.

27. 27. The non-transitory computer-readable medium of claim 25 or 26, wherein the region of interest is defined between a first branch vessel and a second branch vessel.

28. The EEL value is the EEL diameter or EEL area, 28. The non-transitory computer-readable medium of claim 25, wherein the representative EEL value is a representative EEL diameter or a representative EEL area.

29. 29. The non-transitory computer-readable medium of any one of claims 25 to 28, wherein the threshold plaque burden corresponds to a plaque burden between 0% and 50%.

30. 30. The non-transitory computer-readable medium of any one of claims 25 to 29, wherein the threshold number of frames is at least three frames.

31. 31. The non-transitory computer-readable medium of any one of claims 25 to 30, wherein determining the plaque burden is further based on a lumen area for each frame.

32. 32. The non-transitory computer-readable medium of claim 31, wherein determining the plaque burden for each frame comprises calculating as 1 - (the lumen area / the representative EEL value).

33. 33. The non-transitory computer-readable medium of any one of claims 25 to 32, wherein the one or more processors are further configured to identify lesions based on the determined plaque burden.

34. 34. The non-transitory computer-readable medium of claim 33, wherein the one or more processors are further configured to identify candidate treatment zones based on the identified lesion.

35. 35. The non-transitory computer-readable medium of claim 34, wherein when identifying the candidate treatment zone, the one or more processors are further configured to identify frames proximal and distal to the identified lesion, wherein at least one of the plaque burdens of the frames proximal and distal to the identified lesion are below a threshold or a visible EEL arc is equal to or greater than an arc threshold.

36. 34. The non-transitory computer-readable medium of claim 33, wherein if a first lesion is within a minimum gap distance from a second lesion, the one or more processors are further configured to identify the first lesion and the second lesion as a composite lesion.

37. receiving, by one or more processors, vascular data including intravascular image data; identifying, by the one or more processors, a vascular region of interest based on the vascular data; determining, based on the vascular data, predicted external elastic lamina (EEL) values ​​for frames within the region of interest as a function of intimal and medial thickness; determining, by the one or more processors, a plaque load for frames within the region of interest based on the predicted EEL values; The method comprising:

38. 38. The method of claim 37, wherein the region of interest is defined between a first branch vessel and a second branch vessel.

39. 39. The method of claim 37 or 38, wherein the predicted EEL value is further determined as a function of at least one of lumen area or lumen diameter derived from the HK model.

40. 40. The method of any one of claims 37 to 39, wherein determining the plaque burden is further based on the lumen area for each frame.

41. 41. The method of claim 40, wherein determining the plaque burden for each frame comprises calculating 1 - (the lumen area / the predicted EEL value).

42. The method of any one of claims 37 to 41, further comprising the step of the one or more processors identifying lesions based on the determined plaque burden.

43. 43. The method of claim 42, further comprising the one or more processors identifying candidate treatment zones based on the identified lesions.

44. 44. The method of claim 43, wherein identifying the candidate treatment zone further comprises the one or more processors identifying frames proximal and distal to the identified lesion, wherein at least one of the plaque burdens of the frames proximal and distal to the identified lesion are below a threshold or a visible EEL arc is equal to or greater than an arc threshold.

45. 45. The method of claim 44, wherein if a first lesion is within a minimum gap distance from a second lesion, the method further comprises the one or more processors identifying the first lesion and the second lesion as a composite lesion.

46. the one or more processors further comparing the predicted EEL value for the region of interest to a representative EEL value for a proximal segment of the blood vessel; 46. ​​The method of any one of claims 37 to 45, wherein if the representative EEL value for the proximal segment is less than the predicted EEL value for the target region, determining the plaque burden for the target region is based on the representative EEL value of the proximal segment.

47. the one or more processors further comparing the predicted EEL value for the region of interest to a representative EEL value for a distal segment of the blood vessel; 47. The method of claim 37, wherein if the representative EEL value for the distal segment is greater than the predicted EEL value for the target region, determining the plaque burden for the target region is based on the representative EEL value of the distal segment.

48. 48. The method of claim 46 or 47, wherein the representative EEL value is at least one of a mean EEL value, an average EEL value, a median EEL value, or a mode EEL value.

49. 1. A system comprising one or more processors, The one or more processors: receiving vascular data including intravascular image data; identifying a target region of a blood vessel based on the blood vessel data; determining predicted external elastic lamina (EEL) values ​​for frames within the region of interest based on the vascular data as a function of intimal and medial thickness; determining a plaque burden for a frame within the region of interest based on the predicted EEL value; and A system that is configured to:

50. 50. The system of claim 49, wherein the region of interest is defined between a first branch vessel and a second branch vessel.

