Program, information processing device, and information processing method

Abstract

A program according to one aspect is configured such that a plurality of tomographic images of a blood vessel which are produced using an image diagnosis catheter inserted into the blood vessel, each of the acquired tomographic images of the blood vessel is input into a trained model that has been trained so as to output information about a blood vessel when a tomographic images of the blood vessel is input, information about the blood vessel which corresponds to each of the tomographic images of the blood vessel is output, a histogram of a plaque area ratio in the blood vessel is produced on the basis of the output information about the blood vessel which corresponds to each of the tomographic images of the blood vessel, a stent length or a lesion length is acquired, a threshold value of the plaque area ratio is calculated on the basis of the relationship between the produced histogram and the acquired stent length or lesion length, and a processing for outputting the calculated threshold value of the plaque area ratio on a graph showing the relationship between the position of the longer axis and the plaque area ratio is executed by a computer (1).

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a medical image processing apparatus capable of obtaining spatial information in which information regarding blood vessels is accurately depicted from non-invasively obtained spatial information.

Prior Art Document

Patent Document

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the invention according to Patent Document 1 has a problem that it cannot assist in determining the position where a stent is to be placed.

[0005] On one aspect, it is to provide a program or the like capable of assisting in determining the position where a stent is to be placed.

Means for Solving the Problems

[0006] The program for one aspect involves acquiring multiple tomographic images of blood vessels generated using an imaging catheter inserted into a blood vessel, inputting each acquired tomographic image into a learning model trained to output information about blood vessels when a tomographic image of a blood vessel is input, outputting information about the blood vessel corresponding to each tomographic image, generating a histogram of the plaque area ratio in the blood vessel based on the outputted information about the blood vessel corresponding to each tomographic image, acquiring the stent length or lesion length, calculating a threshold for the plaque area ratio based on the relationship between the generated histogram and the acquired stent length or lesion length, and outputting the calculated threshold for the plaque area ratio on a graph showing the relationship between the long axis position and the plaque area ratio. [Effects of the Invention]

[0007] In one respect, it can help determine the optimal placement location for the stent. [Brief explanation of the drawing]

[0008] [Figure 1] This is an explanatory diagram showing an overview of the catheter system. [Figure 2] This is an explanatory diagram illustrating the configuration of a catheter system. [Figure 3] This is an explanatory diagram illustrating the vascular information output model. [Figure 4] This is an explanatory diagram showing an example of a graph illustrating the relationship between the average diameter of a blood vessel and its major axis position in a cross-sectional image of a blood vessel. [Figure 5] This is an explanatory diagram showing an example of a graph illustrating the relationship between plaque area ratio and its long-axis position in a cross-sectional image of a blood vessel. [Figure 6A] This is an explanatory diagram showing an example of a histogram of plaque area ratio. [Figure 6B] This is an explanatory diagram showing an example of a histogram of plaque area ratio. [Figure 6C] This is an explanatory diagram showing an example of a histogram of plaque area ratio. [Figure 7A]It is an explanatory diagram for explaining the process of deriving a function for obtaining the threshold of the plaque area ratio. [Figure 7B] It is an explanatory diagram for explaining the process of deriving a function for obtaining the threshold of the plaque area ratio. [Figure 8] It is an explanatory diagram showing an example of a screen for displaying the threshold of the plaque area ratio on a graph. [Figure 9] It is a flowchart showing the processing procedure when outputting the threshold of the plaque area ratio. [Figure 10] It is a flowchart showing the processing procedure of a subroutine for the process of obtaining information related to blood vessels. [Figure 11] It is a flowchart showing the processing procedure of a subroutine for the process of generating a histogram of the plaque area ratio. [Figure 12] It is a flowchart showing the processing procedure of a subroutine for the derivation process of a function for obtaining the threshold of the plaque area ratio. [Figure 13] It is an explanatory diagram showing an example of a graph showing the relationship between the plaque area ratio and the long axis position in Embodiment 2. [Figure 14] It is a flowchart showing the processing procedure when generating a histogram of the plaque area ratio for each lesion area. [Figure 15] It is a flowchart showing the processing procedure when generating a histogram of the plaque area ratio for each lesion area. [Figure 16A] It is an explanatory diagram for explaining the process of generating a histogram of the plaque area ratio corresponding to a target area. [Figure 16B] It is an explanatory diagram for explaining the process of generating a histogram of the plaque area ratio corresponding to a target area.

Embodiments for Carrying Out the Invention

[0009] Hereinafter, the present invention will be described in detail based on the drawings showing its embodiments.

[0010] (Embodiment 1) Embodiment 1 relates to a form of outputting a threshold value of the plaque burden in a blood vessel based on a tomographic image of the blood vessel generated using an image diagnostic catheter inserted into the blood vessel. When an image diagnostic catheter is inserted into a lumen organ such as a blood vessel and treatment for a lesion is performed, it is necessary to determine the placement position of a stent. Based on the determined placement position, the size of the stent can be determined. The placement position of the stent is determined based on the plaque burden. However, for lesions in actual clinical practice, it is difficult to determine the placement position of the stent even by referring to the plaque burden. In this embodiment, an optimal placement position of the stent can be determined using the threshold value of the plaque burden. The threshold value of the plaque burden will be described later.

