Information processing method, program, and information processing device

Abstract

In this information processing method, a computer executes processing for: acquiring data indicating a vascular tomographic image; inputting the acquired data indicating the vascular tomographic image into a model trained to detect, upon receiving input of data indicating a vascular tomographic image, for one or a plurality of objects of detection, a detection object region indicating the object of detection and a region different from the detection object region, and thereby detecting the detection object region and the region different from the detection object region; and determining the boundary of the detection object region on the basis of the detected detection object region and the region different from the detection object region.

Description

Information processing method, program, and information processing device

[0001] The present invention relates to an information processing method, a program, and an information processing device.

[0002] There is a technology that uses machine learning to support image diagnosis related to vascular treatments such as PCI (Percutaneous Coronary Intervention). For example, Patent Literature 1 discloses a system that uses a neural network to segment a region of interest from a polar image of a coronary artery.

[0003] Special Publication No. 2022-522960

[0004] In one aspect, an object is to provide an information processing method and the like that can suitably determine the boundary of a detection target region in a vascular tomographic image.

[0005] In one aspect, (1) an information processing method includes acquiring data representing a vascular tomographic image, and inputting the acquired data representing the vascular tomographic image into a model that has been trained to detect, for one or more detection targets, a detection target region representing the detection target and a region different from the detection target region, whereby a computer detects the detection target region and a region different from the detection target region, and determines the boundary of the detection target region based on the detected detection target region and the region different from the detection target region.

[0006] (2) In the information processing method of (1) above, the detection target area includes at least one of an area indicating the inside of a stent, an area indicating the lumen of the main trunk of a blood vessel, an area indicating the inside of the external elastic lamina of the main trunk of a blood vessel, an area indicating the inside of the adventitia of the main trunk of a blood vessel, an area indicating the lumen of a side branch of a blood vessel, an area indicating the inside of the external elastic lamina of a side branch of a blood vessel, an area indicating the lumen of the main trunk and side branch of a blood vessel, and an area indicating the inside of the external elastic lamina of the main trunk and side branch of a blood vessel.

[0007] (3) In the information processing method of (1) or (2) above, the detection target region includes at least one of a plaque region, a thrombus region, a hematoma region, a calcification region, a dissection region, a crack region, an extravascular tissue region, and a device region.

[0008] (4) In the information processing method described in any of (1) to (3) above, the model is a model that has been trained to detect a detection target area and an area excluding the detection target area when data representing a vascular tomographic image is input, and by inputting data representing the acquired vascular tomographic image into the model, the detection target area and the area excluding the detection target area are detected, and the boundary of the detection target area is determined by combining the detected detection target area with the area excluding the detection target area.

[0009] (5) In the information processing method described in any of (1) to (4) above, the detection target area includes an area indicating a side branch of a blood vessel, and by inputting data indicating the acquired vascular tomographic image into the model, an area indicating a side branch of a blood vessel, an area indicating a main trunk of a blood vessel, and an area indicating the main trunk and side branches of a blood vessel are detected, an area indicating a side branch of a blood vessel is identified by excluding the area indicating the main trunk of the blood vessel from the detected area indicating the main trunk and side branches of the blood vessel, and a boundary of the area indicating the side branch of a blood vessel is determined by combining the identified area indicating the side branch of a blood vessel with the area indicating the side branch of a blood vessel detected by the model.

[0010] (6) In the information processing method of (5) above, the vascular tomographic image is a transverse image of a cross section intersecting the axial direction of the blood vessel, and the transverse image is output with a marker indicating a representative position of the side branch superimposed based on the boundary of the determined area indicating the side branch.

[0011] (7) The information processing method described in any of (1) to (6) above detects high-brightness or low-brightness areas included in data representing the vascular tomographic image, and determines the boundary of the detection target area based on the detection target area and areas different from the detection target area detected by the model, and the high-brightness or low-brightness areas.

[0012] (8) In the information processing method described in (7) above, the detection target area includes a calcified area, and by inputting data representing the acquired vascular tomographic image into the model, the calcified area, an area representing the inside of the external elastic lamina of the blood vessel, and an area representing the lumen of the blood vessel are detected, an area representing the vascular wall is identified based on the detected area representing the inside of the external elastic lamina of the blood vessel and the area representing the lumen of the blood vessel, a high-brightness area having a brightness equal to or greater than a threshold value is detected within the identified area representing the vascular wall, and the boundary of the calcified area is determined by combining the high-brightness area with the calcified area detected by the model.

[0013] (9) In the information processing method of (8) above, the vascular tomographic image is a transverse image obtained by capturing a cross section intersecting the axial direction of the blood vessel, and the angle occupied by the calcified region in the circumferential direction of the blood vessel is identified based on the determined boundary of the calcified region, and the transverse image showing the identified angle is output.

[0014] (10) In the information processing method of (8) or (9) above, the characteristics of the calcification are determined based on the position of the calcified region relative to the region representing the vascular wall.

[0015] (11) In the information processing method described in any of (7) to (10) above, the detection target region includes an attenuating plaque region, and by inputting data representing the acquired vascular tomographic image into the model, the attenuating plaque region and a region representing the inside of the external elastic lamina of the blood vessel are detected, a low-brightness region having a brightness equal to or less than a threshold outside the detected region representing the inside of the external elastic lamina is detected, and the boundary of the attenuating plaque region is determined by combining the low-brightness region with the attenuating plaque region detected by the model.

[0016] (12) In the information processing method of (11) above, the vascular tomographic image is a transverse image obtained by capturing a cross section intersecting the axial direction of the blood vessel, and the angle occupied by the attenuating plaque region in the circumferential direction of the blood vessel is identified based on the determined boundary of the attenuating plaque region, and the transverse image showing the identified angle is output.

[0017] (13) In the information processing method described in any of (1) to (12) above, the vascular tomographic image is a transverse cross-sectional image captured of a cross section intersecting the axial direction of the blood vessel, the detection target region includes a plaque region and a region representing the lumen of the blood vessel, and the ratio of the plaque region to the cross section of the blood vessel at each position along the axial direction is calculated based on the area of ​​the plaque region and the area of ​​the region representing the lumen of the blood vessel in each transverse cross-sectional image captured continuously along the axial direction, and a longitudinal cross-sectional image is generated that displays the vascular wall at each position along the axial direction in different display modes depending on the ratio based on the multiple transverse cross-sectional images.

[0018] (14) In the information processing method of (13) above, the detection target area further includes an area indicating a side branch of a blood vessel, and the longitudinal section image is generated with a predetermined marker attached to each position along the axial direction depending on whether or not an area indicating a side branch of a blood vessel is detected from each of the transverse section images.

[0019] In one aspect, (15) a program acquires data representing a vascular tomographic image, and when the data representing the vascular tomographic image is input, the program inputs the acquired data representing the vascular tomographic image into a model that has been trained to detect, for one or more detection targets, a detection target area representing the detection target and an area different from the detection target area, thereby detecting the detection target area and an area different from the detection target area, and causes a computer to execute a process of determining the boundary of the detection target area based on the detected detection target area and the area different from the detection target area.

