Information processing method, program, and information processing device

Abstract

Provided is, for example, an information processing method capable of generating an appropriate boundary line even in a cross-sectional image in which a part of a boundary is unclear.　The information processing method causes a computer to execute processing for: acquiring label data (50) in which information relating to a plurality of labels including a first label is respectively associated with each portion of a cross-sectional image 59 of a blood vessel; determining a low-reliability angle range (57) in which a straight line passing through the scanning center of the cross-sectional image 59 passes through a region in which the reliability of the label data (50) is low; setting a plurality of interpolation points (517) in a region which is a boundary between a region to which the first label is assigned and a region to which the first label is not assigned, and which is outside the low-reliability angle range (57); and creating an interpolation line (518) connecting the interpolation points (517) to each other.

Description

Information Processing Method, Program, and Information Processing Apparatus

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

[0002] An image diagnostic apparatus has been proposed that generates a boundary line of a blood vessel wall or a blood flow region for a tomographic image of a blood vessel taken using an IVUS (Intravascular Ultrasound) catheter or an OCT (Optical Coherence Tomography) catheter (Patent Document 1).

[0003] International Publication No. 2014 / 041579

[0004] In Patent Document 1, in an IVUS image or an OCT image, a closed curve is generated by connecting positions where pixel values change steeply on a scanning line. Therefore, it is suitable for processing a cross-sectional image in which the boundary can be clearly identified over the entire circumference even with the naked eye.

[0005] However, for various reasons such as a branch of a blood vessel, calcification, plaque, devices such as a guide wire, and a stent previously placed, a tomographic image with an unclear boundary may be taken. It is difficult for the image diagnostic apparatus of Patent Document 1 to generate a correct boundary line for such a tomographic image.

[0006] On one aspect, an object is to provide an information processing method or the like that can generate an appropriate boundary line even for a tomographic image with an unclear boundary.

[0007] The present invention is an information processing method in which a computer executes a process of: acquiring label data in which information on a plurality of labels including a first label is associated with each part of a tomographic image of a blood vessel; determining a low reliability angle range through which a straight line passing through the scanning center of the tomographic image passes through a region with low reliability of the label data; setting a plurality of interpolation points in a region that is a boundary between a region to which the first label is assigned and a region to which the first label is not assigned and that is outside the low reliability angle range; and creating an interpolation line connecting the interpolation points.

[0008] Herein, an embodiment of the present invention is (2) the information processing method described in (1) above, wherein the label data further includes the probability that the assignment of the first label is correct for each portion, and the region to which the first label is assigned is preferably a region in which the probability of the assignment of the first label is correct is equal to or greater than a predetermined threshold.

[0009] (3) The information processing method described in (1) or (2) above preferably further includes, in the label data, information regarding whether or not a first label has been assigned to each part, and the reliability of the information.

[0010] (4) The information processing method described in any one of (1) to (3) above is further preferably such that the low-reliability angular range is an angular range in which a straight line passing through the scanning center of the transverse layer image does not pass through the boundary line between the region to which the first label is assigned and the region to which the first label is not assigned.

[0011] (5) The information processing method described in any one of (1) to (3) above is further preferably such that the low-reliability angular range is an angular range in which a straight line passing through the scanning center of the transverse layer image passes multiple times through the boundary line between the region to which the first label is assigned and the region to which the first label is not assigned.

[0012] (6) The information processing method described in any one of (1) to (3) above is further preferably such that the low-reliability angular range is an angular range in which a straight line passing through the scanning center of the transverse layer image passes through a region in which the difference between the boundary line between the region to which the first label is assigned and the region to which the first label is not assigned and the smoothed boundary line obtained by smoothing the boundary line is greater than a predetermined threshold.

[0013] (7) The information processing method described in any one of (1) to (3) above is further preferably such that the low reliability angle range is an angle range in which a straight line passing through the scanning center of the transverse layer image passes through the guide wire region or the side branch region.

[0014] (8) In the information processing method described in any one of (1) to (7) above, it is preferable that the interpolation points are more densely arranged in locations where the distance between the scanning center of the transverse layer image and the boundary is greater.

[0015] (9) The information processing method described in any one of (1) to (8) above is further preferably characterized in that the first label is a label indicating that the area is within the median boundary of the main pipe.

[0016] (10) The information processing method described in any one of (1) to (8) above is further preferably characterized in that the first label is a label indicating an extravascular region.