51. 51. The system of claim 49 or 50, wherein the predicted EEL value is further determined as a function of at least one of lumen area or lumen diameter derived from the HK model.

52. 52. The system of any one of claims 49 to 51, wherein determining the plaque burden is further based on a lumen area for each frame.

53. 53. The system of claim 52, wherein determining the plaque burden for each frame comprises calculating 1 - (the lumen area / the predicted EEL value).

54. 54. The system of any one of claims 49 to 53, wherein the one or more processors are further configured to identify lesions based on the determined plaque burden.

55. 55. The system of claim 54, wherein the one or more processors are further configured to identify candidate treatment zones based on the identified lesion.

56. 56. The system of claim 55, wherein when identifying the candidate treatment zone, the one or more processors are further configured to identify frames proximal and distal to the identified lesion, wherein at least one of the plaque burdens of the frames proximal and distal to the identified lesion are below a threshold or a visible EEL arc is equal to or greater than an arc threshold.

57. 57. The system of claim 56, wherein the one or more processors are further configured to identify the first lesion and the second lesion as a composite lesion if the first lesion is within a minimum gap distance from the second lesion.

58. the one or more processors are further configured to compare the predicted EEL value for the region of interest to a representative EEL value for a proximal segment of the blood vessel; 58. The system of any one of claims 49 to 57, wherein if the representative EEL value for the proximal segment is less than the predicted EEL value for the target region, determining the plaque burden for the target region is based on the representative EEL value of the proximal segment.

59. the one or more processors are further configured to compare the predicted EEL value for the region of interest to a representative EEL value for a distal segment of the blood vessel; 59. The system of any one of claims 49 to 58, wherein if the representative EEL value for the distal segment is greater than the predicted EEL value for the target region, determining the plaque burden for the target region is based on the representative EEL value of the distal segment.

60. 60. The system of claim 58 or 59, wherein the representative EEL value is at least one of a mean EEL value, an average EEL value, a median EEL value, or a mode EEL value.

61. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to: receiving vascular data including intravascular image data; Identifying a target region of a blood vessel based on the blood vessel data; determining predicted external elastic lamina (EEL) values ​​for frames within the region of interest based on the vascular data as a function of intimal and medial thickness; determining a plaque burden for a frame within the region of interest based on the predicted EEL value; A non-transitory computer-readable medium that causes

62. 62. The non-transitory computer-readable medium of claim 61, wherein the region of interest is defined between a first branch vessel and a second branch vessel.

63. 63. The non-transitory computer-readable medium of claim 61 or 62, wherein the predicted EEL value is further determined as a function of at least one of a lumen area or a lumen diameter derived from an HK model.

64. 64. The non-transitory computer readable medium of any one of claims 61 to 63, wherein determining the plaque burden is further based on a lumen area for each frame.

65. 65. The non-transitory computer-readable medium of claim 64, wherein determining the plaque burden for each frame comprises calculating 1 - (the lumen area / the predicted EEL value).

66. 66. The non-transitory computer-readable medium of any one of claims 61 to 65, wherein the one or more processors are further configured to identify lesions based on the determined plaque burden.

67. 67. The non-transitory computer-readable medium of claim 66, wherein the one or more processors are further configured to identify candidate treatment zones based on the identified lesion.

68. 68. The non-transitory computer-readable medium of claim 67, wherein when identifying the candidate treatment zone, the one or more processors are further configured to identify frames proximal and distal to the identified lesion, wherein at least one of the plaque burdens of the frames proximal and distal to the identified lesion is below a threshold and / or a visible EEL arc is equal to or greater than an arc threshold.

69. 69. The non-transitory computer-readable medium of claim 68, wherein if a first lesion is within a minimum gap distance from a second lesion, the one or more processors are further configured to identify the first lesion and the second lesion as a composite lesion.

70. the one or more processors are further configured to compare the predicted EEL value for the region of interest to a representative EEL value for a proximal segment of the blood vessel; 70. The non-transitory computer-readable medium of any one of claims 61 to 69, wherein if the representative EEL value for the proximal segment is less than the predicted EEL value for the target region, determining the plaque burden for the target region is based on the representative EEL value of the proximal segment.

71. the one or more processors are further configured to compare the predicted EEL value for the region of interest to a representative EEL value for a distal segment of the blood vessel; 71. The non-transitory computer-readable medium of any one of claims 61 to 70, wherein if the representative EEL value for the distal segment is greater than the predicted EEL value for the target region, determining the plaque burden for the target region is based on the representative EEL value of the distal segment.

72. 72. The non-transitory computer-readable medium of claim 70 or 71, wherein the representative EEL value is at least one of a mean EEL value, an average EEL value, a median EEL value, or a mode EEL value.

Images