[0020] In one aspect, (16) an information processing device is an information processing device that includes a control unit, and when the control unit acquires data representing a vascular tomographic image and inputs the data representing the vascular tomographic image, the control unit inputs the acquired data representing the vascular tomographic image into a model that has been trained to detect, for one or more detection targets, a detection target area representing the detection target and an area different from the detection target area, thereby detecting the detection target area and an area different from the detection target area, and determining the boundary of the detection target area based on the detected detection target area and the area different from the detection target area.

[0021] In one aspect, the boundary of the detection target region in the vascular tomographic image can be suitably determined.

[0022] FIG. 1 is an explanatory diagram showing an example of the configuration of an image diagnostic system. FIG. 2 is a block diagram showing an example of the configuration of a server. FIG. 3 is a block diagram showing an example of the configuration of an image diagnostic apparatus. FIG. 4 is an explanatory diagram showing an example of processing for detecting a detection target region from a vascular tomographic image. FIG. 5 is an explanatory diagram showing an overview of embodiment 1. FIG. 6 is a flowchart showing the procedure for processing to generate a detection model. FIG. 7 is a flowchart showing the procedure for processing to detect a detection target region. FIG. 8 is an explanatory diagram showing an overview of embodiment 2. FIG. 9 is an explanatory diagram showing an example of a display of a detection result of a vascular side branch. FIG. 10 is a flowchart showing the procedure for processing to detect a detection target region according to embodiment 2. FIG. 11 is an explanatory diagram showing an overview of embodiment 3. FIG. 12 is an explanatory diagram showing an example of a display of a detection result of a calcified region. FIG. 13 is a flowchart showing the procedure for processing to detect a detection target region according to embodiment 3. FIG. 14 is an explanatory diagram showing an overview of a modified example. FIG. 15 is an explanatory diagram showing an overview of embodiment 16. FIG. 16 is a flowchart showing the procedure for processing to detect a detection target region according to embodiment 16. FIG. 17 is an explanatory diagram showing an overview of embodiment 17.

[0023] The present invention will be described in detail below with reference to the drawings showing embodiments thereof. (Embodiment 1) Fig. 1 is an explanatory diagram showing an example of the configuration of an imaging diagnostic system. In this embodiment, an imaging diagnostic system will be described that detects a detection target region relating to a predetermined detection target (e.g., the lumen of a blood vessel, the external elastic lamina, plaque, etc.) from a vascular tomographic image obtained by capturing a cross section of a patient's blood vessel. The imaging diagnostic system includes a server 1 and an imaging diagnostic device 2. Each device is communicatively connected via a network N.

[0024] The server 1 is a server computer capable of various information processing and transmitting and receiving information. Note that the device corresponding to the server 1 may be a device such as a personal computer. The server 1 performs machine learning to learn predetermined training data, and generates a detection model 50 (see FIG. 5 ) that detects a detection target region and a region different from the detection target region using a vascular tomographic image as input.

[0025] The diagnostic imaging device 2 is an imaging device that captures medical images of a patient's blood vessels, and has an IVUS (Intravascular Ultrasound) function that generates vascular tomographic images by transmitting and receiving ultrasound, and an OCT (Optical Coherence Tomography) function that generates vascular tomographic images by transmitting and receiving light, using, for example, a catheter 201. The catheter 201 is a medical instrument that is inserted into a patient's blood vessels, and the tip of the catheter 201 is provided with an ultrasound transmitting and receiving unit that transmits and receives ultrasound, and an optical transmitting and receiving unit that transmits and receives light. The diagnostic imaging device 2 generates and displays IVUS images and OCT images based on the ultrasound and light received by the catheter 201.

[0026] In this embodiment, the diagnostic imaging device 2 is described as a device having both IVUS and OCT functions, but the diagnostic imaging device 2 may be a device having only one of the functions. Also, the vascular tomographic image may be captured by a method other than IVUS or OCT.

[0027] Data of the detection model 50 generated by the server 1 is installed in the image diagnostic device 2, and the image diagnostic device 2 inputs a vascular tomographic image captured using the catheter 201 into the detection model 50 and detects a detection target region and a region different from the detection target region. As will be described later, the image diagnostic device 2 identifies the detection target region based on the region different from the detection target region, and compares the identified detection target region with the detection target region detected by the detection model 50 to determine the boundary of the detection target region.

[0028] In this embodiment, the diagnostic imaging device 2 detects the detection target region using the detection model 50, but the server 1 on the cloud may execute the processing using the detection model 50. Alternatively, for example, a general-purpose computer connected to the diagnostic imaging device 2 may execute the processing. In this way, the processing entity that executes the series of processes is not particularly limited.

[0029] FIG. 2 is a block diagram showing an example configuration of the server 1. The server 1 includes a control unit 11, a main memory unit 12, a communication unit 13, and an auxiliary memory unit 14. The control unit 11 has one or more arithmetic processing devices, such as a central processing unit (CPU), a microprocessing unit (MPU), or a graphics processing unit (GPU), and performs various information processing, control processing, and the like, by reading and executing a program P1 stored in the auxiliary memory unit 14. The main memory unit 12 is a temporary storage area, such as a static random access memory (SRAM) or a dynamic random access memory (DRAM), and temporarily stores data required for the control unit 11 to execute arithmetic processing. The communication unit 13 is a communication module for performing communication-related processing and transmits and receives information to and from the outside. The auxiliary memory unit 14 is a non-volatile storage area, such as a large-capacity memory or a hard disk, and stores the program P1 (program product) and other data required for the control unit 11 to execute processing.

[0030] The auxiliary storage unit 14 may be an external storage device connected to the server 1. The server 1 may be a multi-computer consisting of multiple computers, or may be a virtual machine virtually constructed by software.

[0031] Furthermore, in this embodiment, the server 1 is not limited to the above configuration, and may include, for example, an input unit that accepts operation input, a display unit that displays images, etc. Furthermore, the server 1 may be provided with a reading unit that reads a portable storage medium 1a such as a CD (Compact Disk)-ROM or a DVD (Digital Versatile Disc)-ROM, and may read and execute the program P1 from the portable storage medium 1a.

[0032] FIG. 3 is a block diagram showing an example configuration of the imaging diagnostic device 2. The imaging diagnostic device 2 includes a control unit 21, a main memory unit 22, a communication unit 23, a display unit 24, an input unit 25, an image processing unit 26, and an auxiliary memory unit 27. The control unit 21 has one or more arithmetic processing units such as CPUs, and performs various information processing, control processing, etc. by reading and executing a program P2 stored in the auxiliary memory unit 27. The main memory unit 22 is a temporary storage area such as RAM, and temporarily stores data necessary for the control unit 21 to execute arithmetic processing. The communication unit 23 is a communication module for performing communication-related processing and transmits and receives information to and from the outside. The display unit 24 is a display screen such as a liquid crystal display, and displays images. The input unit 25 is an operation interface such as a keyboard or mouse, and accepts operation inputs from the user. The image processing unit 26 is an image processing module that processes signals transmitted and received via the catheter 201 and generates a vascular tomographic image.