[0011] FIG. 1 is an explanatory diagram showing an overview of the catheter system 10. The catheter system 10 of this embodiment includes an information processing device 1, an image diagnostic catheter 2, and an MDU (Motor Drive Unit) 3. The image diagnostic catheter 2 is connected to the information processing device 1 via the MDU 3.

[0012] The image diagnostic catheter 2 may be an IVUS (Intravascular Ultrasound) catheter used for generating an ultrasonic tomographic image from the inside of a blood vessel, that is, an image diagnostic catheter for generating an ultrasonic tomographic image, a catheter for generating an optical tomographic image such as for OCT (Optical Coherence Tomography) or OFDI (Optical Frequency Domain Imaging) using near-infrared light, or the like. Hereinafter, for simplicity, the image diagnostic catheter 2 will be read as the catheter 2.

[0013] The information processing device 1 is connected to a display device 4 and an input device 5. The input device 5 is, for example, an input device such as a keyboard, mouse, trackball, or microphone. The display device 4 is a liquid crystal display or an organic EL (electroluminescence) display, and displays images and information output from the information processing device 1. The display device 4 and the input device 5 may be stacked together to form a touch panel. The input device 5 and the information processing device 1 may be configured as a single unit.

[0014] The computer 1 according to this embodiment acquires multiple tomographic images of blood vessels generated using a catheter 2 inserted into the blood vessel. The computer 1 inputs each acquired tomographic image of a blood vessel into the blood vessel information output model 171, which will be described later, and outputs information about the blood vessel corresponding to each tomographic image of a blood vessel.

[0015] Computer 1 generates a histogram of plaque area ratios in each vessel based on the information about the vessel corresponding to the tomographic image of each vessel output. Computer 1 obtains the stent length or lesion length. Computer 1 calculates a threshold for plaque area ratio based on the relationship between the generated histogram and the obtained stent length or lesion length. Computer 1 outputs the calculated threshold for plaque area ratio on a graph showing the relationship between the long axis position and the plaque area ratio.

[0016] Figure 2 is an explanatory diagram illustrating the configuration of the catheter system 10. As mentioned above, the catheter system 10 includes an information processing device 1, a catheter 2, and an MDU 3. The information processing device 1 is a dedicated ultrasound diagnostic device, or a personal computer, tablet, or smartphone with ultrasound diagnostic device functionality. The information processing device 1 is connected to various imaging diagnostic devices 6, such as an X-ray angiography device, an X-ray CT (Computed Tomography) device, an MRI (Magnetic Resonance Imaging) device, a PET (Positron Emission Tomography) device, or an ultrasound diagnostic device, via an HIS (Hospital Information System), etc. For simplicity, the information processing device 1 will be read as computer 1 below.

[0017] Computer 1 comprises a control unit 11, a storage unit 12, a communication unit 13, a catheter control unit 14, a display unit 15, an input unit 16, a large-capacity storage unit 17, and a reading unit 18. Each component is connected by bus B.

[0018] The control unit 11 includes arithmetic processing units such as a CPU (Central Processing Unit), MPU (Micro-Processing Unit), and GPU (Graphics Processing Unit), and performs various information processing and control processing related to the computer 1 by reading and executing the control program 1P (information processing program, program product) stored in the memory unit 12. In Figure 2, the control unit 11 is described as a single processor, but it may be a multi-processor.

[0019] The storage unit 12 includes memory elements such as RAM (Random Access Memory) and ROM (Read Only Memory), and stores the control program 1P or data necessary for the control unit 11 to execute processing. The storage unit 12 also temporarily stores data necessary for the control unit 11 to execute arithmetic processing.

[0020] The communication unit 13 is an interface for communication between the computer 1 and the network. The catheter control unit 14 controls the MDU 3 and generates images based on sensor signals. Alternatively, the control unit 11 may perform the functions of the catheter control unit 14. The display unit 15 is an interface connecting the display device 4 and the bus. The input unit 16 is an interface connecting the input device 5 and the bus.

[0021] The large-capacity storage unit 17 includes a recording medium such as an HDD (Hard disk drive) or an SSD (Solid State Drive). The large-capacity storage unit 17 stores a vascular information output model 171. The vascular information output model 171 is an estimator that outputs (estimates) information about blood vessels based on cross-sectional images of blood vessels, and is a trained model generated by machine learning.

[0022] In this embodiment, the storage unit 12 and the large-capacity storage unit 17 may be configured as a single storage device. Furthermore, the large-capacity storage unit 17 may be composed of multiple storage devices. Moreover, the large-capacity storage unit 17 may be an external storage device connected to the computer 1.

[0023] The reading unit 18 reads a portable storage medium 1a, including a CD (Compact Disc)-ROM or DVD (Digital Versatile Disc)-ROM. The control unit 11 may read the control program 1P from the portable storage medium 1a via the reading unit 18 and store it in the large-capacity storage unit 17. Alternatively, the control unit 11 may download the control program 1P from another computer via a network N or the like and store it in the large-capacity storage unit 17. Furthermore, the control unit 11 may also read the control program 1P from the semiconductor memory 1b.