[0017] (11) The information processing method described in any one of (1) to (10) above is preferably further described in terms of creating interpolation lines when the first label is a label indicating that the region is within the medial boundary of the main vessel, and when the first label is a label indicating that the region is extravascular.

[0018] (12) The information processing method described in any one of (1) to (11) above is preferably further expressed by superimposing the interpolation line and the cross-sectional image.

[0019] (13) Preferably, the program is one in which a computer performs the following processes: acquires label data by associating information about multiple labels, including a first label, with each part of a transverse image of a blood vessel; determines a low-confidence angular range in which a straight line passing through the scanning center of the transverse image passes through a region with low confidence in the label data; sets multiple interpolation points in a region that is the boundary between a region to which the first label is assigned and a region to which the first label is not assigned, and is outside the low-confidence angular range; and creates an interpolation line connecting the interpolation points.

[0020] (14) Preferably, an information processing device having a control unit, wherein the control unit acquires label data by associating information about a plurality of labels, including a first label, with each portion of a transverse image of a blood vessel, determines a low-confidence angle range in which a straight line passing through the scanning center of the transverse image passes through a region with low confidence in the label data, sets a plurality of interpolation points in a region that is the boundary between a region to which the first label is assigned and a region to which the first label is not assigned, and is outside the low-confidence angle range, and creates an interpolation line connecting the interpolation points.

[0021] In one respect, this provides an information processing method that can generate appropriate boundary lines even in cross-sectional images where parts of the boundary are unclear.

[0022] This is an explanatory diagram illustrating the outline of the procedure for creating a boundary line. This is an explanatory diagram illustrating the configuration of the information processing system. This is an explanatory diagram illustrating an example of the procedure for determining a low-confidence angle range. This is an explanatory diagram illustrating an example of the procedure for determining a low-confidence angle range. This is an explanatory diagram illustrating an example of the procedure for determining a low-confidence angle range. This is an explanatory diagram illustrating an example of the procedure for determining a low-confidence angle range. This is an explanatory diagram illustrating an example of the procedure for setting the first interpolation point. This is a flowchart illustrating the processing flow of the program. This is an example screen. This is an explanatory diagram illustrating the configuration of the information processing system of Embodiment 2.

[0023] [Embodiment 1] Figure 1 is an explanatory diagram illustrating the outline of the procedure for creating a boundary line. A transverse image 59 of the blood vessel is acquired from the catheter system 30 (see Figure 2). The catheter system 30 in this embodiment performs radial scanning using a sensor 322 (see Figure 2) inserted inside the blood vessel. Details of the catheter system 30 will be described later.

[0024] The cross-sectional image 59 shown in the upper left of Figure 1 is in the so-called RT (polar coordinate) format, formed by arranging scan line data in parallel according to the scan angle. In this image, the upper end of the cross-sectional image 59 is the sensor 322. The horizontal direction of the cross-sectional image 59 corresponds to the scan angle θ, and the vertical direction of the cross-sectional image 59 corresponds to the distance R from the sensor 322.

[0025] Label data 50 is generated, which associates information about multiple labels with each part of the transverse image 59. Specifically, the label data 50 is data that associates the probability that each part constituting the transverse image 59 is a region corresponding to various labels, such as the lumen region of the main vessel, the lumen region of a side branch, the intramedial region, the guidewire region 551 (see Figure 7), the calcified region 553 (see Figure 7), the plaque region, or the extravascular region. Each part is, for example, one pixel. Each part may be of any size, for example, 2 pixels vertically and horizontally.

[0026] Here, "main vessel" refers to the blood vessel being treated, specifically the vessel into which the imaging catheter 32 (see Figure 2) is inserted. "Side branch" refers to a blood vessel branching off from the main vessel. In subsequent explanations, the luminal region of the main vessel may be simply referred to as the luminal region, and the luminal region of a side branch may be referred to as the side branch region 552 (see Figure 7).

[0027] The "medial boundary" includes the IEL (internal elastic lamina), located at the boundary between the intima and media of a blood vessel, and the EEM (external elastic membrane) and EEL (external elastic lamina), located at the boundary between the media and adventitia. The "region within the medial boundary" is the region sandwiched between the IEL and the EEM or EEL, and refers to the region corresponding to the media itself.

[0028] The "extravascular region" refers to the area outside the tunica media of a blood vessel. The extravascular region includes various organs adjacent to the blood vessel being observed using the imaging catheter 32, as well as the area outside the skin. Other blood vessels adjacent to the blood vessel being observed are also classified as part of the extravascular region.

