Program, information processing method, and information processing apparatus

Abstract

To provide a program and the like that calculate a representative value related to a blood vessel with high reproducibility.SOLUTION: A program causes a computer to execute processing of: acquiring a cross-sectional shape 57 of a blood vessel including the shape of a lumen region 41 of the blood vessel and the shape of a dissection flap region 42 present inside the lumen region 41; generating a post-correction cross-sectional shape 55 on the basis of the shape of the lumen region 41 and the shape of the dissection flap region 42; calculating the center of gravity 67 of a post-correction lumen region 48 in the post-correction cross-sectional shape 55; and calculating a representative value related to the blood vessel on the basis of the center of gravity 67.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

[0002] A program has been proposed that quantitatively outputs the angle of a dissected portion of a blood vessel based on a transverse cross-sectional image taken using an image acquisition catheter inserted into the blood vessel (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2022 / 209652 Summary of the Invention [Problem to be solved by the invention]

[0004] In the program of Patent Document 1, the angle of the part where dissection occurs is quantitatively calculated based on the center of the transverse slice image, i.e., the central axis of the imaging catheter. The angle of the part where dissection occurs is an example of a representative value related to the blood vessel. The representative value is a value that quantitatively expresses the state of the blood vessel.

[0005] However, the position of the imaging catheter inside the blood vessel changes because the inside of the blood vessel is thicker than the outer shape of the imaging catheter. Therefore, the reference point for evaluating the angle changes, making it difficult to obtain the angle of the dissection with high reproducibility.

[0006] In one aspect, an object of the present invention is to provide a program or the like that calculates representative values ​​related to blood vessels with high reproducibility. [Means for solving the problem]

[0007] The present invention relates to (1) acquiring a cross-sectional shape of a blood vessel, the cross-sectional shape including a shape of a lumen region of the blood vessel and a shape of a dissected flap region present inside the lumen region; generating a corrected cross-sectional shape based on the shape of the lumen region and the shape of the dissection flap region; Calculating a center of gravity of a corrected lumen region in the corrected cross-sectional shape; A representative value for the blood vessel is calculated based on the center of gravity. It is a program that causes a computer to execute processing.

[0008] Here, an embodiment of the present invention is (2) The program described in (1) above, The corrected cross-sectional shape is preferably generated by deforming the dissection flap region so that it conforms to the outer wall of the lumen region.

[0009] (3) The program described in (2) above, A boundary point corresponding to a boundary with the blood vessel is obtained from the dissection flap region; Calculating a first straight line that passes through the boundary point and is tangent to an outer wall of the lumen region; It is preferable that the corrected cross-sectional shape is generated by dividing the dissection flap region by a plurality of straight lines parallel to the first straight line, and translating each of the divided regions in a first direction parallel to the first straight line until they touch the inner surface of the blood vessel.

[0010] (4) The program described in (3) above, the border point corresponds to a point at which dissection of the dissection flap region from the blood vessel ends; The first direction is preferably a direction that returns the dissection flap region to a state prior to dissection from the blood vessel.

[0011] (5) The program described in (1) to (4) above, obtaining a cross-sectional image of the blood vessel; Displaying the cross-sectional image; The cross-sectional shape is preferably obtained based on a user tracing of the lumen region and the dissection flap region on the cross-sectional image.

[0012] (6) The program described in (1) to (5) above, The cross-sectional shape is preferably a schematic diagram created based on a cross-sectional image of the blood vessel.

[0013] (7) The program described in (1) to (6) above, It is preferable that the area of ​​the corrected lumen region and the area of ​​the corrected dissection flap region in the corrected cross-sectional shape are equal to the area of ​​the lumen region and the area of ​​the dissection flap region in the cross-sectional shape, respectively.

[0014] (8) The program described in (1) to (7) above, The representative value is preferably an angle formed by straight lines connecting the center of gravity with both ends of the range in which the dissection flap region is dissected from the blood vessel.

[0015] (9) The program according to (1) to (7) above, The representative value is preferably an angle formed by straight lines connecting the center of gravity with both ends of a lesion area occurring in the tissue of the blood vessel.

[0016] (10) The program according to (8) or (9) above, It is preferable that a transverse cross-sectional image of the blood vessel is acquired and the straight line and the angle are displayed by being superimposed on the transverse cross-sectional image.

[0017] (11) The program according to (9) above, The lesion area is preferably a calcified area or a plaque area.

[0018] (12) The information processing method is obtaining a cross-sectional shape of the blood vessel, the cross-sectional shape including a shape of a lumen region of the blood vessel and a shape of a dissected flap region present within the lumen region; generating a corrected cross-sectional shape based on the shape of the lumen region and the shape of the dissection flap region; Calculating a center of gravity of a corrected lumen region in the corrected cross-sectional shape; A representative value for the blood vessel is calculated based on the center of gravity. Preferably, the processing is computer implemented.

[0019] (13) The information processing device comprises: An information processing device including a control unit, The control unit is obtaining a cross-sectional shape of the blood vessel, the cross-sectional shape including a shape of a lumen region of the blood vessel and a shape of a dissected flap region present within the lumen region; generating a corrected cross-sectional shape based on the shape of the lumen region and the shape of the dissection flap region; Calculating a center of gravity of a corrected lumen region in the corrected cross-sectional shape; It is preferable to calculate a representative value for the blood vessel based on the center of gravity. Effect of the Invention

[0020] In one aspect, it is possible to provide a program or the like that calculates a representative value related to a blood vessel with high reproducibility. [Brief description of the drawings]