[0033] The auxiliary storage unit 27 is a non-volatile storage area such as a hard disk or large-capacity memory, and stores a program P2 (program product) and other data necessary for the control unit 21 to execute processing. The auxiliary storage unit 27 also stores a detection model 50. The detection model 50 is a machine learning model generated by learning predetermined training data, and is a trained model that detects a detection target region, etc., using a vascular tomographic image captured by the image diagnostic device 2 as input.

[0034] The diagnostic imaging device 2 may be provided with a reading unit for reading a portable storage medium 2a such as a CD-ROM, and may read and execute the program P2 from the portable storage medium 2a.

[0035] Fig. 4 is an explanatory diagram showing an example of processing for detecting a detection target region from a vascular tomographic image. Fig. 4 illustrates how the lumen, stent, external elastic membrane (EEM), etc. of a blood vessel are detected from a transverse cross-sectional image captured from a cross section intersecting (orthogonal to) the axial direction of the blood vessel. Based on Fig. 4, we will explain the problems involved in detecting various objects from a vascular tomographic image using a machine learning model.

[0036] In this embodiment, the term "vascular tomographic image" is basically explained as meaning a transverse cross-sectional image (an image of a cross section intersecting the axial direction of the blood vessel), but this embodiment is not limited to this, and the detection target may also be detected from a longitudinal cross-sectional image of the blood vessel (an image of a cross section along the axial direction of the blood vessel).

[0037] When detecting a desired detection target from a vascular tomographic image, it is possible to perform processing related to semantic segmentation using, for example, a machine learning model. Semantic segmentation is a method for identifying targets from an image on a pixel-by-pixel basis, and is performed using, for example, a convolutional neural network (CNN) such as U-net.

[0038] By training a machine learning model for semantic segmentation to learn the features of various objects, it is possible to detect multiple objects from a single image. For example, as shown on the left side of Figure 4, it is possible to detect the lumen of a blood vessel, the external elastic lamina, and the boundaries of a stent placed in the blood vessel from a vascular tomographic image, and it is also possible to identify the area of ​​each object on a pixel-by-pixel basis, as shown by the hatching in the center of Figure 4.

[0039] However, the more detection targets a single model is trained on, the more complex the problem the model must solve, resulting in a decrease in accuracy.In addition, training data must be prepared for each segmented region, which increases costs.

[0040] To address the above issues, it is conceivable to create a dedicated model for each detection target and divide the detection task into multiple parts. However, this increases the amount of calculation required, which leads to problems such as a decrease in computer processing speed. Furthermore, the increase in models increases the computer's implementation capacity, which increases hardware costs.

[0041] Therefore, in this embodiment, a mode will be described in which detection is performed using a single model while improving detection accuracy by dividing the detection task into multiple parts.

[0042] Fig. 5 is an explanatory diagram showing an overview of the first embodiment. Fig. 5 illustrates how a detection target region ("inLumen") indicating the main lumen of a blood vessel and a region ("inLumen_background") different from the detection target region are detected using the same detection model 50, and the results of both detections are combined to finally determine the boundary of the detection target region. An overview of this embodiment will be described based on Fig. 5.

[0043] The detection model 50 is a machine learning model that has learned predetermined training data, and when a vascular tomographic image is input, the model detects, for one or more detection targets, a detection target region indicating the detection target and a region different from the detection target region. For example, the detection model 50 is a model (CNN) related to semantic segmentation such as U-net, and detects the detection target region and a region different from the detection target region on a pixel-by-pixel basis.

[0044] The detection model 50 may be a neural network other than a CNN. The detection model 50 may also be a machine learning model other than a neural network, such as a decision tree or a support vector machine (SVM).

[0045] Furthermore, in this embodiment, a vascular tomographic image generated (constructed) from data on reflected ultrasound waves is input to the detection model 50, but this embodiment is not limited to this, and raw data of reflected waves that form the basis of the vascular tomographic image may also be used as input to the detection model 50. In other words, the detection model 50 only needs to be able to receive data representing a vascular tomographic image as input, and this data is not limited to image data.

[0046] Furthermore, the image diagnostic device 2 according to this embodiment captures IVUS images (ultrasound tomographic images) and OCT images (optical coherence tomographic images) as described above, but a separate detection model 50 may be constructed for each image, or a common detection model 50 may be used to detect a detection target region from each image.

[0047] The detection target region includes at least one of a region showing the inside of a stent, a region showing the lumen of the main trunk of a blood vessel, a region showing the inside of the external elastic lamina of the main trunk of a blood vessel, a region showing the inside of the adventitia of the main trunk of a blood vessel, a region showing the lumen of a side branch of a blood vessel, a region showing the inside of the external elastic lamina of a side branch of a blood vessel, a region showing the lumen of the main trunk and side branch of a blood vessel, a region showing the inside of the external elastic lamina of the main trunk and side branch of a blood vessel, a plaque region, a thrombus region, a hematoma region, a calcification region, a dissection region, a crack region, an extravascular tissue region, and a device region. Note that "device" refers to a device such as a guidewire inserted into a patient's blood vessel.

[0048] Note that the above is an example of the detection target region, and other detection targets may be detected. The detection model 50 according to this embodiment detects the detection target region representing each of these multiple detection targets. As an example, FIG. 5 illustrates the case of detecting the main trunk lumen of a blood vessel.

[0049] The region different from the detection target region refers to a region of an object different from the detection target whose boundary is to be finally determined. In this embodiment, in order to finally determine the boundary of the main trunk lumen of the blood vessel, the region excluding the region indicating the main trunk lumen of the blood vessel (detection target region) is detected as a region different from the detection target region.

[0050] The server 1 generates the detection model 50 by learning training data for generating the detection model 50. The training data is data in which a label indicating the position of the correct detection target region (region indicating the main trunk lumen of the blood vessel) is added to a training vascular tomographic image.

[0051] The server 1 inverts the detection target region indicated by the label attached to the training vascular tomographic image to identify the region excluding the detection target region (the region indicating the outside of the main trunk lumen of the blood vessel). The server 1 attaches a label indicating the position of the identified region to the training vascular tomographic image.

[0052] The server 1 generates a trained detection model 50 that detects a detection target region and a region excluding the detection target region when a vascular tomographic image is input based on newly labeled training data. The server 1 detects the detection target region and a region excluding the detection target region by inputting the training vascular tomographic image to the detection model 50. Specifically, the server 1 inputs the training vascular tomographic image to the detection model 50, and outputs label data (label images) that indicate, with a binary value, whether each pixel belongs to the target region or not for each of the detection target region and the region excluding the detection target region. The server 1 compares each detected region with the region indicated by the correct label and adjusts parameters such as weights between neurons so that the two regions match. The server 1 sequentially inputs each training vascular tomographic image to the detection model 50 to perform learning, and ultimately generates a detection model 50 with optimized parameters.

[0053] The diagnostic imaging device 2 detects the detection target region using the detection model 50 generated as described above. First, as shown in Fig. 5 , the diagnostic imaging device 2 inputs a vascular tomographic image into the detection model 50, thereby detecting the detection target region (region indicating the main lumen of the blood vessel) and a region excluding the detection target region (region indicating the outside of the main lumen of the blood vessel). Specifically, the diagnostic imaging device 2 inputs the vascular tomographic image into the detection model 50, and outputs label data (label image) indicating, with a binary value, whether or not each pixel belongs to the target region, for each of the detection target region and the region excluding the detection target region.