[0029] The label data 50 is output by a multi-label segmentation model that accepts, for example, a transverse image 59 as input. The label data 50 may also be generated by a segmentation model that accepts a transverse image 59 as input and outputs labels for specific regions such as the lumen region and the confidence level of those labels for each portion. The label data 50 may also be generated by a combination of multiple segmentation models. Since multi-label segmentation models and segmentation models are well known, a detailed explanation is omitted.

[0030] Based on the label data 50, the probability that each part is, for example, a lumen region is extracted. By dividing the region into areas where the probability of being a lumen region is above a predetermined threshold, such as 50 percent, and other areas, the boundary line between the lumen region and the non-lumen region in the transverse image 59, i.e., the internal elastic lamina, can be defined. Data regarding the defined boundary line may be added to the label data 50.

[0031] In Figure 1, as a generalization, the first label region 511, where the probability of being the first label is above a threshold in the transverse layer image 59, is shown with a downward-sloping hatch, while the other regions, the non-first label region 512, are shown in white. The boundary between the first label region 511 and the non-first label region 512 is labeled as the first label boundary 515. The aforementioned luminal region is an example of a first label. The first label boundary 515 is an example of the boundary line between a region that has been assigned a first label and a region that has not.

[0032] The first label boundary 515 may be created by applying known image processing such as smoothing and edge extraction to the transverse layer image 59 instead of using the label data 50. The first label boundary 515 may also be generated by a learning model trained to accept the input of the transverse layer image 59 and output the first label boundary 515.

[0033] As mentioned above, data relating to the boundary line may be added to the label data 50. Therefore, if the first label boundary 515 is generated without generating the label data 50, the label data 50 may consist of data relating to the first label boundary 515.

[0034] A low-confidence angular range 57 is determined when a straight line passing through the scanning center of the transverse layer image 59 passes through a region with low confidence in the label data 50 or the first label boundary 515. In RT format (polar coordinate format) images, the straight line passing through the scanning center of the transverse layer image 59 is a straight line parallel to the axis in the distance R direction, and the low-confidence angular range 57 is represented by a vertically elongated rectangle. A specific example of the method for determining the low-confidence angular range 57 will be described later.

[0035] The reliability of the first label boundary 515 is low in the low-reliability angular range 57, but it can be expected to be sufficiently high in areas other than the low-reliability angular range 57. Multiple first interpolation points 517 are set in the area of ​​the first label boundary 515 other than the low-reliability angular range 57. A specific example of the method for determining the positions of the first interpolation points 517 will be described later.

[0036] The coordinate transformation converts the position of the first interpolation point 517 from RT format to XY format (orthogonal coordinate format). The method for converting between RT format and XY format images is well known, so its explanation is omitted. In Figure 1, the figure after conversion to XY format also includes the low-confidence angle range 57 for reference, in addition to the first interpolation point 517. However, the low-confidence angle range 57 does not need to be converted to XY format.

[0037] For example, a first interpolation line 518 is created that connects the first interpolation points 517 using any interpolation method such as spline interpolation. The first interpolation line 518 is superimposed on the cross-sectional image 59 converted to XY format. The first interpolation line 518 is an example of a boundary line whose accuracy has been improved by modifying the first label boundary 515.

[0038] If the first label is the lumen region, the first interpolation line 518 is a closed curve representing the outer circumference of the lumen region. If the first label is the region encompassing both the lumen region and the medial boundary region, the first interpolation line 518 is a closed curve representing the external elastic plate. Because the influence of the low-confidence angle range 57 can be avoided, the shape of the outer circumference of the lumen region or the external elastic plate, for example, can be drawn with good reproducibility. By using the first interpolation line 518, the area, maximum diameter, minimum diameter, etc. of the outer circumference of the lumen region or the external elastic plate, for example, can be quantified with good reproducibility.

[0039] By continuing to process in RT format until the stage of setting the first interpolation point 517, it is possible to prevent errors, noise, and computational load caused by conversion processing between RT format and XY format.

[0040] If these disadvantages are acceptable, for example, the creation of the first label boundary 515, the determination of the low-confidence angle range 57, or the setting of the first interpolation point 517 may be performed in an XY format, and the intermediate results may be displayed on the display unit 25 (see Figure 2), etc. The user can confirm that the correct processing is proceeding by viewing the intermediate results in an XY format that matches the actual structure of blood vessels.