[0024] The MDU3 rotates its sensor and shaft inside the catheter 2. The catheter control unit 14 generates one image for each rotation of the sensor. The generated images are transverse images centered on the probe portion of the catheter 2 and approximately perpendicular to the probe portion. The MDU3 can also be moved forward and backward while rotating its sensor and shaft inside the catheter 2. By performing a pullback operation, which rotates the sensor while pulling it towards the MDU3 at a constant speed, the catheter control unit 14 continuously generates multiple transverse images approximately perpendicular to the catheter 2 at predetermined intervals.

[0025] Next, we will explain the process of outputting a threshold for the plaque area ratio in blood vessels. First, computer 1 performs segmentation processing on the tomographic image of the blood vessels using the blood vessel information output model 171, which will be described later.

[0026] Figure 3 is an explanatory diagram illustrating the vascular information output model 171. The vascular information output model 171 is used as a program module that is part of artificial intelligence software. The vascular information output model 171 is an estimator with a pre-constructed neural network that takes a cross-sectional image of a blood vessel as input and outputs a classification result that classifies the regions in the cross-sectional image of the blood vessel. The vascular information output model 171 is implemented using a segmentation network such as U-Net (Convolutional Networks for Biomedical Image Segmentation). The vascular information output model 171 includes a region extraction unit 71 and an output unit 72.

[0027] The region extraction unit 71 includes a multilayer encoder layer and a multilayer decoder layer connected thereafter. Each encoder layer includes a pooling layer and a convolutional layer. Semantic segmentation is used to assign a label to each pixel that makes up the input tomographic image of the blood vessel. That is, when a tomographic image of a blood vessel is input to the region extraction unit 71, the region extraction unit 71 assigns a label to each pixel, classifying it into the lumen region, the tunica media region, the collateral branch region, and other regions. The output unit 72 outputs the classification result, classifying each pixel of the tomographic image of the blood vessel into the lumen region, tunica media region, collateral branch region, and other regions to which a label has been assigned.

[0028] The vascular information output model 171 may also be composed of a SegNet model or an FCN (Fully Convolutional Network) model, etc.

[0029] Furthermore, the vascular information output model 171 is not limited to the segmentation network described above, but may be implemented using other models such as RCNN (Regions with Convolutional Neural Network), Fast RCNN, Faster RCNN, SSD (Single Shot Multibook Detector), YOLO (You Only Look Once), SVM (Support Vector Machine), Bayesian networks, Transformer networks, regression trees, or random forests.

[0030] Next, computer 1 acquires information about blood vessels based on the classification results output from the blood vessel information output model 171. The information about blood vessels includes the average lumen diameter, average vessel diameter, area, location of side branches, or plaque area ratio. The plaque area ratio (plaque Burden) is the ratio of plaque area to the vascular medial area (corresponding to the vessel area), and is expressed by the following formula (1). Plaque area ratio = (vascular area - lumen area) / vascular area …(1)

[0031] Specifically, computer 1 inputs a tomographic image of a blood vessel into a vascular information output model 171 and outputs a classification result that categorizes each pixel of the tomographic image of the blood vessel into the luminal region, the tunica media region, the collateral branch region, and other regions, with each pixel being labeled.

[0032] For example, computer 1 calculates the lumen area based on the number of pixels in the lumen region output from the vascular information output model 171. Computer 1 calculates the vascular area based on the number of pixels in the media region and the lumen region output from the vascular information output model 171. Based on the calculated lumen area and vascular area, computer 1 uses the above-described formula (1) to calculate the plaque area ratio corresponding to the tomographic image of the blood vessel.

[0033] Computer 1 calculates the average lumen diameter and average vessel diameter based on the calculated lumen area and vessel area. The average lumen diameter may be calculated using the lumen area and pi as follows: Average lumen diameter = 2 × √(lumen area / π), or it may be calculated using the output lumen region, minimum lumen diameter, and maximum lumen diameter as follows: Average lumen diameter = (minimum lumen diameter + maximum lumen diameter) / 2. The calculation process for the average vessel diameter is the same as that for the average lumen diameter, so the explanation is omitted.

[0034] Computer 1 obtains the position (e.g., coordinates) of a collateral branch based on the region of the collateral branch output from the vascular information output model 171.

[0035] It should be noted that the process is not limited to the acquisition of information about blood vessels as described above. For example, when a tomographic image of a blood vessel is input, a blood vessel information output model 171 may be constructed that directly outputs information about the blood vessel corresponding to the tomographic image of that blood vessel. In this case, computer 1 acquires multiple tomographic images of blood vessels generated using a catheter 2 inserted into the blood vessel. Computer 1 inputs each acquired tomographic image of a blood vessel into the blood vessel information output model 171, which has been trained to output information about blood vessels when a tomographic image of a blood vessel is input, and outputs information about the blood vessel corresponding to each tomographic image of a blood vessel.

[0036] Next, we will explain the process of calculating a threshold for the plaque area ratio in a given blood vessel based on information about that blood vessel.