[0021] [Figure 1] FIG. 11 is an explanatory diagram illustrating an overview of a process for calculating a representative value. [Diagram 2] FIG. 2 is an explanatory diagram illustrating a configuration of an information processing device. [Diagram 3] FIG. 2 is an explanatory diagram for explaining a record layout of a tomographic image DB. [Figure 4] FIG. 2 is an explanatory diagram illustrating an outline of a procedure for creating classification data. [Diagram 5] FIG. 11 is an explanatory diagram for explaining a procedure for calculating the center of gravity. [Figure 6] FIG. 11 is an explanatory diagram for explaining a representative value. [Figure 7] FIG. 11 is an explanatory diagram for explaining a representative value. [Figure 8] 10 is a flowchart illustrating a process flow of a program for creating classification data. [Figure 9] 11 is a flowchart illustrating the flow of processing of a program for calculating and displaying a representative value. [Figure 10] 13 is a flowchart illustrating a process flow of a subroutine for generating corrected classification data. [Figure 11] This is an example screen. [Figure 12] 13 is an example of a graph showing representative values. [Figure 13] FIG. 13 is an explanatory diagram illustrating a carpet view. [Figure 14] FIG. 2 is an explanatory diagram for explaining a classification model. [Figure 15] FIG. 13 is an explanatory diagram for explaining a procedure for determining a boundary point according to the third embodiment. [Figure 16] FIG. 13 is an explanatory diagram for explaining a procedure for determining a boundary point according to the fourth embodiment. [Figure 17] FIG. 13 is an explanatory diagram illustrating a configuration of an information processing device according to a fifth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] [Embodiment 1] FIG. 1 is an explanatory diagram for explaining an overview of the process of calculating the representative value. A user such as a doctor inserts an image acquisition catheter into a blood vessel. The image acquisition catheter is for three-dimensional scanning, and can continuously capture multiple transverse cross-sectional images 58 with the scanning plane changed little by little in the axial direction. In the following explanation, the transverse cross-sectional images 58 created by one three-dimensional scanning may be referred to as a set of transverse cross-sectional images 58.

[0023] In addition, in Fig. 1, a transverse layer image 58 in a so-called XY (orthogonal coordinate) format constructed according to an actual shape is illustrated as an example. The transverse layer image 58 may be in a so-called RT (polar coordinate) format constructed by arranging scanning lines in parallel in the order of scanning angles. Conversion between the RT format and the XY format may be performed during the process described using Fig. 1. The method of conversion between the RT format and the XY format is well known, so the description will be omitted.

[0024] Based on each transverse cross-sectional image 58, classification data 57 is created. The classification data 57 is data obtained by classifying each pixel constituting the transverse cross-sectional image 58 into a plurality of regions, such as the lumen region 41, the dissection flap region 42, the vascular wall region 43, the catheter region 44, and the extraluminal region 45. The regions shown in Fig. 1 are examples of the regions constituting the classification data 57. The classification data 57 may include, for example, a guidewire region, a stent region, a calcification region 47 (see Fig. 7), a plaque region, and a dissection lumen region 46 (see Fig. 15), etc.

[0025] Based on the classification data 57, a classification image can be created in which the pixels constituting the transverse cross-sectional image 58 are colored differently for each region. A set of classification images is created based on a set of transverse cross-sectional images 58. The classification images obtained by visualizing the classification data 57 are illustrative diagrams showing the cross-sectional shapes of blood vessels. A method for creating the classification data 57 will be described in detail later.

[0026] As the classification data 57 is created, the dissection flap region 42 is extracted from each transverse layer image 58. The dissection flap region 42 is a region that exists inside the lumen region 41, where a part of the inner surface of the blood vessel wall is thinly dissected. Fig. 1 shows the dissection flap region 42 whose base is continuous with the blood vessel wall region 43 and whose end protrudes near the center of the lumen region 41.

[0027] The dissection flap region 42 moves inside the lumen region 41 due to, for example, contact with the imaging catheter and the influence of pulsation of blood flow. Therefore, when the dimensions of the dissection flap region 42 are quantified based on the classification data 57, the quantification cannot be performed with high reproducibility.

[0028] Both the vascular wall region 43 and the dissection flap region 42 are biological tissue regions 40 (see FIG. 16). However, the dissection flap region 42 does not function as a blood vessel, which is a blood flow path. Therefore, in the following description, the vascular wall region 43 may be referred to as a blood vessel.

[0029] The control unit 21 (see FIG. 2) moves the dissection flap region 42 so that it comes into contact with the inner surface of the vascular wall region 43, and creates corrected classification data 55. In FIG. 1, the corrected classification data 55 is illustrated using a corrected classification image in which pixels are colored differently for each region. The corrected classification image obtained by visualizing the corrected classification data 55 is an example of a corrected cross-sectional shape in this embodiment. Note that in the classification data 57 and the corrected classification data 55 shown in the lower right of FIG. 1, the extraluminal region 45 is omitted from illustration.

[0030] The shapes of the extraluminal region 45 and the vascular wall region 43 are the same in the classification data 57 and the corrected classification data 55. The corrected dissection flap region 49 shown in the corrected classification data 55 is the region into which the control unit 21 has deformed the dissection flap region 42. The area of ​​the corrected dissection flap region 49 is equal to the area of ​​the dissection flap region 42.

[0031] The corrected lumen region 48 shown in the corrected classification data 55 is a region surrounded by the vascular wall region 43 and the dissection flap region 42. A method in which the control unit 21 creates the corrected classification data 55 based on the classification data 57 will be described in detail later.

[0032] The control unit 21 calculates the center of gravity 67 of the corrected lumen region 48. Since the method of calculating the center of gravity of a planar figure is well known, detailed explanation will be omitted. The control unit 21 calculates a representative value related to the blood vessel based on the center of gravity 67.

[0033] The representative values ​​for blood vessels are, for example, the angle of the dissected area, the angle of the calcified portion, the angle of the plaque portion, etc. By using the center of gravity 67 as a reference, these representative values ​​can be calculated with high reproducibility.

[0034] An example of a method for evaluating the reliability of the calculated representative value will be described. For example, the control unit 21 calculates a three-dimensional approximation curve based on the position of the center of gravity 67 calculated from several transverse cross-sectional images 58 that sandwich the transverse cross-sectional image 58 to be evaluated. The control unit 21 may calculate an approximation line instead of the approximation curve. The approximation curve or approximation line can be calculated by, for example, the least squares method.

[0035] The control unit 21 calculates the distance between the center of gravity 67 calculated from the cross-sectional image 58 of the evaluation target and the approximation curve. If the distance is greater than a predetermined threshold, the control unit 21 determines that the accuracy of the center of gravity 67 and the representative value of the cross-sectional image 58 of the evaluation target is low.