[0041] Figure 2 is an explanatory diagram illustrating the configuration of the information processing system 10. The information processing system 10 includes an information processing device 20 and a catheter system 30. The catheter system 30 includes a catheter control device 31, an image acquisition catheter 32, and an MDU (Motor Driving Unit) 33. The image acquisition catheter 32 is connected to the catheter control device 31 via the MDU 33.

[0042] The image acquisition catheter 32 is either an OCT catheter or an IVUS catheter. The image acquisition catheter 32 may also be a dual-sensor type catheter that has the functions of both an OCT catheter and an IVUS catheter.

[0043] The imaging catheter 32 includes a sensor 322, a shaft 323, and a sheath 324. The sheath 324 is a resin tube with a tip portion bent in a substantially crank shape. A guide wire insertion hole (not shown) is provided on the tip side of the bent portion.

[0044] The shaft 323 is inserted into the sheath 324. The sensor 322 is fixed to the tip of the shaft 323. When the imaging catheter 32 is an OCT catheter, the sensor 322, which is an OCT sensor, is a ball lens that emits light from a light source and receives reflected light. When the imaging catheter 32 is an IVUS catheter, the sensor 322, which is an IVUS sensor, is an ultrasonic sensor that transmits and receives ultrasonic waves.

[0045] In the following description, the side farther from the MDU 33 of the imaging catheter 32 is referred to as the tip side. A catheter image such as a cross-sectional image 59 of a blood vessel (see FIG. 3) is output from the catheter control device 31.

[0046] The MDU 33 rotates the sensor 322 and the shaft 323 inside the sheath 324. By rotating while the sensor 322 transmits and receives light or ultrasonic waves, radial scanning is performed. The catheter control device 31 generates one image for each rotation of the sensor 322. The generated image is a cross-sectional image 59 centered on the sensor 322 and substantially perpendicular to the sheath 324. In the following description, a series of processes from the transmission and reception of data by the sensor 322 to the generation of one cross-sectional image 59 may be referred to as the imaging of the cross-sectional image 59.

[0047] The MDU 33 can also move forward and backward while rotating the sensor 322 and the shaft 323 inside the sheath 324. By a pullback operation of rotating the sensor 322 while pulling it toward the MDU 33 at a constant speed, the catheter control device 31 realizes a three-dimensional scan of continuously imaging a plurality of cross-sectional images 59 substantially perpendicular to the sheath 324 at a predetermined interval.

[0048] The information processing apparatus 20 includes a control unit 21, a main memory device 22, an auxiliary storage device 23, a communication unit 24, a display unit 25, an input unit 26, and a bus. The control unit 21 is an arithmetic control device that executes the program of the present embodiment. One or more CPUs (Central Processing Units), GPUs (Graphics Processing Units), or multi-core CPUs, etc. are used for the control unit 21. The control unit 21 is connected to each hardware component constituting the information processing apparatus 20 via a bus.

[0049] The main memory device 22 is a storage device such as SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), or flash memory. Information necessary during the processing performed by the control unit 21 and the program being executed by the control unit 21 are temporarily stored in the main memory device 22.

[0050] The auxiliary storage device 23 is a storage device such as SRAM, flash memory, hard disk, or magnetic tape. Programs to be executed by the control unit 21 and various data necessary for the execution of the programs are stored in the auxiliary storage device 23. The communication unit 24 is an interface that performs communication between the information processing apparatus 20 and a network.

[0051] The display unit 25 is, for example, a liquid crystal display device or an organic EL (Electro Luminescence) display device. The input unit 26 is an input device such as a keyboard, mouse, trackball, or microphone. The display unit 25 and the input unit 26 may be integrally laminated to form a touch panel.

[0052] The display unit 25 may be a connection interface that connects the information processing apparatus 20 and an external display device. A communication unit 24 that connects data to an external display device via a network may realize the functions of the display unit 25 and the input unit 26.

[0053] The information processing device 20 in this embodiment is an information device such as a general-purpose personal computer, tablet, smartphone, or server computer. The information processing device 20 may also be a mainframe computer, a virtual machine operating on a mainframe computer, a cloud computing system, a quantum computer, or multiple personal computers performing distributed processing. The information processing device 20 may be configured integrally with, for example, a catheter control device 31 or a hospital information system (HIS) not shown in the figures.