[0054] Next, the image diagnostic device 2 identifies the detection target region by inverting the region excluding the detection target region. The image diagnostic device 2 compares the detection target region identified by inverting the region excluding the detection target region with the detection target region detected by the detection model 50, and determines the boundary of the detection target region. For example, the image diagnostic device 2 determines a region (pixels) belonging to either of the two as the detection target region (OR operation) and determines the boundary. Alternatively, the image diagnostic device 2 may determine a region (pixels) belonging to both as the detection target region (AND operation) and determine the boundary.

[0055] The diagnostic imaging device 2 displays the detection result of the detection target region. For example, the diagnostic imaging device 2 displays a vascular tomographic image in which a boundary line is superimposed on the detection target region (the lumen of the main trunk of the blood vessel) as in Fig. 4. The diagnostic imaging device 2 also calculates the lumen diameter of the blood vessel based on the boundary line of the lumen and displays it together with the vascular tomographic image.

[0056] While the lumen of the trunk of a blood vessel has been cited above as an example of a detection target, the diagnostic imaging device 2 similarly detects detection target regions for other detection targets using the detection model 50. That is, the diagnostic imaging device 2 detects regions such as the inside of a stent and the inside of the external elastic lamina of the trunk of a blood vessel as detection target regions using the detection model 50, detects regions excluding these detection target regions (inverted regions), and determines the boundaries of the detection target regions by combining these regions. The diagnostic imaging device 2 also superimposes and displays the boundaries of these detection targets on the vascular tomographic image, and also calculates and displays the diameter of the external elastic lamina (EEM diameter), etc.

[0057] In this way, the diagnostic imaging device 2 detects the detection target region and a region different from the detection target region (a region excluding the detection target region) using the detection model 50, and determines the boundary of the detection target region based on the detection target region and the region different from the detection target region. This makes it possible to improve detection accuracy while suppressing an increase in calculation costs.

[0058] FIG. 6 is a flowchart showing the steps of the process for generating the detection model 50. The process for generating the detection model 50 by machine learning will be described with reference to FIG. 6 . The control unit 11 of the server 1 acquires training data for generating the detection model 50 (step S11). The training data is data in which a label indicating the position of the correct detection target region is added to a training vascular tomographic image. The vascular tomographic image is, for example, a transverse image obtained by capturing a cross section intersecting (orthogonal to) the axial direction of the blood vessel. The detection target region is, for example, a region indicating the main trunk lumen of the blood vessel, but is not limited thereto.

[0059] The control unit 11 inverts the detection target region indicated by the training data, and assigns a label indicating the position of the region excluding the detection target region to the training vascular tomographic image (step S12).

[0060] The control unit 11 generates a detection model 50 that has been trained based on the training data to detect a detection target region and a region excluding the detection target region when a vascular tomographic image is input (step S13). For example, the control unit 11 generates a neural network such as a CNN as the detection model 50. The control unit 11 inputs a training vascular tomographic image into the detection model 50 to detect the detection target region and a region excluding the detection target region. The control unit 11 compares each detected region with a correct region and optimizes parameters such as the weights between neurons so that the two match, thereby generating the detection model 50. The control unit 11 ends the series of processes.

[0061] 7 is a flowchart showing the procedure for the detection process of the detection target region. The process for detecting the detection target region using the detection model 50 and determining its boundary will be described with reference to FIG. 7 . The control unit 21 of the image diagnostic device 2 acquires a vascular tomographic image of a patient's vascular cross section (step S31). The control unit 21 inputs the acquired vascular tomographic image into the detection model 50, thereby detecting the detection target region (e.g., a region showing the main trunk lumen of the blood vessel) and a region excluding the detection target region (a region obtained by inverting the detection target region) (step S32).

[0062] The control unit 21 determines the boundary of the detection target area by combining the detection target area detected in step S32 with the area excluding the detection target area (step S33). Specifically, the control unit 21 identifies the detection target area by inverting the area excluding the detection target area. The control unit 21 determines the boundary of the detection target area by performing an OR operation or an AND operation between the identified detection target area and the detection target area detected by the detection model 50.

[0063] The control unit 21 displays a vascular tomographic image in which a boundary line is superimposed on the detection target region (step S34), and ends the series of processes.

[0064] As described above, according to the first embodiment, the boundary of the detection target region in the vascular tomographic image can be suitably determined.

[0065] Second Embodiment In the present embodiment, a configuration for detecting a side branch of a blood vessel from a tomographic image will be described. Note that the same reference numerals will be used to designate the same components as those in the first embodiment, and the description thereof will be omitted.

[0066] 8 is an explanatory diagram showing an outline of the second embodiment. The outline of the present embodiment will be explained based on the second embodiment.

[0067] In this embodiment, the detection model 50 detects side branches of blood vessels in addition to the lumen of the main trunk of the blood vessel and the external elastic lamina of the main trunk, etc. Specifically, as shown in Fig. 8 , the detection model 50 detects a region indicating side branches of blood vessels as a detection target region, and detects a region indicating the main trunk of the blood vessel and a region indicating the main trunk and side branches of the blood vessel as regions different from the detection target region.

[0068] The "region representing a side branch of a blood vessel" may be a region representing the lumen of the side branch of a blood vessel, or may be a region representing the inside of the external elastic lamina of the side branch. Similarly, the "region representing the main trunk of a blood vessel" and the "region representing the main trunk and side branch of a blood vessel" may be a region representing the lumen, or may be a region representing the inside of the external elastic lamina. In this embodiment, both the lumen and the external elastic lamina are detected. Figure 8 illustrates an example of detecting the lumen.

[0069] The server 1 generates a detection model 50 using training data for a training vascular tomographic image, to which labels indicating the positions of the correct area indicating the side branch of the blood vessel, the area indicating the main trunk of the blood vessel, and the areas indicating the main trunk and side branches of the blood vessel are assigned.

[0070] The diagnostic imaging device 2 detects regions indicating side branches of blood vessels using the detection model 50 generated as described above. First, the diagnostic imaging device 2 detects regions indicating side branches of blood vessels, regions indicating the main trunk of the blood vessel, and regions indicating the main trunk and side branches of the blood vessel by inputting a vascular tomographic image into the detection model 50. Next, the diagnostic imaging device 2 identifies regions indicating side branches of blood vessels by removing the region indicating the main trunk of the blood vessel from the detected regions indicating the main trunk and side branches of the blood vessel.

[0071] The diagnostic imaging device 2 determines the boundary of the region indicating the side branch of the blood vessel by combining the identified region indicating the side branch of the blood vessel with the region indicating the side branch of the blood vessel detected by the detection model 50. That is, the diagnostic imaging device 2 determines the boundary of the region indicating the side branch of the blood vessel by performing an OR operation or an AND operation on both.