[0036] The control unit 21 outputs the centroid 67 and the representative value determined to have low accuracy with a flag. When the control unit 21 displays the centroid 67 and the representative value with the flag on the display screen described later, the control unit 21 displays them in a different manner from the other centroids 67 and the representative values, for example, by changing the color or attaching a mark. The control unit 21 may calculate the representative value by using the intersection of the approximation curve and the transverse layer image 58 instead of the centroid 67 determined to have low accuracy.

[0037] The control unit 21 may not calculate a representative value for the cross-sectional image 58 including the center of gravity 67 determined to have low accuracy, and may output, for example, "N / A (Not Applicable)" instead of the representative value. The control unit 21 may add a flag to the cross-sectional image 58 or the classification data 57 including the center of gravity 67 determined to have low accuracy.

[0038] The image acquisition catheter used to capture the transverse cross-sectional image 58 is, for example, an IVUS (Intravascular Ultrasound) catheter used to generate an ultrasonic tomographic image from inside a blood vessel. The image acquisition catheter may be a catheter for generating an optical tomographic image, such as an OCT (Optical Coherence Tomography) catheter or an OFDI (Optical Frequency Domain Imaging) catheter for generating an optical tomographic image using near-infrared light.

[0039] The imaging catheter is not limited to being used for three-dimensional scanning. If an imaging catheter without three-dimensional scanning capability is used, the transverse image 58 is taken at the position where the user inserts the imaging catheter.

[0040] 2 is an explanatory diagram for explaining the configuration of the information processing device 20. The information processing device 20 includes a control unit 21, a main memory device 22, an auxiliary memory device 23, a communication unit 24, a display unit 25, an input unit 26, and a bus. The control unit 21 is an arithmetic and control device that executes the program of this embodiment. The control unit 21 uses one or more central processing units (CPUs), graphics processing units (GPUs), multi-core CPUs, or the like. The control unit 21 is connected to each hardware unit constituting the information processing device 20 via the bus.

[0041] The main memory device 22 is a storage device such as an SRAM (Static Random Access Memory), a DRAM (Dynamic Random Access Memory), a flash memory, etc. The main memory device 22 temporarily stores information required during processing performed by the control unit 21 and programs being executed by the control unit 21.

[0042] The auxiliary storage device 23 is a storage device such as an SRAM, a flash memory, a hard disk, or a magnetic tape. A tomographic image DB (Database) 36, programs to be executed by the control unit 21, and various data required for executing the programs are stored in the auxiliary storage device 23. The tomographic image DB 36 may be stored in an external large-capacity storage device connected to the information processing device 20. The communication unit 24 is an interface that performs communication between the information processing device 20 and a network.

[0043] The display unit 25 is, for example, a liquid crystal display panel or an organic EL (electro-luminescence) panel, etc. The input unit 26 is, for example, a keyboard, a mouse, a pen tablet, etc. The display unit 25 and the input unit 26 may be stacked to form a touch panel.

[0044] The information processing device 20 is a general-purpose personal computer, a tablet, a mainframe computer, a virtual machine running on a mainframe computer, or a quantum computer. The information processing device 20 may be configured with hardware such as a plurality of personal computers or a mainframe computer that perform distributed processing. The information processing device 20 may be configured with a cloud computing system. The information processing device 20 may be configured with hardware such as a plurality of personal computers or a mainframe computer that operate in cooperation with each other.

[0045] 3 is an explanatory diagram for explaining the record layout of the tomogram DB 36. The tomogram DB 36 is a database in which transverse images 58 created by three-dimensional scanning and classification data 57 are recorded. The tomogram DB 36 has a 3D scan ID field, a tomogram number field, a tomogram field, and a classification data field. The tomogram field and the classification data field each have an RT format field and an XY format field.

[0046] The 3D scan ID field records a 3D scan ID assigned to each three-dimensional scan. The slice number field records a number indicating the order of the cross-sectional images 58 created in one three-dimensional scan. The RT format field of the cross-sectional image field records the cross-sectional images 58 in RT format. The XY format field of the cross-sectional image field records the cross-sectional images 58 in XY format. The RT format field of the classification data field records classification data 57 in RT format. The XY format field of the classification data field records classification data 57 in XY format.

[0047] In addition, only the transverse cross-sectional image 58 in the RT format is recorded in the tomographic image DB 36, and the control unit 21 may create the transverse cross-sectional image 58 in the XY format by coordinate conversion as necessary. Instead of the DB in which the transverse cross-sectional image 58 is recorded, a DB in which data related to the scanning line before the transverse cross-sectional image 58 is created may be used.

[0048] Similarly, only one of the classification data 57 in the RT format and the classification data 57 in the XY format may be recorded in the tomographic image DB 36, and the control unit 21 may create the other classification data 57 by coordinate conversion as necessary. Instead of the transverse cross-sectional image 58, a DB in which data related to the scanning line before creating the transverse cross-sectional image 58 is recorded may be used. The tomographic image DB 36 may not have a tomographic image field, and only the classification data 57 created based on the transverse cross-sectional image 58 may be recorded.

[0049] In this embodiment, a case will be described as an example in which the tomographic image DB 36 is created in advance and recorded in the auxiliary storage device 23. Only the transverse images 58 may be recorded in the tomographic image DB 36, and the control unit 21 may create classification data 57 based on each of the set of transverse images 58 and record the classification data 57 in the tomographic image DB 36.

[0050] When the control unit 21 creates the classification data 57, a program for creating the classification data 57 based on the cross-sectional image 58 is stored in the auxiliary storage device 23 or an external large-capacity storage device connected to the information processing device 20.

[0051] 4 is an explanatory diagram for explaining an outline of a procedure for creating the classification data 57. In this embodiment, a case will be described in which a user skilled in interpreting transverse cross-sectional images 58 operates the information processing device 20 to manually create the classification data 57. The classification data 57 may be created using hardware different from the information processing device 20.

[0052] A large number of cross-sectional images 58 taken in various cases are recorded in the tomographic image field of the tomographic image DB 36. The control unit 21 extracts records in which classification data 57 have not yet been recorded from the tomographic image DB 36. The control unit 21 displays on the display unit 25 the cross-sectional images 58 in XY format recorded in the extracted records.