[0037] Computer 1 acquires multiple tomographic images of blood vessels. Computer 1 inputs each acquired tomographic image of a blood vessel into the blood vessel information output model 171 and outputs a classification result that categorizes each pixel of the tomographic image of each blood vessel into the lumen region, tunica media region, collateral branch region, and other regions. Based on the output classification result, Computer 1 acquires information about the blood vessel corresponding to each tomographic image of the blood vessel (average lumen diameter, average vessel diameter, area, location of collateral branches, or plaque area ratio, etc.). Based on the acquired information about the blood vessel, Computer 1 generates a histogram of the plaque area ratio in that blood vessel.

[0038] A histogram is a graph that represents the plaque area ratio in a blood vessel as the number of tomographic images (frequency) of that blood vessel for each predetermined class width (interval size) of plaque area ratio. The horizontal axis of the histogram represents the plaque area ratio, and the vertical axis represents the number of tomographic images. The class width can be, for example, 10%, 5%, or 1%. Below, the histogram generation process will be explained based on Figures 4 to 7 (Figures 7A and 7B).

[0039] Figure 4 is an explanatory diagram showing an example of a graph illustrating the relationship between the average diameter of a blood vessel and its major axis position in a cross-sectional image of a blood vessel.

[0040] Computer 1 acquires multiple tomographic images of blood vessels. Computer 1 inputs each acquired tomographic image of a blood vessel into the blood vessel information output model 171 and outputs information about the blood vessel corresponding to each tomographic image. From the outputted information about the blood vessels, Computer 1 acquires the average lumen diameter and average vessel diameter corresponding to each tomographic image of a blood vessel. Based on the acquired average lumen diameter and average vessel diameter, Computer 1 generates a graph 91 showing the relationship between the average diameter of the blood vessel and its major axis position. Computer 1 outputs the generated graph 91 to the display device 4.

[0041] The display device 4 displays the graph 91 output from the computer 1. As shown in the figure, the horizontal axis of the graph 91 indicates the major axis position, and the unit is millimeters (mm). The vertical axis of the graph 91 indicates the average diameter of the blood vessels (average lumen diameter and average vessel diameter), and the unit is millimeters (mm). A line 11a showing the average vessel diameter and a line 11b showing the average lumen diameter are displayed on the graph 91.

[0042] Although Figure 4 illustrates an example of a line graph, it is not limited to this, and other graphs such as bar graphs may also be used. In this embodiment, graph 91 is shown as an example to explain the relationship between the average diameter of blood vessels and the position of the long axis, but the generation of graph 91 is not essential.

[0043] Figure 5 is an explanatory diagram showing an example of a graph illustrating the relationship between plaque area ratio and long-axis position in a cross-sectional image of a blood vessel. Computer 1 obtains the plaque area ratio corresponding to the cross-sectional image of each blood vessel from the blood vessel information output from the blood vessel information output model 171. Based on the obtained plaque area ratio, Computer 1 generates a graph 92 showing the relationship between plaque area ratio and long-axis position.

[0044] Computer 1 outputs the generated graph 92 to the display device 4. The display device 4 displays the graph 92 output from Computer 1. As shown in the figure, the horizontal axis of graph 92 indicates the major axis position, and the unit is millimeters (mm). The vertical axis of graph 92 indicates the plaque area ratio (%).

[0045] Computer 1 generates a histogram of the plaque area ratio in each blood vessel (Figures 6A, 6B, and 6C) based on the plaque area ratio corresponding to the tomographic image of each blood vessel.

[0046] Figures 6A, 6B, and 6C are explanatory diagrams showing examples of histograms of plaque area ratios. Figure 6A is an explanatory diagram showing an example of a histogram generated with the first class width. Figure 6B is an explanatory diagram showing an example of a histogram generated with the second class width. Figure 6C is an explanatory diagram showing an example of a histogram generated with the third class width. In the following, we will explain examples where the first class width is in 10% increments, the second class width is in 5% increments, and the third class width is in 1% increments.

[0047] Computer 1 obtains the class width of the plaque area ratio. For example, if the class width of the plaque area ratio is previously stored in the memory unit 12 or the large-capacity memory unit 17, Computer 1 obtains the class width from the memory unit 12 or the large-capacity memory unit 17. Alternatively, Computer 1 may accept input of the class width from the operator or other person.

[0048] Computer 1 generates a histogram of plaque area percentages based on the acquired class width. Computer 1 outputs the generated plaque area percentage histogram to the display device 4. The display device 4 displays the plaque area percentage histogram output from Computer 1. As shown in the figure, the horizontal axis of the histogram shows the plaque area percentage (%). The vertical axis of the histogram shows the number of tomographic images of the blood vessel for each predetermined class width of plaque area percentage, in units of images.

[0049] Furthermore, the generated plaque area ratio histogram can be automatically resized to change the class width. Specifically, Computer 1 adds the number of tomographic images (frequency) to the rightmost class in the histogram. Computer 1 calculates the stent length based on the added number of tomographic images. For example, if 30 tomographic images are acquired in a 1 mm interval in the longitudinal direction, 990 tomographic images will be acquired in a 33 mm interval in the longitudinal direction. Computer 1 determines whether two or more stent lengths are included in the same class. Stent lengths include, for example, 9 mm, 12 mm, 15 mm, 18 mm, 21 mm, 24 mm, 28 mm, 33 mm, or 38 mm. Note that the stent length may be determined according to the manufacturer of the stent specified by the user.