[0072] 9 is an explanatory diagram showing an example of the display of the detection results of vascular side branches. The diagnostic imaging device 2 displays the detection results of the detection target region (in this embodiment, the region indicating the vascular side branch) in the same manner as in the first embodiment. Here, the diagnostic imaging device 2 superimposes a marker 91 indicating a representative position of the vascular side branch on the vascular tomographic image (transverse image) based on the boundary of the region indicating the vascular side branch determined above. For example, the diagnostic imaging device 2 identifies the center of gravity of the region indicating the vascular side branch as the representative position of the vascular side branch, and displays the marker 91 at the identified representative position.

[0073] 9, two markers 91 are displayed, which represent the representative position of the side branch based on the lumen of the blood vessel and the representative position of the side branch based on the external elastic lamina of the blood vessel. The diagnostic imaging device 2 detects an area indicating the lumen of the main trunk of the blood vessel and an area indicating the inside of the external elastic lamina of the main trunk of the blood vessel using the detection model 50, and superimposes the boundary line indicating the lumen of the main trunk of the blood vessel and the boundary line indicating the external elastic lamina of the main trunk of the blood vessel on the vascular tomographic image, and also superimposes a marker 91 indicating the representative position of the side branch based on the lumen of the blood vessel and a marker 91 indicating the representative position of the side branch based on the external elastic lamina of the blood vessel.

[0074] 10 is a flowchart showing the procedure of the detection process of the detection target region according to embodiment 2. The control unit 21 of the image diagnostic apparatus 2 acquires a vascular tomographic image (step S221). The control unit 21 inputs the acquired vascular tomographic image to the detection model 50, thereby detecting a region indicating a side branch of the blood vessel (detection target region), a region indicating the main trunk of the blood vessel, and a region indicating the main trunk and side branches of the blood vessel (regions different from the detection target region) (step S222).

[0075] The control unit 21 identifies an area representing a side branch of the blood vessel by excluding the area representing the main trunk of the blood vessel from the area representing the detected main trunk and lumen of the blood vessel (step S223). The control unit 21 determines the boundary of the area representing the side branch of the blood vessel by combining the identified area representing the side branch of the blood vessel with the area representing the side branch of the blood vessel detected in step S222 (step S224). The control unit 21 displays a vascular tomographic image (transverse image) on which a marker indicating a representative position of the side branch is superimposed based on the determined boundary of the area representing the side branch of the blood vessel (step S225), and ends the series of processes.

[0076] As described above, according to the second embodiment, the boundary of the region indicating the side branch of the blood vessel can be suitably determined.

[0077] Third Embodiment In this embodiment, a mode for detecting calcified regions from a vascular tomographic image will be described.

[0078] 11 is an explanatory diagram showing an outline of the third embodiment. The outline of this embodiment will be described with reference to FIG.

[0079] The detection model 50 according to this embodiment detects calcified regions of vascular tissue, in addition to the lumen of the main trunk of the blood vessel and the external elastic lamina of the main trunk. Specifically, as shown in Fig. 11 , the detection model 50 detects calcified regions as detection target regions, and also detects a region indicating the inside of the external elastic lamina of the blood vessel (main trunk) and a region indicating the lumen of the blood vessel (main trunk) as regions different from the detection target region.

[0080] The server 1 generates the detection model 50 by using training data to which labels indicating the correct position of the calcified region, the position of the region indicating the inside of the external elastic lamina of the blood vessel, and the position of the region indicating the lumen of the blood vessel are assigned to the training vascular tomographic image. Note that in this embodiment, a polar coordinate image (an image in which one axis represents the distance from the tip of the catheter 201 and the other axis represents the rotation angle) is used instead of a Cartesian coordinate image as the vascular tomographic image to be input to the detection model 50.

[0081] The diagnostic imaging device 2 detects calcified regions using the detection model 50 generated as described above. First, the diagnostic imaging device 2 inputs a vascular tomographic image into the detection model 50, thereby detecting calcified regions ("Calc"), regions representing the inside of the external elastic lamina of the blood vessel ("inVessel"), and regions representing the lumen of the blood vessel ("inLumen").

[0082] Next, the diagnostic imaging device 2 identifies a region representing the vascular wall ("Plaque+media") based on the detected region representing the inside of the external elastic lamina of the blood vessel and the region representing the lumen of the blood vessel. The diagnostic imaging device 2 detects, as calcified regions, high-intensity regions with brightness equal to or greater than a threshold within the identified region representing the vascular wall. Specifically, the diagnostic imaging device 2 detects, as calcified regions, portions with brightness equal to or greater than the threshold (calcified surface) and a region (acoustic shadow) behind the portions with brightness equal to or greater than the threshold as viewed from the tip of the catheter 201 (lower in FIG. 11 ). The threshold used as a reference for detecting calcified regions may be determined according to the brightness value of the adventitia of the blood vessel.

[0083] The diagnostic imaging device 2 determines the boundaries of the calcified regions by combining the calcified regions (high-brightness regions) detected above with the calcified regions detected by the detection model 50. Specifically, the diagnostic imaging device 2 calculates the degree of agreement (e.g., Jaccard coefficient) between the two, and determines the regions where the degree of agreement is equal to or greater than a specified value as calcified regions, thereby determining their boundaries.

[0084] The above-described method for detecting calcified regions is a process that is mainly applied to IVUS images. Normally, calcified regions can be observed in OCT images as regions with thickness in the depth direction from the center of the image. Therefore, brightness correction (detection of high-brightness regions and determination of boundaries) is not performed on OCT images, and only the detection of calcified regions is performed using the detection model 50.

[0085] 12 is an explanatory diagram showing an example of the display of the detection results of calcified regions. The diagnostic imaging device 2 displays the detection results of the detection target region (calcified region) as in Embodiments 1 and 2. In this embodiment, the diagnostic imaging device 2 identifies the angle that the calcified region to be detected occupies in the circumferential direction of the blood vessel, and displays a vascular tomographic image (transverse image) showing this angle.

[0086] 12, the diagnostic imaging device 2 displays a circular arc 131 superimposed on the transverse image in accordance with the angular range of the calcified region as viewed from the center of gravity of the blood vessel lumen. The diagnostic imaging device 2 also displays the angle and thickness of the calcified region at the bottom right of the screen. Since the thickness of the calcified region cannot be identified from the IVUS image, it is identified from the results of calcified region detection using the OCT image.

[0087] 13 is a flowchart showing the steps of a detection process for a detection target region according to embodiment 3. The control unit 21 of the image diagnostic apparatus 2 acquires a vascular tomographic image (step S321). The control unit 21 inputs the acquired vascular tomographic image into the detection model 50, thereby detecting a calcified region (detection target region), a region representing the inside of the external elastic lamina of the blood vessel, and a region representing the lumen of the blood vessel (a region different from the detection target region) (step S322).

[0088] The control unit 21 identifies a region representing the vascular wall based on the detected region representing the inside of the external elastic lamina of the blood vessel and the region representing the blood vessel lumen (step S323).The control unit 21 detects a high-brightness region whose brightness is equal to or greater than a threshold within the identified region representing the vascular wall (step S324).The control unit 21 determines the boundary of the calcified region by combining the high-brightness region with the calcified region detected by the detection model 50 (step S325).