[0053] The user operates the input unit 26 to draw boundary lines separating the regions depicted in the cross-sectional image 58, thereby creating a trace diagram 56. For example, the control unit 21 superimposes the lines drawn by the user on the cross-sectional image 58 displayed on the display unit 25. The user can trace the cross-sectional image 58 while looking at the superimposed lines. The user specifies the classification of each region surrounded by the boundary lines drawn by tracing. In this way, classification data 57 in XY format corresponding to one cross-sectional image 58 is completed.

[0054] The control unit 21 may display the cross-sectional image 58 and the line drawn by the user side by side on the display unit 25. The user may create a so-called schema diagram by extracting features of the cross-sectional image 58 while looking at the cross-sectional image 58. The control unit 21 uses the schema diagram created by the user based on the cross-sectional image 58 for the classification data 57.

[0055] The control unit 21 creates the RT format classification data 57 based on the XY format classification data 57. The control unit 21 records the classification data 57 in the classification data field of the same record as the transverse cross-sectional image 58. With the above, one record of the tomographic image DB 36 is completed.

[0056] After the user creates classification data 57 corresponding to the first cross-sectional image 58 of a set of cross-sectional images 58, the control unit 21 may automatically generate boundary lines or classification data 57 based on the second and subsequent cross-sectional images 58 using a known automatic tracking technique or the like, and display them on the display unit 25. The user corrects the displayed boundary lines or classification data 57 as necessary to complete the classification data 57.

[0057] 5 is an explanatory diagram for explaining the procedure for calculating the center of gravity 67. The procedure for calculating the center of gravity 67 for one piece of classification data 57 recorded in the tomographic image DB 36 will be explained using the diagrams labeled (a) to (f) in FIG. 5 in order.

[0058] The control unit 21 acquires one piece of classified data 57 from the tomographic image DB 36. The acquired classified data 57 is shown in (a). The classified data 57 includes a lumen region 41, a dissection flap region 42, a vascular wall region 43, a catheter region 44, and an extraluminal region 45. The dissection flap region 42 has a base that is continuous with the vascular wall region 43, and an end that protrudes near the center of the lumen region 41.

[0059] As shown in (b), the control unit 21 determines a boundary point 66. The boundary point 66 is the boundary between the dissection flap region 42 and the vascular wall region 43, and corresponds to the dissection end point where dissection between the vascular wall region 43 and the dissection flap region 42 ends.

[0060] For example, the control unit 21 determines, among the end points of the boundary between the base of the dissection flap region 42 and the vascular wall region 43, the end point closer to the tip of the dissection flap region 42, i.e., the edge-like portion at the end of the dissection flap region 42, as the boundary point 66. The control unit 21 may determine, among the end points of the boundary between the base of the dissection flap region 42 and the vascular wall region 43, the one that forms a larger angle between the inner periphery of the vascular wall region 43 and the outer periphery of the dissection flap region 42, as the boundary point 66. The control unit 21 may determine the boundary point 66 so that the line connecting the boundary points 66 in one set of classification data 57 is the shortest.

[0061] As shown in (c), the control unit 21 determines an approximation circle 68 that approximates the inner circumference of the vascular wall region 43 in the vicinity of the boundary point 66. As shown in (d), the control unit 21 determines a first straight line 61 that is a tangent to the approximation circle 68 and passes through the boundary point 66. The first straight line 61 is a straight line that is tangent to the outer wall of the lumen region 41, i.e., the inner surface of the vascular wall region 43, at the boundary point 66.

[0062] The controller 21 determines the second straight line 62 at the initial position where the first straight line 61 does not overlap with the dissociation flap region 42 when the first straight line 61 is translated toward the tip side of the dissociation flap region 42. As a result, the dissociation flap region 42 is sandwiched between the first straight line 61 and the second straight line 62.

[0063] The control unit 21 divides the dissection flap region 42 by a plurality of straight lines parallel to the first straight line 61 to form rectangular divided regions each having a very small width. As shown by the thick arrow in (d), the control unit 21 translates each divided region in a first direction parallel to the first straight line 61 to bring the end of the divided region into contact with the inner circumference of the vascular wall region 43. Here, the first direction is the direction in which the dissection flap region 42 is returned to the state before being dissected from the vascular wall region 43. The entire dissection flap region 42 moves toward the inner surface of the vascular wall region 43, and becomes the state of the corrected dissection flap region 49 in (e).

[0064] The corrected dissection flap region 49 is a portion where the dissection flap region 42 is deformed to fit along the inner surface of the vascular wall region 43, i.e., the outer wall of the lumen region 41. As shown in (e), one end of the corrected dissection flap region 49 is at approximately the same position as the boundary point 66.

[0065] The boundary line between the vascular wall region 43 and the corrected dissection flap region 49 is interpreted as corresponding to the dissection surface formed by dissection of the vascular wall region 43 and the vascular wall region 43. Therefore, the intersection point between the inner circumference of the vascular wall region 43 and the second straight line 62 is described as a provisional dissection start point 65.

[0066] The control unit 21 determines the area surrounded by the vascular wall region 43 and the corrected dissection flap region 49 as the corrected lumen region 48. In other words, the catheter region 44 in the classification data 57 is integrated into the corrected lumen region 48. In this manner, the corrected classification data 55 is completed. As shown in (f), the control unit 21 calculates the center of gravity 67 of the corrected lumen region 48.

[0067] Fig. 6 is an explanatory diagram for explaining the representative value. Using Fig. 6, the dissociation angle θ1, which is a representative value indicating the range in which dissociation has occurred, will be explained. The control unit 21 calculates the dissociation angle θ1, which is the angle formed by the first measurement line 641 connecting the center of gravity 67 and the boundary point 66, and the second measurement line 642 connecting the center of gravity 67 and the tentative dissociation start point 65. The method of calculating the angle formed by two straight lines on the same plane is well known, so the explanation will be omitted.