[0050] If Computer 1 determines that two or more stent lengths (e.g., 21 mm and 24 mm) are included in the same class, it changes the class width to one that represents a finer interval. For example, Computer 1 changes the class width from one with 10% increments to one with 5% increments. Note that for the class width change process described above, lesion length may be used instead of stent length. Computer 1 then regenerates a histogram of plaque area ratios based on the plaque area ratio according to the changed class width.

[0051] Figures 7A and 7B are explanatory diagrams illustrating the process of deriving the function for determining the threshold plaque area ratio. The function for determining the threshold plaque area ratio represents the relationship between the stent length and the threshold plaque area ratio. Figure 7A is an explanatory diagram illustrating the process of calculating sample points based on class widths. Figure 7A illustrates an example where the class width is in 10% increments.

[0052] Computer 1 calculates sample points to generate a regression curve representing the relationship between stent length and threshold plaque area ratio. The sample points are generated based on a histogram of plaque area ratios (Figure 6A). Specifically, Computer 1 adds the number of tomographic images to the class width of the plaque area ratio, starting from the rightmost class. Computer 1 calculates the major axis position based on the added number of tomographic images. Computer 1 calculates the plaque area ratio value corresponding to the calculated major axis position as a sample point.

[0053] For example, Computer 1 adds the number of tomographic images corresponding to a class width of "0.7 to 0.8" based on the plaque area ratio histogram shown in Figure 6A. For example, the added number of tomographic images is 50. Computer 1 calculates the long axis position based on the added number of tomographic images. For example, if 30 tomographic images are acquired in a 1 mm interval in the longitudinal direction, the added 50 tomographic images are calculated to be approximately 1.67 mm in the longitudinal direction. Computer 1 calculates the plaque area ratio value corresponding to the calculated long axis position of 1.67 mm as a sample point. As shown in the figure, sample point 12a is displayed on the graph.

[0054] Alternatively, Computer 1 adds up the number of tomographic images corresponding to a class width of "0.6 to 0.7". For example, the added number of tomographic images is 820. Computer 1 calculates the longitudinal position based on the added number of tomographic images. For example, if 30 tomographic images are acquired in a 1 mm interval in the longitudinal direction, the added 820 tomographic images are calculated to be approximately 27.33 mm in the longitudinal direction. Computer 1 calculates the plaque area ratio value corresponding to the calculated longitudinal position of 27.33 mm as a sample point. As shown in the figure, sample point 12b is displayed on the graph.

[0055] In this way, by adding the number of tomographic images sequentially from the rightmost class, the sample points corresponding to each class width can be calculated. Note that Figure 7A illustrates an example where the class width is in 10% increments, but this method can be applied similarly to other class widths. The larger the class width, the more sample points are calculated, making it possible to generate a more accurate regression curve. The regression curve will be explained in Figure 7B.

[0056] Figure 7B is an explanatory diagram illustrating the process of creating a regression curve using sample points. As shown in the figure, computer 1 uses the calculated sample points as a sequence of points to determine the correlation between stent length and the threshold for plaque area ratio, and generates an approximated regression curve, for example, using the least squares method. As shown in the figure, the generated regression curve 12c is displayed on the graph. The threshold for plaque area ratio can be determined using the function representing the regression curve.

[0057] The function representing the regression curve (the function for determining the threshold for plaque area ratio) is expressed by the following equation (2). Threshold for plaque area ratio = f(stent length) …(2)

[0058] Computer 1 calculates a threshold for the plaque area ratio when the stent length is input to equation (2) described above. Computer 1 outputs the calculated threshold for the plaque area ratio to the display device 4.

[0059] Although Figures 7A and 7B illustrate the calculation process for the threshold plaque area ratio relative to stent length, the same process can be applied to calculating the threshold relative to lesion length.

[0060] Figure 8 is an explanatory diagram showing an example of a screen displaying the plaque area ratio threshold on a graph. Note that the same reference numerals are used for content that overlaps with Figure 5, and explanations are omitted. Computer 1 calculates the plaque area ratio threshold using equation (2) described above. Computer 1 outputs the calculated plaque area ratio threshold on graph 92 and outputs graph 92, including the plaque area ratio threshold, to the display device 4. As shown in the figure, the display device 4 displays graph 92 output from computer 1. Additionally, an identification line 13a indicating the plaque area ratio threshold is displayed on graph 92.

[0061] Figure 9 is a flowchart showing the processing procedure for outputting a threshold for the plaque area ratio. The control unit 11 of the computer 1 acquires multiple tomographic images of blood vessels generated using the catheter 2 inserted into the blood vessel from the catheter control unit 14 (step S101). The control unit 11 then executes a subroutine for acquiring information about blood vessels using the blood vessel information output model 171 (step S102).

[0062] The control unit 11 executes a subroutine for generating a histogram of plaque area ratios (step S103). The control unit 11 also executes a subroutine for deriving a function for determining the threshold of the plaque area ratio (step S104). The subroutines for acquiring information about blood vessels, generating histograms, and deriving functions will be described later.

[0063] The control unit 11 obtains the stent length (step S105). For example, if the stent length has been previously stored in the storage unit 12 or the large-capacity storage unit 17, the control unit 11 may obtain the stored stent length from the storage unit 12 or the large-capacity storage unit 17. Alternatively, the control unit 11 may receive input of the stent length from the operator or the like via the input unit 16 from the input device 5. Note that instead of the stent length, the control unit 11 may obtain the lesion length.