[0089] The control unit 21 identifies the angle occupied by the calcified region in the circumferential direction of the blood vessel based on the determined boundary of the calcified region (step S326), and displays a vascular tomographic image (transverse image) showing the identified angle (step S327), thereby completing the series of processes.

[0090] As described above, according to the third embodiment, the boundaries of calcified regions can be suitably determined.

[0091] (Modification) In the third embodiment, a configuration in which a calcified region is detected from a vascular tomographic image is described. In this modification, a configuration in which the characteristics of calcification are determined based on the position of the detected calcified region within the vascular wall will be described.

[0092] Fig. 14 is an explanatory diagram showing an outline of a modified example. In Fig. 14, solid lines indicate the lumen of the blood vessel and the external elastic lamina, dashed lines indicate the center line between the lumen of the blood vessel and the external elastic lamina, and hatched areas indicate calcified areas. An outline of this modified example will be described based on Fig. 14.

[0093] As explained in the third embodiment, the diagnostic imaging device 2 identifies a region indicative of a vascular wall using the detection model 50, identifies a region within the region indicative of the vascular wall where the brightness is equal to or greater than a threshold as a calcified region, and determines the boundary of the calcified region by comparing it with the calcified region detected by the detection model 50. In this modification, the diagnostic imaging device 2 determines the characteristics of the calcification, such as whether the calcified tissue is superficial, deep, or superficial, based on the position of the calcified region whose boundary has been determined relative to the region indicative of the vascular wall.

[0094] Specifically, the diagnostic imaging device 2 identifies the center line between the lumen of the blood vessel and the external elastic lamina from the region representing the blood vessel wall. The diagnostic imaging device 2 then determines whether the calcified region is located inside the center line and whether the calcified region is in contact with the lumen of the blood vessel. If the diagnostic imaging device 2 determines that the calcified region is located outside the center line, it determines that the calcification is deep. If the diagnostic imaging device 2 determines that the calcified region is located inside the center line and is not in contact with the lumen, it determines that the calcification is superficial. If the diagnostic imaging device 2 determines that the calcified region is in contact with the lumen, it determines that the calcification is superficial.

[0095] In this way, the diagnostic imaging apparatus 2 determines the characteristics of calcification depending on the position of the calcified region within the region representing the vascular wall.

[0096] Other than the above, this modification is the same as the third embodiment, and therefore detailed explanations of the flowchart and other details will be omitted.

[0097] Fourth Embodiment In this embodiment, a mode for detecting an attenuating plaque region from a vascular tomographic image will be described.

[0098] 15 is an explanatory diagram showing an outline of the fourth embodiment. The outline of this embodiment will be described with reference to FIG.

[0099] The detection model 50 according to this embodiment detects attenuating plaque regions that attenuate ultrasonic waves in IVUS images, in addition to the external elastic lamina in the main trunk of the blood vessel. Specifically, as shown in Fig. 15 , when a vascular tomographic image is input, the detection model 50 detects the attenuating plaque region as a detection target region, and also detects a region indicating the inside of the external elastic lamina of the blood vessel as a region different from the detection target region.

[0100] The server 1 generates a detection model 50 using training data in which labels indicating the position of the correct attenuating plaque region and the position of the region indicating the inside of the external elastic lamina of the blood vessel are added to a training vascular tomographic image.

[0101] The diagnostic imaging device 2 detects an attenuating plaque region from a vascular tomographic image using the detection model 50 generated as described above. First, the diagnostic imaging device 2 inputs the vascular tomographic image into the detection model 50, thereby detecting an attenuating plaque region and a region showing the inside of the external elastic lamina of the blood vessel.

[0102] Next, the diagnostic imaging device 2 detects a low-intensity region with a brightness below a threshold outside the region representing the inside of the external elastic lamina of the blood vessel as an attenuating plaque region. For example, the diagnostic imaging device 2 sets a threshold based on the brightness of the region outside the external elastic lamina and detects the region below the threshold.

[0103] The diagnostic imaging device 2 determines the boundary of the attenuating plaque region by combining the attenuating plaque region (low brightness region) detected above with the attenuating plaque region detected by the detection model 50. The diagnostic imaging device 2 identifies the common region by performing an OR operation or an AND operation on both, and determines the boundary.

[0104] As in Embodiments 1 to 3, the diagnostic imaging device 2 displays the detection result of the detection target region (attenuating plaque region). In this embodiment, as in Embodiment 3, the diagnostic imaging device 2 specifies the angle occupied by the attenuating plaque region in the circumferential direction of the blood vessel (the angle when the center of gravity of the lumen of the blood vessel is the center) based on the boundary of the attenuating plaque region determined above, and displays a vascular tomographic image (transverse image) showing the specified angle as an arc (see FIG. 12 ).

[0105] The diagnostic imaging device 2 may simultaneously detect calcified regions along with attenuating plaque regions using the method described in embodiment 3. When displaying the angles occupied by the attenuating plaque regions and calcified regions in the circumferential direction of the blood vessel as arcs 131 (see FIG. 12 ), the diagnostic imaging device 2 preferably displays the arcs 131 indicating the angles of each region in different display modes (for example, different display colors).

[0106] 16 is a flowchart showing the procedure of the detection process for the detection target region according to embodiment 4. The control unit 21 of the image diagnostic apparatus 2 acquires a vascular tomographic image (step S421). The control unit 21 inputs the acquired vascular tomographic image into the detection model 50 to detect an attenuating plaque region and a region showing the inside of the external elastic lamina of the blood vessel (step S422).

[0107] The control unit 21 detects a low-brightness region, the brightness of which is equal to or less than a threshold, outside the region indicating the inside of the detected external elastic lamina of the blood vessel (step S423).The control unit 21 determines the boundary of the attenuating plaque region by combining the low-brightness region with the attenuating plaque region detected by the detection model 50 (step S424).

[0108] The control unit 21 identifies the angle occupied by the attenuating plaque region in the circumferential direction of the blood vessel based on the determined boundary of the attenuating plaque region (step S425).The control unit 21 displays a vascular tomographic image (transverse image) showing the identified angle (step S426), and ends the series of processes.

[0109] As described above, according to the fourth embodiment, the boundary of the attenuating plaque region can be suitably determined.

[0110] Fifth Embodiment In the first to fourth embodiments, the embodiments have been described in which the lumen, side branches, calcified regions, etc. of blood vessels are detected. In the present embodiment, a description will be given of a form in which the detection results of each region are presented (displayed).

[0111] 17 and 18 are explanatory diagrams showing an overview of embodiment 5. Fig. 17 and 18 show an example of an analysis screen of a vascular tomographic image displayed by the image diagnostic device 2 based on the detection results of each detection target region by the detection model 50. The overview of this embodiment will be described with reference to Fig. 17 and 18.

[0112] The diagnostic imaging device 2 captures vascular tomographic images (transverse images) at each position along the axial direction of the blood vessel in response to the pullback operation of the catheter 201. The diagnostic imaging device 2 detects various types of detection target regions from the vascular tomographic images at each position using the techniques described in the first to fourth embodiments. That is, the diagnostic imaging device 2 detects regions representing the lumen of the blood vessel, regions representing the inside of the external elastic lamina, regions representing side branches, plaque regions, calcification regions, etc., and determines their boundaries.