[0068] 6, the representative value is calculated for the range in which the dissection flap region 42 is dissected from the vascular wall region 43. The dissection angle θ1 is the angle between the straight lines connecting the center of gravity 67 and both ends of the dissection range.

[0069] Fig. 7 is an explanatory diagram for explaining the representative value. The calcification angle θ2, which is a representative value representing the calcified range, will be explained using Fig. 7. In Fig. 7, a calcified region 47 occurs inside a vascular wall region 43. The range of the calcified region 47 is determined by the user, similarly to the dissection flap region 42 and the vascular wall region 43. The control unit 21 may determine the calcified region 47 by extracting a high-brightness region from a transverse slice image 58.

[0070] The control unit 21 calculates the calcification angle θ2, which is the angle between a first measurement line 641 connecting the center of gravity 67 and one end of the calcified region 47, and a second measurement line 642 connecting the center of gravity 67 and the other end of the calcified region 47. In Fig. 7, the target for calculating the representative value is the calcified region 47. The calcification angle θ2 is the angle between the straight lines connecting the center of gravity 67 and both ends of the calcified region 47.

[0071] In a similar manner, the control unit 21 may calculate a representative value for the range of any lesion region that has occurred in the tissue of the vascular wall region 43, such as a plaque region and an atheroma region. The control unit 21 may calculate a representative value for the range of a hematoma region. A hematoma region is a region that corresponds to a hematoma that is generated when blood that has invaded a dissection cavity region 46 (described below) from the lumen region 41 remains inside the vascular wall without returning to the lumen region 41. The control unit 21 may calculate a representative value for a treatment portion, such as a region of poor adhesion between a stent and a vascular wall.

[0072] Fig. 8 is a flowchart illustrating the process flow of a program for creating classified data 57. The program in Fig. 8 is a program for creating classified data 57 according to the procedure described using Fig. 4. As in Fig. 4, the program for creating classified data 57 will be described taking as an example a case where it is executed by information processing device 20.

[0073] The control unit 21 acquires one transverse image 58 from the tomographic image DB 36 (step S501). Specifically, the control unit 21 extracts a record in which classification data 57 has not yet been recorded from the tomographic image DB 36. The control unit 21 acquires the transverse image 58 in the XY format recorded in the extracted record.

[0074] The control unit 21 displays the cross-sectional image 58 on the display unit 25 (step S502). The control unit 21 accepts the drawing of a boundary line by the user via the input unit 26 (step S503). The control unit 21 accepts the classification designation by the user for the area surrounded by the boundary line via the input unit 26 (step S504).

[0075] The control unit 21 displays the areas for which the classification designation has been accepted on the display unit 25 in a manner such as coloring the areas with a color determined for each classification (step S505). The user checks whether the areas have been designated as intended, and performs an operation to instruct the end when designation of all areas has been completed.

[0076] The control unit 21 determines whether or not an instruction to end the process has been received from the user (step S506). If it is determined that an instruction to end the process has not been received (NO in step S506), the control unit 21 returns to step S504. Although not shown in the flowchart, the control unit 21 may receive an instruction to return to step S503.

[0077] When it is determined that the instruction to end has been received (YES in step S506), the control unit 21 generates classification data 57 in RT format based on the classification data 57 in XY format. The control unit 21 records the classification data 57 in the classification data field of the record from which the cross-sectional image 58 was acquired in step S501 (step S507). Thereafter, the control unit 21 ends the process.

[0078] 9 is a flowchart explaining the processing flow of a program that calculates and displays a representative value. The control unit 21 acquires one piece of classification data 57 from a set of classification data 57 recorded in the tomographic image DB 36 (step S521). The control unit 21 starts a subroutine for generating corrected classification data (step S522). The subroutine for generating corrected classification data is a program for generating corrected classification data 55 from classification data 57 according to the procedure explained using FIG. 5. The processing flow of the subroutine for generating corrected separated data will be described later.

[0079] The control unit 21 calculates the center of gravity 67 of the corrected lumen region 48 based on the corrected classification data 55 (step S523). The control unit 21 generates a first measurement line 641 connecting the center of gravity 67 and one end of the object for which the representative value is calculated (step S524). The control unit 21 generates a second measurement line 642 connecting the center of gravity 67 and the other end of the object for which the representative value is calculated (step S525).

[0080] The control unit 21 calculates a representative value, which is an angle between the first measurement line 641 and the second measurement line 642 (step S526). The control unit 21 records the representative value and the fault number of the classification data 57 acquired in step S521 in the main storage device 22 or the auxiliary storage device 23 in association with each other (step S527).

[0081] The control unit 21 determines whether the processing of one set of classification data 57 is completed (step S528). If it is determined that the processing is not completed (NO in step S528), the control unit 21 returns to step S521. If it is determined that the processing is completed (YES in step S528), the control unit 21 displays the representative value on the display unit 25 (step S529). After that, the control unit 21 ends the processing.

[0082] 10 is a flowchart illustrating the process flow of the corrected classification data generation subroutine. The corrected classification data generation subroutine is a program that generates corrected classification data 55 from classification data 57 according to the procedure described using FIG.

[0083] The control unit 21 extracts a boundary line between the dissection flap region 42 and the vascular wall region 43 from the classification data 57 being processed (step S541). The control unit 21 determines a boundary point 66, which is one end point of the boundary line (step S542). The method of determining the boundary point 66 may be any of the methods exemplified using FIG. 5, and therefore description of the flowchart will be omitted.

[0084] The control unit 21 creates an approximation circle 68 that approximates the inner circumference of the vascular wall region 43 near the boundary point 66 (step S543). Specifically, the control unit 21 creates the approximation circle 68 such that the circle has a center closer to the lumen region 41 than the boundary point 66, passes through the boundary point 66, and the error with respect to the inner circumference of the vascular wall region 43 within a predetermined range near the boundary point 66 is minimized.

[0085] The control unit 21 creates a first straight line 61 that is a tangent to the approximation circle 68 and passes through the boundary point 66 (step S544). The control unit 21 creates a second straight line 62 at the last position where the first straight line 61 does not overlap with the dissection flap region 42 when the first straight line 61 is translated toward the end point of the dissection flap region 42 (step S545). The control unit 21 calculates a tentative dissection start point 65 that is the intersection point between the inner circumference of the vascular wall region 43 and the second straight line 62 (step S546).