[0064] The control unit 11 substitutes the acquired stent length (or lesion length) into the derived function to calculate the threshold for the plaque area ratio (step S106). The control unit 11 outputs the calculated threshold for the plaque area ratio to the display device 4 via the display unit 15 (step S107). The control unit 11 terminates the process.

[0065] Figure 10 is a flowchart showing the processing procedure of a subroutine for acquiring information about blood vessels. The control unit 11 acquires a tomographic image of one blood vessel from the tomographic images of multiple blood vessels acquired in the processing of step S101 (Figure 9) (step S01). The control unit 11 inputs the acquired tomographic image of the blood vessel into the blood vessel information output model 171 (step S02) and outputs a classification result that classifies the luminal region, tunica media region, collateral branch region, and other regions, with each pixel of the tomographic image of the blood vessel being labeled (step S03).

[0066] The control unit 11 acquires information about blood vessels based on the output classification results (step S04). The information about blood vessels includes the average lumen diameter, average vessel diameter, area (lumen area and vessel area), the location of side branches, or the plaque area ratio.

[0067] Specifically, the control unit 11 calculates the lumen area based on the number of pixels in the lumen region output from the vascular information output model 171. The control unit 11 calculates the vascular area based on the number of pixels in the media region and the lumen region output from the vascular information output model 171. Based on the calculated lumen area and vascular area, the control unit 11 uses the above-described equation (1) to calculate the plaque area ratio corresponding to the tomographic image of the blood vessel.

[0068] The control unit 11 calculates the average lumen diameter and average vessel diameter based on the calculated lumen area and vessel area. For example, the control unit 11 calculates the average lumen diameter using the lumen area and pi, with the formula: average lumen diameter = 2 × √(lumen area / π). The control unit 11 also calculates the average vessel diameter using the vessel area and pi, with the formula: average vessel diameter = 2 × √(vascular area / π). The control unit 11 obtains the position (e.g., coordinates) of the side branches based on the region of the side branches output from the vessel information output model 171.

[0069] The control unit 11 determines whether the tomographic image of a given blood vessel is the last tomographic image among the tomographic images of multiple blood vessels (step S05). If the control unit 11 determines that the tomographic image of the given blood vessel is the last tomographic image (YES in step S05), it outputs information about the blood vessel corresponding to each tomographic image (step S06). The control unit 11 terminates the subroutine for acquiring information about the blood vessels and returns. If the control unit 11 determines that the tomographic image of the given blood vessel is not the last tomographic image (NO in step S05), it returns to the process in step S01.

[0070] Figure 11 is a flowchart showing the processing procedure of a subroutine for generating a histogram of plaque area percentages. The control unit 11 of the computer 1 obtains the plaque area percentage corresponding to the tomographic image of each vessel from the information on the vessels output from the vascular information output model 171 (step S11). The control unit 11 obtains the class width (for example, in increments of 10%) of the plaque area percentages that has been stored in the storage unit 12 in advance (step S12). The control unit 11 may also accept the setting of the class width by input from the operator or the like via the input unit 16 from the input device 5.

[0071] The control unit 11 generates a histogram of plaque area ratios based on the acquired plaque area ratios according to the acquired class width (step S13). The control unit 11 adds the number of tomographic images in order from the rightmost class in the histogram (step S14). The control unit 11 calculates the stent length based on the added number of tomographic images (step S15). For example, if 1 mm corresponds to 30 tomographic images, then 990 tomographic images corresponds to approximately 33 mm.

[0072] The control unit 11 determines whether two or more stent lengths (for example, 21 mm and 24 mm) are included in the same class (step S16). If the control unit 11 determines that two or more stent lengths are not included in the same class (NO in step S16), it terminates the histogram generation subroutine and returns.

[0073] If the control unit 11 determines that two or more stent lengths are included in the same class (YES in step S16), it changes the class width to a class width that represents a finer interval (step S17). For example, the control unit 11 may change the class width from 10% increments to 5% increments. The control unit 11 returns to the process in step S13 and regenerates a histogram of plaque area ratios based on the plaque area ratio according to the changed class width.

[0074] Figure 12 is a flowchart showing the processing procedure of the subroutine for deriving the function for determining the threshold of the plaque area ratio. The control unit 11 calculates sample points for generating a regression curve that represents the relationship between the stent length and the threshold of the plaque area ratio (step S21). Specifically, the control unit 11 adds the number of tomographic images to the class width of the plaque area ratio, starting from the rightmost class. The computer 1 calculates the major axis position based on the added number of tomographic images. The computer 1 calculates the plaque area ratio value corresponding to the major axis position calculated on the graph as a sample point.

[0075] The control unit 11 uses the calculated sample points as a sequence of points to determine the correlation between the stent length and the threshold for the plaque area ratio, thereby deriving a function that represents an approximated regression curve, for example, using the least squares method (step S22). The control unit 11 then finishes the subroutine for function derivation and returns.

[0076] According to this embodiment, it is possible to output a threshold value for the plaque area ratio in a blood vessel based on a tomographic image of the blood vessel generated using the catheter 2.