[0113] The analysis screens shown in FIGS. 17 and 18 use the detection results of the detection target region to display an image that allows the user (healthcare worker) to determine the placement position (landing zone) of the stent, etc.

[0114] The analysis screen includes a first image display field 210, a second image display field 220, a third image display field 230, and a fourth image display field 240. The first image display field 210 is a display field for displaying an OCT image (transverse cross-sectional image). The second image display field 220 is a display field for displaying an IVUS image (transverse cross-sectional image). The third image display field 230 and the fourth image display field 240 are display fields for displaying longitudinal cross-sectional images of blood vessels.

[0115] The diagnostic imaging device 2 generates (reconstructs) a longitudinal tomographic image from each transverse slice image (e.g., IVUS image) captured successively along the axial direction of the blood vessel, and displays the longitudinal tomographic image in the third image display field 230. The diagnostic imaging device 2 also displays, in the fourth image display field 240, longitudinal tomographic images in which the display color is changed or markers 241 are added depending on the amount of plaque and the presence or absence of side branches.

[0116] Specifically, the diagnostic imaging device 2 calculates the ratio of the plaque region to the cross section of the blood vessel at each position along the axial direction based on the area of ​​the plaque region and the area of ​​the region representing the blood vessel lumen in each transverse cross-sectional image captured consecutively along the axial direction of the blood vessel. This ratio is calculated, for example, by (area of ​​the plaque region) / (area of ​​the plaque region + area of ​​the region representing the blood vessel lumen). The diagnostic imaging device 2 displays the blood vessel wall at each position along the axial direction on the longitudinal cross-sectional image in a different display mode depending on the calculated ratio. For example, the diagnostic imaging device 2 displays blood vessel wall portions where the calculated ratio is equal to or greater than a threshold in a different display color from the remaining portions. For convenience, in Figures 17 and 18, the different display colors are illustrated by hatching.

[0117] Furthermore, the image diagnostic device 2 displays a marker 241 at each position on the longitudinal tomographic image depending on whether or not an area indicating a side branch of the blood vessel has been detected from each transverse tomographic image captured along the axial direction of the blood vessel. When a cursor 242 (described later) is positioned on the marker 241 on the longitudinal tomographic image, a marker 91 indicating a representative position of the side branch is displayed on the transverse tomographic image displayed in the first image display field 210 or the second image display field 220 (see FIG. 9 ).

[0118] In addition, although Figures 17 and 18 illustrate vertical bar-shaped markers 241 as markers 241 indicating side branches, the markers 241 may also be marks (e.g., triangular marks) indicating frames at representative positions of each side branch.

[0119] Furthermore, when the diagnostic imaging device 2 receives an operation input specifying a marker 241, it may generate and display a longitudinal tomographic image showing a longitudinal section taken along the center line of the side branch corresponding to the specified marker 241. Fig. 19 is an explanatory diagram showing a longitudinal tomographic image taken along the center line of the side branch. When the diagnostic imaging device 2 receives a specification input of the marker 241, the diagnostic imaging device 2 displays, as shown in Fig. 19, a longitudinal tomographic image in which the blood vessel is cut so as to pass through the center line of the main blood vessel (or the center line of the transverse image) and the center line of the side branch, with the center line of the side branch as the horizontal direction. This allows the user to properly understand the structure of the blood vessel.

[0120] 17 and 18 , the explanation will be continued. The diagnostic imaging apparatus 2 displays a cursor 242 on the longitudinal tomographic images displayed in the third image display field 230 and the fourth image display field 240. The cursor 242 is an object for specifying the position of the transverse images (OCT image and IVUS image) displayed in the first image display field 210 and the second image display field 220. The cursor 242 is initially positioned at the most constricted position of the blood vessel (the position where the diameter of the blood vessel lumen is smallest).

[0121] The diagnostic imaging device 2 receives an operation input from the user to move the cursor 242 along the axial direction (lateral direction). In accordance with the operation of moving the cursor 242, a transverse slice image corresponding to the position of the cursor 242 is displayed in the first image display field 210 and the second image display field 220, respectively.

[0122] The diagnostic imaging device 2 displays a transverse cross-sectional image corresponding to the position of the cursor 242 in the first image display field 210 and the second image display field 220, and also superimposes the boundary line indicating the lumen of the blood vessel, the boundary line indicating the external elastic lamina of the blood vessel, etc. on the transverse cross-sectional image. The diagnostic imaging device 2 also calculates the lumen diameter (L) of the blood vessel, the blood vessel diameter (diameter of the external elastic lamina, V), etc., and displays them in association with the transverse cross-sectional image.

[0123] Furthermore, when a calcified region or an attenuated plaque region is detected from the transverse image, the diagnostic imaging device 2 displays an arc 131 on the transverse image, which indicates the angle that the region occupies in the circumferential direction of the blood vessel. The diagnostic imaging device 2 also displays information such as the angle and thickness of the calcified region, etc., along with the arc 131. This allows appropriate consideration of a method for vessel preparation.

[0124] As described above, the user operates the cursor 242 to check the transverse cross-sectional image at each position and is presented with information such as the lumen diameter. The user also moves the reference cursors 243 and 244. Once the positions of the reference cursors 243 and 244 have been determined, the diagnostic imaging device 2 calculates and displays the distance between the reference cursors 243 and 244, as shown in Fig. 18. This allows the placement position (landing zone) of the stent to be determined, and the stent size to be appropriately determined.

[0125] FIG. 20 is a flowchart showing the procedure for displaying a vascular image. It is assumed that the diagnostic imaging device 2 detects a detection target region using the detection model 50 in advance and detects the vascular lumen, side branches, plaque regions, and the like from the vascular tomographic image. The control unit 21 of the diagnostic imaging device 2 calculates the ratio of the plaque region to the cross section of the blood vessel at each position along the axial direction based on the area of ​​the plaque region and the area of ​​the region representing the vascular lumen in each transverse cross-sectional image captured consecutively along the axial direction of the blood vessel (step S601). Based on each transverse cross-sectional image captured consecutively along the axial direction of the blood vessel, the control unit 21 generates and displays a longitudinal cross-sectional image that displays the vascular wall at each position along the axial direction in a different display mode (e.g., different display colors) according to the ratio calculated in step S601 (step S602). Specifically, as described above, the control unit 21 changes the display color of the vascular wall according to the ratio of the plaque region, and generates and displays a longitudinal cross-sectional image that displays a marker 241 at each position along the axial direction depending on whether a region representing a side branch of the blood vessel has been detected in each transverse cross-sectional image.

[0126] The control unit 21 receives an operation input to move the cursor 242 displayed on the longitudinal tomographic image along the axial direction (step S603), and displays the transverse tomographic image corresponding to the position to which the cursor 242 has been moved in the first image display field 210 and the second image display field 220 (step S604).

[0127] The control unit 21 receives an operation input to move the reference cursors 243 and 244 displayed on the longitudinal tomographic image along the axial direction (step S605). The control unit 21 calculates and displays the distance between the reference cursors 243 and 244 (step S606), and ends the series of processes.