[0086] The control unit 21 divides the dissection flap region 42 into rectangular divided regions having minute widths by a plurality of straight lines parallel to the first straight line 61 (step S547). The control unit 21 selects one divided region (step S548). The control unit 21 moves the selected divided region in a direction parallel to the first straight line 61 to just before the vascular wall region 43 (step S549).

[0087] Specifically, the control unit 21 changes the classification of the pixels included in the divided region selected in step S548 from the dissection flap region 42 to the lumen region 41. The control unit 21 changes the classification of the pixels included in the range obtained by moving the divided region in the direction parallel to the first straight line 61 to just before the vascular wall region 43 to the corrected dissection flap region 49.

[0088] The control unit 21 determines whether or not the processing of the divided regions divided in step S547 is completed (step S550). If it is determined that the processing is not completed (NO in step S550), the control unit 21 returns to step S548. If it is determined that the processing is completed (YES in step S550), the control unit 21 ends the processing.

[0089] 11 is an example of a screen. A center of gravity 67, a first measurement line 641, a second measurement line 642, and a message indicating that the angle between the first measurement line 641 and the second measurement line 642 is θ1 are superimposed on a single cross-sectional image 58. A position field 71 and a feature field 72 are arranged at the top of the screen.

[0090] The position field 71 indicates which of the set of transverse cross-sectional images 58 is displayed using the distance from the start position of the three-dimensional scan. The feature amount field 72 indicates the angle of the portion θ1.

[0091] Note that a plurality of feature amounts calculated based on one transverse image 58 may be displayed on one screen. When a plurality of feature amounts are displayed, for example, positions corresponding to the respective feature amounts are superimposed on the transverse image 58, and a plurality of feature amount fields 72 are displayed on the screen. As a specific example, for example, the angle of the hematoma region may be displayed alongside the dissection angle θ1.

[0092] As described above, a case where a flag indicating low accuracy is attached to the center of gravity 67, the representative value, or the cross-sectional image 58 will be described. The control unit 21 displays the center of gravity 67 and the representative value in a manner different from other centers of gravity 67 and representative values, for example, by changing the color or attaching a mark. The control unit 21 may display a mark indicating low accuracy on the screen.

[0093] The classification data 57 may be displayed, for example, in a light color, superimposed on the cross-sectional image 58. The classification data 57 may be displayed next to the cross-sectional image 58. The user can easily confirm the basis for calculating the feature amount.

[0094] The control unit 21 may receive an instruction from the user to change the angle of the first measurement line 641 or the second measurement line 642. The control unit 21 may calculate and display an angle based on the center of gravity 67 for a range specified by the user.

[0095] FIG. 12 is an example of a graph displaying representative values. The horizontal axis indicates the position of the transverse cross-sectional image 58 as a distance from the start position of the three-dimensional scan. The unit of the horizontal axis is millimeters. The vertical axis indicates the feature amount. The unit of the feature amount is angle. The feature amount for each transverse cross-sectional image 58 is displayed in the graph.

[0096] The portions at both ends of the graph where the feature values ​​are zero mean that the target for which the feature values ​​are to be calculated is not included in the transverse image 58. The graph shown in Fig. 12 allows the user to easily grasp the change in the feature values ​​over the entire area where the 3D scan was performed.

[0097] 13 is an explanatory diagram for explaining the carpet view. In the following description, an image in the carpet view format is referred to as a carpet image 59. When an instruction to display the carpet image 59 is received from a user, the control unit 21 outputs the carpet image 59.

[0098] The horizontal direction of the carpet image 59 indicates the longitudinal direction of the blood vessel. The horizontal direction of the carpet image 59 indicates the angle based on the center of gravity 67. Black color means that a target region, such as a calcified region 47 or a plaque region, exists on a determination line 646 defined by the longitudinal position and angle.

[0099] An outline of a method for creating a carpet image 59 will be described below, taking as an example a case where the target region is a calcified region 47. As described above, the control unit 21 calculates the center of gravity 67 for the classification data 57. The control unit 21 determines an angle reference line 645 extending from the center of gravity 67 in, for example, an upward direction of the transverse slice image 58.

[0100] The control unit 21 determines whether or not a determination line 646 that passes through the center of gravity 67 and is separated from the angle reference line 645 by an angle θ passes through the calcified region 47, and records the slice number, the angle θ, and the determination result in association with each other. The control unit 21 repeats the determination and recording for θ from 0 degrees to 360 degrees. The control unit 21 executes the above processing for a set of transverse slice images 58.

[0101] The control unit 21 colors the longitudinal position corresponding to the tomography number and the position corresponding to the angle θ according to whether or not it is determined that a calcified region 47 exists. That is, a vertically elongated region surrounded by a two-dot chain line in Fig. 13 is formed based on one piece of classification data 57. In this way, a carpet image 59 showing the distribution of classification data 57 is completed with the center of gravity 67 as a reference.

[0102] The control unit 21 may determine whether the determination line 646 passes through each of a plurality of target regions, and display the carpet image 59 in a different color for each target region. For example, the control unit 21 can display the distribution of the calcified regions 47 and the distribution of the plaque regions on one carpet image 59.

[0103] When the determination line 646 passes through a plurality of destination regions, the control unit 21 colors the position corresponding to the determination line 646, for example, using a color corresponding to the destination region close to the center of gravity. The control unit 21 may also use a color different from the color defined for each of the destination regions that are passed through. For example, when red is assigned to the first destination region and blue is assigned to the second destination region, the control unit 21 may use purple, which is a mixture of red and blue. By using a mixture of colors, a display that is easy for the user to intuitively understand is realized. The control unit 21 may also use a color different from the mixed color, such as yellow.

[0104] According to this embodiment, the representative value of the blood vessel can be calculated with high reproducibility. The representative value is not limited to the angle. For example, the minimum or maximum distance between the center of gravity 67 and the inner circumferential surface of the blood vessel wall region 43 may be used as the representative value.