[0077] (Embodiment 2) Embodiment 2 relates to a method for detecting lesion areas based on a threshold for the plaque area ratio in blood vessels. Note that explanations of content that overlaps with Embodiment 1 will be omitted.

[0078] Figure 13 is an explanatory diagram showing an example of a graph illustrating the relationship between plaque area ratio and major axis position in Embodiment 2. Note that the same reference numerals are used for elements that overlap with Figure 5, and their explanations are omitted. Computer 1 outputs a threshold value for the plaque area ratio in a given blood vessel based on tomographic images of multiple blood vessels. Note that the output process for the plaque area ratio threshold is the same as in Embodiment 1, and therefore its explanation is omitted.

[0079] Computer 1 detects lesion areas based on the outputted plaque area ratio threshold. Specifically, Computer 1 detects areas where the plaque area ratio is greater than or equal to the threshold. Computer 1 then detects multiple lesion areas based on a predetermined distance for each lesion area relative to the detected area. The predetermined distance for each lesion area is, for example, 3 mm.

[0080] Computer 1 determines whether the distance (difference) between plaque area ratio values ​​in an area that exceeds a threshold is less than or equal to a predetermined distance. If Computer 1 determines that the distance between plaque area ratio values ​​is less than or equal to the predetermined distance, it detects them as the same lesion area. If Computer 1 determines that the distance between plaque area ratio values ​​exceeds the predetermined distance, it detects them as different lesion areas. In this way, multiple lesion areas can be detected based on a predetermined distance for each lesion area.

[0081] Computer 1 outputs the plaque area threshold and multiple lesion areas on graph 92. Computer 1 also outputs the multiple lesion areas on graph 92 in a way that distinguishes each lesion area, for example, by color or pattern. Computer 1 outputs graph 92 to display device 4. Display device 4 displays graph 92 output from computer 1. As shown in the figure, an identification line 13a indicating the plaque area threshold and multiple lesion areas 13b are displayed on graph 92.

[0082] Furthermore, the placement positions of both ends of the stent can be determined based on the detected lesion area. Specifically, computer 1 determines the placement positions of both ends of the stent based on the plaque area ratio. The stent placement positions (landing zones) should preferably be in areas free of plaque and normal, but if there are no normal areas, they may be in areas where the plaque area ratio is below a predetermined threshold (e.g., 50%), or in the area with the smallest plaque area ratio.

[0083] Computer 1 outputs the determined placement positions of both ends of the stent onto graph 92 and outputs graph 92 to display device 4. Display device 4 displays graph 92 output from computer 1. As shown in the figure, the distal end 13c of the stent, which indicates the distal end of the stent, and the proximal end 13d of the stent, which indicates the proximal end of the stent, are displayed on graph 92.

[0084] Figures 14 and 15 are flowcharts showing the processing procedure for generating a histogram of plaque area ratio for each lesion area. Note that the same reference numerals are used for elements that overlap with Figure 9, and their explanations are omitted. After executing the process in step S106, the control unit 11 of computer 1 detects areas where the plaque area ratio exceeds a threshold (step S111). Alternatively, the control unit 11 may use a pre-set plaque area ratio threshold without executing steps S101 to S106.

[0085] The control unit 11 detects multiple lesion areas based on a predetermined distance (e.g., 2 mm) for each lesion area (step S112). Specifically, the control unit 11 determines whether the distance (difference) between the plaque area ratio values ​​for the detected area is less than or equal to the predetermined distance. If the control unit 11 determines that the distance between the plaque area ratio values ​​is less than or equal to the predetermined distance, it detects them as the same lesion area. If the control unit 11 determines that the distance between the plaque area ratio values ​​exceeds the predetermined distance, it detects them as different lesion areas.

[0086] The control unit 11 acquires one lesion region from among the multiple lesion regions it has detected (step S113). The control unit 11 executes a subroutine to generate a histogram of plaque area ratios for the acquired lesion region (step S114). The control unit 11 executes a subroutine to derive a function for determining the threshold of the plaque area ratio for the acquired lesion region (step S115). The control unit 11 acquires the stent length from the storage unit 12 (step S116). Alternatively, the control unit 11 may acquire the lesion length instead of the stent length. The control unit 11 calculates the threshold of the plaque area ratio by substituting the acquired stent length (or lesion length) into the derived function (step S117).

[0087] The control unit 11 determines whether the detected lesion area is the last lesion area from among the multiple lesion areas (step S118). If the control unit 11 determines that the lesion area is the last lesion area (YES in step S118), it outputs a histogram of the plaque area ratio generated for each lesion area to the display device 4 via the display unit 15 (step S119). The control unit 11 terminates the process. If the control unit 11 determines that the lesion area is not the last lesion area (NO in step S118), it returns to the process in step S113 and acquires the next lesion area after the lesion area.

[0088] According to this embodiment, it is possible to detect multiple lesion areas based on a threshold for plaque area ratio.

[0089] According to this embodiment, it is possible to generate a histogram of plaque area ratio for each lesion region.

[0090] According to this embodiment, displaying multiple lesion areas on a graph makes it possible to assist in selecting the appropriate stent size.