[0128] As described above, according to the fifth embodiment, it is possible to preferably support users (medical personnel) who perform intravascular treatment such as PCI.

[0129] The embodiments disclosed herein are illustrative in all respects and should not be considered limiting. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims.

[0130] The matters described in each embodiment can be combined with each other. Furthermore, the independent claims and dependent claims described in the claims can be combined with each other in any combination, regardless of the reference format. Furthermore, the claims use a format in which a claim references two or more other claims (multiple claim format), but this is not limited to this. A multiple claim (multi-multi claim) that references at least one other multiple claim may also be used.

[0131] REFERENCE SIGNS LIST 1 Server 11 Control unit 12 Main memory unit 13 Communication unit 14 Auxiliary memory unit P1 Program 2 Image diagnostic device 21 Control unit 22 Main memory unit 23 Communication unit 24 Display unit 25 Input unit 26 Image processing unit 27 Auxiliary memory unit P2 Program 50 Detection model

Claims

1. An information processing method in which a computer executes a process of acquiring data representing a vascular tomographic image, inputting the acquired data representing the vascular tomographic image into a model that has been trained to detect, for one or more detection targets, a detection target region representing the detection target and a region different from the detection target region, thereby detecting the detection target region and a region different from the detection target region, and determining the boundary of the detection target region based on the detected detection target region and the region different from the detection target region.

2. The information processing method of claim 1, wherein the detection target area includes at least one of an area indicating the inside of a stent, an area indicating the lumen of the main trunk of a blood vessel, an area indicating the inside of the external elastic lamina of the main trunk of a blood vessel, an area indicating the inside of the adventitia of the main trunk of a blood vessel, an area indicating the lumen of a side branch of a blood vessel, an area indicating the inside of the external elastic lamina of a side branch of a blood vessel, an area indicating the lumen of the main trunk and side branch of a blood vessel, and an area indicating the inside of the external elastic lamina of the main trunk and side branch of a blood vessel.

3. The information processing method according to claim 1, wherein the detection target region includes at least one of a plaque region, a thrombus region, a hematoma region, a calcification region, a dissection region, a crack region, an extravascular tissue region, and a device region.

4. The information processing method according to claim 1, wherein the model is a trained model that, when data representing a vascular tomographic image is input, detects the detection target area and the area excluding the detection target area by inputting data representing the acquired vascular tomographic image into the model, and determines the boundary of the detection target area by combining the detected detection target area with the area excluding the detection target area.

5. An information processing method as described in claim 1, wherein the detection target area includes an area indicating a side branch of a blood vessel; by inputting data indicating the acquired vascular tomographic image into the model, an area indicating a side branch of a blood vessel, an area indicating a main trunk of a blood vessel, and an area indicating the main trunk and side branches of a blood vessel are detected; an area indicating a side branch of a blood vessel is identified by excluding the area indicating the main trunk of the blood vessel from the detected area indicating the main trunk and side branches of the blood vessel; and a boundary of the area indicating the side branch of a blood vessel is determined by combining the identified area indicating the side branch of a blood vessel with the area indicating the side branch of a blood vessel detected by the model.

6. The information processing method according to claim 5, wherein the vascular tomographic image is a transverse image obtained by capturing a cross section intersecting with the axial direction of the blood vessel, and the transverse image is output with a marker indicating a representative position of the side branch superimposed thereon based on the boundary of the determined area indicating the side branch.

7. The information processing method according to claim 1, further comprising the steps of: detecting high-luminance areas or low-luminance areas included in data showing the vascular tomographic image; and determining a boundary of the detection target area based on the detection target area and areas different from the detection target area detected by the model, and the high-luminance areas or low-luminance areas.

8. An information processing method as described in claim 7, further comprising the steps of: detecting a calcified region, an area showing the inside of the external elastic lamina of the blood vessel, and an area showing the lumen of the blood vessel by inputting data showing the acquired vascular tomographic image into the model; identifying an area showing the blood vessel wall based on the detected area showing the inside of the external elastic lamina of the blood vessel and the area showing the lumen of the blood vessel; detecting a high-brightness area having a brightness equal to or greater than a threshold within the identified area showing the blood vessel wall; and determining the boundary of the calcified region by combining the high-brightness area with the calcified region detected by the model.

9. The information processing method according to claim 8, wherein the vascular tomographic image is a transverse image obtained by capturing a cross section intersecting with the axial direction of the blood vessel, the angle occupied by the calcified region in the circumferential direction of the blood vessel is identified based on the determined boundary of the calcified region, and the transverse image showing the identified angle is output.

10. The information processing method according to claim 8, further comprising determining characteristics of the calcification based on the position of the calcified region relative to the region representing the vascular wall.

11. The information processing method of claim 7, wherein the detection target area includes an attenuating plaque area, and by inputting data representing the acquired vascular tomographic image into the model, an attenuating plaque area and an area representing the inside of the external elastic lamina of the blood vessel are detected, a low-brightness area having a brightness below a threshold outside the detected area representing the inside of the external elastic lamina is detected, and the boundary of the attenuating plaque area is determined by combining the low-brightness area with the attenuating plaque area detected by the model.

12. The information processing method according to claim 11, wherein the vascular tomographic image is a transverse cross-sectional image captured of a cross section intersecting the axial direction of the blood vessel, and the angle occupied by the attenuating plaque region in the circumferential direction of the blood vessel is identified based on the determined boundary of the attenuating plaque region, and the transverse cross-sectional image showing the identified angle is output.

13. The information processing method according to claim 1, wherein the vascular tomographic image is a transverse cross-sectional image captured of a cross section intersecting the axial direction of the blood vessel, the detection target region includes a plaque region and a region showing the lumen of the blood vessel, and the ratio of the plaque region to the cross section of the blood vessel at each position along the axial direction is calculated based on the area of ​​the plaque region and the area of ​​the region showing the lumen of the blood vessel in each transverse cross-sectional image captured successively along the axial direction, and a longitudinal cross-sectional image is generated that displays the blood vessel wall at each position along the axial direction in a different display manner according to the ratio based on a plurality of the transverse cross-sectional images.

14. An information processing method as described in claim 13, wherein the detection target area further includes an area indicating a side branch of a blood vessel, and the longitudinal section image is generated with a predetermined marker attached to each position along the axial direction depending on whether or not an area indicating a side branch of a blood vessel is detected from each of the transverse section images.

15. A program causing a computer to execute a process of acquiring data representing a vascular tomographic image, inputting the acquired data representing the vascular tomographic image into a model that has been trained to detect, for one or more detection targets, a detection target region representing the detection target and a region different from the detection target region, thereby detecting the detection target region and a region different from the detection target region, and determining the boundary of the detection target region based on the detected detection target region and the region different from the detection target region.

16. An information processing device having a control unit, wherein the control unit: acquires data representing a vascular tomographic image; when the data representing the vascular tomographic image is input, detects the detection target region and a region different from the detection target region for one or more detection targets by inputting the acquired data representing the vascular tomographic image into a model that has been trained to detect the detection target region and a region different from the detection target region; and determines the boundary of the detection target region based on the detected detection target region and the region different from the detection target region.

Images