[0105] The center of gravity 67 is not limited to the center of gravity of the corrected lumen region 48. For example, it may be the center of gravity of the vascular wall region 43, the center of gravity of the external elastic lamina, or the center of gravity of the stent region. Note that it is preferable that the center of gravity 67 is the center of gravity of a region surrounding the entire circumference of the imaging catheter.

[0106] Note that a single transverse slice image 58 may include multiple dissection flap regions 42. In such a case, the control unit 21 calculates the center of gravity 67 after performing the above-mentioned process for each dissection flap region 42. The control unit 21 calculates a representative value based on the calculated center of gravity 67. The control unit 21 may calculate a representative value associated with each dissection flap region 42.

[0107] [Embodiment 2] This embodiment relates to a representative value calculation method for automatically generating classification data 57 from a cross-sectional image 58 using a classification model 31 generated by machine learning. Explanation of parts common to the first embodiment will be omitted.

[0108] 14 is an explanatory diagram for explaining the classification model 31. The classification model 31 is a model that receives an input of a transverse cross-sectional image 58 and assigns, to each pixel constituting the transverse cross-sectional image 58, a label relating to an area depicted in the pixel. The classification model 31 may be a model that assigns a label to each small area composed of a plurality of pixels. The classification model 31 is generated by a known machine learning method such as semantic segmentation using the tomographic image DB 36 as training data.

[0109] The classification model 31 may be configured by combining a plurality of models, such as a model that outputs the probability that each pixel is a lumen region 41, a model that outputs the probability that each pixel is not a lumen region 41, a model that outputs the probability that each pixel is a dissection flap region 42, and a model that outputs the probability that each pixel is not a dissection flap region 42. The classification model 31 may be configured to output a sequence that lists, for each pixel, the probability that each pixel is a lumen region 41, the probability that each pixel is not a lumen region 41, the probability that each pixel is not a dissection flap region 42, the probability that each pixel is not a dissection flap region 42, and the like.

[0110] 4 and 7, the classification data 57 can be automatically generated from the cross-sectional images 58. For example, in a clinical setting, the classification data 57 can be generated in parallel while a set of cross-sectional images 58 is captured, thereby realizing the information processing device 20 that displays highly reproducible features in real time.

[0111] For a user who is not accustomed to interpreting the transverse image 58, it is difficult to quickly create the trace image 56 based on the transverse image 58 as described with reference to Fig. 4. For such a user, even if the user observes the transverse image 58 in which dissection has occurred in the blood vessel, it is difficult to imagine the shape of the lumen region 41 before the dissection has occurred. Therefore, it is difficult to appropriately determine the position of the center of gravity 67 and to obtain a feature amount with high reproducibility.

[0112] According to this embodiment, by using the classification model 31, even a user who is not familiar with interpreting the transverse cross-sectional image 58 can easily confirm highly reproducible feature amounts.

[0113] [Embodiment 3] This embodiment relates to a representative value calculation method for generating classification data 57 using a classification model 31 capable of classifying a dissection cavity region 46. Explanation of parts common to the second embodiment will be omitted.

[0114] Fig. 15 is an explanatory diagram for explaining the procedure for determining the boundary point 66 in the embodiment 3. The procedure for determining the center of gravity 67 in the classification data 57 will be explained using the diagrams marked with (a) and (b) in Fig. 15 in order.

[0115] The control unit 21 acquires one transverse image 58 from the tomographic image DB 36 and inputs it to the classification model 31. Classification data 57 shown in (a) is output from the classification model 31. The classification model 31 used in this embodiment is capable of classifying the lumen region 41, the dissection flap region 42, the vascular wall region 43, the catheter region 44, and the extraluminal region 45, as well as the dissection lumen region 46.

[0116] As shown in (b), the control unit 21 determines, among the end points of the boundary line between the dissection flap region 42 and the vascular wall region 43, the end point that contacts the dissection cavity region 46 as the boundary point 66. Note that, if the dissection cavity region 46 is not classified, the control unit 21 determines the boundary point 66 in the same manner as in the first embodiment.

[0117] According to this embodiment, when the dissection cavity region 46 is classified, it is possible to easily determine which of the end points of the boundary line dividing the dissection flap region 42 and the vascular wall region 43 should be determined as the boundary point 66.

[0118] [Embodiment 4] This embodiment relates to a representative value calculation method for generating classification data 57 using a classification model 31 that does not distinguish between a dissection flap region 42 and a vascular wall region 43. Descriptions of parts common to the second embodiment will be omitted.

[0119] Fig. 16 is an explanatory diagram for explaining a procedure for determining a boundary point 66 in the fourth embodiment. The procedure for determining a center of gravity 67 in the classified data 57 will be explained using the diagrams labeled (a) to (c) in Fig. 16 in order.

[0120] The control unit 21 acquires one transverse image 58 from the tomographic image DB 36 and inputs it to the classification model 31. Classification data 57 shown in (a) is output from the classification model 31. The classification model 31 used in this embodiment does not distinguish between the dissection flap region 42 and the vascular wall region 43, and outputs classification data 57 in which both are classified as biological tissue region 40.

[0121] 16, the pixels constituting the transverse cross-sectional image 58 are classified into a biological tissue region 40, a lumen region 41, a catheter region 44, and a dissection lumen region 46. As shown in (b), the control unit 21 divides the biological tissue region 40 into a tubular blood vessel wall region 43 and a dissection flap region 42 protruding inward from the blood vessel wall region 43. Note that the illustration of the extraluminal region 45 is omitted from (b) onward.

[0122] As shown in (c), the control unit 21 determines, among the end points of the boundary line dividing the dissection flap region 42 and the vascular wall region 43, the end point that contacts the dissection cavity region 46 as the boundary point 66. Note that, if the dissection cavity region 46 has not been classified, the control unit 21 determines the boundary point 66 in the same manner as in the first embodiment.

[0123] According to this embodiment, a feature amount with high reproducibility can be calculated using a classification model 31 that has not been trained to separate the dissection flap region 42 and the vascular wall region 43.

[0124] Instead of using the classification model 31, the control unit 21 may extract the biological tissue region 40 from the transverse slice image 58 based on a speckle pattern or the like.