[0091] (Embodiment 3) Embodiment 3 relates to a method for generating a histogram of plaque area ratios corresponding to a target area set (specified) by the operator or other relevant party. Note that explanations of content that overlaps with Embodiments 1 and 2 will be omitted.

[0092] Figures 16A and 16B are explanatory diagrams illustrating the process of generating a histogram of the plaque area ratio corresponding to the target region. Figure 16A is an explanatory diagram showing an example of a graph illustrating the relationship between plaque area ratio and major axis position. Content that overlaps with Figure 13 is denoted by the same reference numerals and its explanation is omitted. Figure 16A includes the region setting field 14a. The region setting field 14a is a field for setting the target region for generating the histogram.

[0093] When computer 1 receives a setting operation in the area setting field 14a, it acquires the set target area. Computer 1 generates a histogram of plaque area ratios corresponding to the acquired target area. The histogram generation process is the same as in Embodiment 1, so the explanation is omitted. Computer 1 outputs the generated plaque area ratio histogram to the display device 4. The display device 4 displays the plaque area ratio histogram output from computer 1.

[0094] Figure 16B is an explanatory diagram showing an example of a histogram of plaque area ratio. The histogram shown in Figure 16B is generated based on the target area set in area setting field 14a in Figure 16A. Note that the histogram is the same as in Figures 6A, 6B, and 6C, so the explanation is omitted. If multiple target areas are set in area setting field 14a, computer 1 generates a histogram of plaque area ratio for each target area.

[0095] According to this embodiment, it is possible to generate a histogram of plaque area ratios corresponding to any target region.

[0096] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of the invention is indicated by the claims, not in the sense described above, and all modifications within the sense and scope equivalent to the claims are intended. [Explanation of Symbols]

[0097] 1. Information processing equipment (computer) 11 Control Unit 12 Storage section 13 Communications Department 14 Catheter Control Unit 15 Display 16 Input section 17 Mass storage 171 Vascular Information Output Model 71 Region extraction part 72 Output section 18 Reading section 1a Portable storage medium 1b Semiconductor memory 1P Control Program 2. Diagnostic imaging catheters (catheters) 4 Display device 5 Input devices 6. Diagnostic imaging equipment 10 Catheter Systems

Claims

1. Multiple cross-sectional images of blood vessels are obtained using an imaging catheter inserted into the blood vessel. A trained model, which is designed to output information about blood vessels when given a cross-sectional image of a blood vessel as input, is given the acquired cross-sectional image of each blood vessel as input, and the model outputs information about the blood vessel corresponding to each cross-sectional image. Based on the information about the blood vessels corresponding to the tomographic images of each blood vessel output, a histogram of the plaque area ratio in the said blood vessel is generated. Obtain the stent length or lesion length, Based on the relationship between the generated histogram and the acquired stent length or lesion length, the threshold for the plaque area ratio is calculated. The calculated threshold for plaque area ratio is output on a graph showing the relationship between the long axis position and the plaque area ratio. A program that instructs a computer to perform a process.

2. Multiple lesion regions are detected based on the threshold of the plaque area ratio. For each detected lesion area, a histogram of the plaque area ratio is generated. The program according to claim 1 that causes a process to be executed.

3. The detected lesion areas are displayed on the graph in a way that distinguishes them from each other. The program according to claim 2 that causes a process to be executed.

4. The class width of the aforementioned histogram is obtained, Based on the acquired class width, a function is derived that represents the relationship between stent length or lesion length and the threshold for plaque area ratio. A program according to claim 1 or 2 that causes a process to be executed.

5. The information relating to the blood vessels includes the average lumen diameter, average vessel diameter, area, location of side branches, or plaque area ratio. The program according to claim 1 or 2.

6. A first acquisition unit that acquires multiple tomographic images of blood vessels generated using an imaging diagnostic catheter inserted into a blood vessel, A first output unit takes a tomographic image of a blood vessel as input and outputs information about the blood vessel corresponding to that image, into a learning model that has been trained to output information about the blood vessel. A generation unit generates a histogram of the plaque area ratio in each blood vessel based on the information about the blood vessel corresponding to the tomographic image of each blood vessel output. A second acquisition unit for acquiring the stent length or lesion length, A calculation unit calculates a threshold for the plaque area ratio based on the relationship between the generated histogram and the acquired stent length or lesion length. A second output unit outputs the calculated threshold value of the plaque area ratio onto a graph showing the relationship between the long axis position and the plaque area ratio. An information processing device characterized by comprising:

7. Multiple cross-sectional images of blood vessels are obtained using an imaging catheter inserted into the blood vessel. A trained model, which is designed to output information about blood vessels when given a cross-sectional image of a blood vessel as input, is given the acquired cross-sectional image of each blood vessel as input, and the model outputs information about the blood vessel corresponding to each cross-sectional image. Based on the information about the blood vessels corresponding to the tomographic images of each blood vessel output, a histogram of the plaque area ratio in the said blood vessel is generated. Obtain the stent length or lesion length, Based on the relationship between the generated histogram and the acquired stent length or lesion length, the threshold for the plaque area ratio is calculated. The calculated threshold for plaque area ratio is output on a graph showing the relationship between the long axis position and the plaque area ratio. An information processing method characterized by executing a process.

Images