[0125] [Embodiment 5] 17 is an explanatory diagram for explaining the configuration of an information processing device 20 according to the fifth embodiment. This embodiment relates to a form in which the information processing device 20 according to this embodiment is realized by combining and operating a general-purpose computer 90 and a program 97. Explanations of parts common to the first embodiment will be omitted.

[0126] The computer 90 includes a reading unit 29 in addition to the control unit 21, main memory device 22, auxiliary memory device 23, communication unit 24, display unit 25, input unit 26 and bus described above.

[0127] The program 97 is recorded in a portable recording medium 96. The control unit 21 reads the program 97 via the reading unit 29 and stores it in the auxiliary storage device 23. The control unit 21 may also read out the program 97 stored in a semiconductor memory 98 such as a flash memory implemented in the computer 90. Furthermore, the control unit 21 may download the program 97 from another server computer (not shown) connected via the communication unit 24 and a network (not shown), and store it in the auxiliary storage device 23.

[0128] The program 97 is installed as a control program of the computer 90, and is loaded into and executed by the main storage device 22. In this manner, the information processing device 20 described in the first embodiment is realized. The program 97 in this embodiment is an example of a program product.

[0129] A computer program can be deployed to be executed on a single computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

[0130] The technical features (constitutive elements) described in each embodiment can be combined with each other, and by combining them, new technical features can be formed. The embodiments disclosed herein are illustrative in all respects and should not be considered as 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 scope and meaning equivalent to the claims.

[0131] Independent claims and dependent claims described in the claims may be combined with each other in any combination, regardless of the reference format. Furthermore, the claims may be written in a format in which a claim references two or more other claims (multiple claim format), but this is not limited to this format. The claims may also be written in a format in which a multiple claim references at least one other claim (multi-multi claim format). [Explanation of symbols]

[0132] 20 Information processing device 21 Control section 22 Main storage 23 Auxiliary storage device 24 Communications Department 25 Display section 26 Input section 29 Reading unit 31 Classification Models 36 Tomographic Image DB 40 Biological Tissue Area 41 Luminal area 42 Dissection flap region 43 Blood vessel wall area 44 Catheter Area 45 Extracavitary area 46 Dissection cavity area 47 Calcified area 48 Luminal area after correction 49 Corrected dissociation flap region 55 Corrected classification data (corrected cross-sectional shape) 56 Trace diagram 57 Classification data (cross-sectional shape) 58 Cross section image 61 1st straight line 62 Second straight line 641 First measurement line 642 Second measurement line 645 Angle Reference Line 646 Judgment line 65 Starting point of provisional dissociation 66 Boundary Points 67 Center of gravity 68 Approximate Circle 71 Position field 72 Feature column 90 Computer 96 Portable recording media 97 Program 98 Semiconductor Memory

Claims

1. obtaining a cross-sectional shape of the blood vessel, the cross-sectional shape including a shape of a lumen region of the blood vessel and a shape of a dissected flap region present within the lumen region; generating a corrected cross-sectional shape based on the shape of the lumen region and the shape of the dissection flap region; Calculating a center of gravity of a corrected lumen region in the corrected cross-sectional shape; A representative value for the blood vessel is calculated based on the center of gravity. A program that causes a computer to carry out processing.

2. The corrected cross-sectional shape is generated by deforming the dissection flap region to conform to the outer wall of the lumen region. The program according to claim 1.

3. A boundary point corresponding to a boundary with the blood vessel is obtained from the dissection flap region; Calculating a first straight line that passes through the boundary point and is tangent to an outer wall of the lumen region; The corrected cross-sectional shape is generated by dividing the dissection flap region by a plurality of straight lines parallel to the first straight line, and translating each of the divided regions in a first direction parallel to the first straight line until the divided regions come into contact with the inner surface of the blood vessel. The program according to claim 2.

4. the border point corresponds to a point at which dissection of the dissection flap region from the blood vessel ends; The first direction is a direction in which the dissection flap region is returned to a state before being dissected from the blood vessel. The program according to claim 3.

5. obtaining a cross-sectional image of the blood vessel; Displaying the cross-sectional image; The cross-sectional shape is obtained based on a user tracing the lumen region and the dissection flap region on the cross-sectional image. The program according to claim 1.

6. The cross-sectional shape is a schematic diagram created based on a cross-sectional image of the blood vessel. The program according to claim 1.

7. The area of ​​the corrected lumen region and the area of ​​the corrected dissection flap region in the corrected cross-sectional shape are equal to the area of ​​the lumen region and the area of ​​the dissection flap region in the cross-sectional shape, respectively. The program according to claim 1.

8. The representative value is an angle formed by lines connecting the center of gravity and both ends of the range in which the dissection flap region is dissected from the blood vessel. The program according to any one of claims 1 to 7.

9. The representative value is an angle formed by lines connecting the center of gravity and both ends of a lesion area occurring in the tissue of the blood vessel. The program according to any one of claims 1 to 7.

10. The program according to claim 8, further comprising: acquiring a transverse cross-sectional image of the blood vessel; and displaying the straight line and the angle superimposed on the transverse cross-sectional image.

11. The lesion area is a calcified area or a plaque area. The program according to claim 9.

12. obtaining a cross-sectional shape of the blood vessel, the cross-sectional shape including a shape of a lumen region of the blood vessel and a shape of a dissected flap region present within the lumen region; generating a corrected cross-sectional shape based on the shape of the lumen region and the shape of the dissection flap region; Calculating a center of gravity of a corrected lumen region in the corrected cross-sectional shape; A representative value for the blood vessel is calculated based on the center of gravity. An information processing method in which processing is performed by a computer.

13. An information processing device including a control unit, The control unit is obtaining a cross-sectional shape of the blood vessel, the cross-sectional shape including a shape of a lumen region of the blood vessel and a shape of a dissected flap region present within the lumen region; generating a corrected cross-sectional shape based on the shape of the lumen region and the shape of the dissection flap region; Calculating a center of gravity of a corrected lumen region in the corrected cross-sectional shape; A representative value for the blood vessel is calculated based on the center of gravity. Information processing device.

Images