[0054] The catheter control device 31 can perform multi-label segmentation on the captured transverse image 59 and create label data 50 that associates label information with each part of the transverse image 59. The catheter control device 31 may also be able to create the first label boundary 515 as described using Figure 1. The catheter control device 31 transmits the RT-format transverse image 59, the XY-format transverse image 59, and the label data 50 to the information processing device 20.

[0055] Based on the label data 50, the control unit 21 divides the image into, for example, a first label region 511 where the probability of being the first label is above a threshold, a non-first label region 512 which is the other region, and the cross-sectional image 59, and creates a boundary line such as the first label boundary 515 which is the boundary between the two. Based on the boundary line such as the first label boundary 515, the control unit 21 creates interpolation lines such as the first interpolation line 518.

[0056] The catheter control device 31 may, after performing multi-label segmentation or the like on the acquired transverse image 59, binarize the probability for each label, for example, by determining whether the probability of it being the first label is above or below a threshold. In this case, the label data 50 created by the catheter control device 31 is data that associates information indicating whether each part of the transverse image 59 is, for example, the first label region 511 or the non-first label region 512, that is, information about each label that has been binarized.

[0057] If the label data 50 is data that associates binarized information about each label with each part of the transverse layer image 59, the catheter control device 31 may add information about boundary lines such as the first label boundary 515 to the label data 50 and transmit it to the information processing device 20.

[0058] When the catheter control device 31 creates a boundary line such as the first label boundary 515, the label data 50 may not include data associating label information with each part of the transverse layer image 59, but may only include information about the boundary line such as the first label boundary 515. The control unit 21 creates interpolation lines such as the first interpolation line 518 based on the information about the boundary line such as the first label boundary 515.

[0059] The catheter control device 31 does not need to generate label data 50. In this case, the catheter control device 31 transmits RT-format transverse image 59 and XY-format transverse image 59 to the information processing device 20. The control unit 21 creates boundary lines such as the first label boundary 515 and the first interpolation line 518 based on the transverse image 59. Alternatively, the catheter control device 31 may create only the XY-format transverse image 59, and the control unit 21 may create the XY-format transverse image 59.

[0060] The catheter control device 31 may transmit the signals related to the scan lines received from the sensor 322 to the information processing device 20. Based on the signals related to the scan lines, the catheter control device 31 can create a transverse layer image 59, boundary lines such as the first label boundary 515, and the first interpolation line 518, etc.

[0061] If it is not necessary to create the first interpolation line 518 in real time during catheterization, the information processing device 20 and the catheter system 30 do not need to be connected. The control unit 21 reads the transverse layer image 59 recorded in the HIS or the like and creates label data 50, boundary lines such as the first label boundary 515, and interpolation lines such as the first interpolation line 518.

[0062] In the following explanation, we will mainly use the case where the control unit 21 performs software-based processing as an example. The processes and models explained using flowcharts may each be implemented by dedicated hardware.

[0063] Figure 3 is an explanatory diagram illustrating an example of the procedure for determining the low-reliability angular range 57. Using Figure 3, the process for when the first label boundary 515 is interrupted at the edge on the farther side of the distance R from the sensor 322 will be explained. This phenomenon occurs, for example, when the display range of the transverse layer image 59 is small and a part of the lumen region or the outer elastic plate is outside the display range of the transverse layer image 59. The control unit 21 determines that the angular range in which the first label boundary 515 is not depicted is the low-reliability angular range 57.

[0064] Figure 4 is an explanatory diagram illustrating an example of a procedure for determining the low-confidence angle range 57. Using Figure 4, we will explain the process when the first label boundary 515 is not fixed in one place for a given scanning angle θ, or when the line indicating the first label boundary 515 is interrupted midway. Such phenomena occur, for example, when artifacts such as multiple echoes or image noise are present.

[0065] The control unit 21 determines that an angular range in which multiple first label boundaries 515 are not fixed at a single location for a given scanning angle θ is a low-reliability angular range 57. In other words, the control unit 21 determines that an angular range in which the distance R corresponding to the first label boundary 515 is not uniquely determined for a given scanning angle θ is a low-reliability angular range 57.

[0066] Figure 5 is an explanatory diagram illustrating an example of a procedure for determining the low confidence angle range 57. The control unit 21 smooths the first label boundary 515 to create a smoothed boundary 516. The smoothed boundary 516 can be calculated, for example, by moving average the first label boundary 515 using a window function that moves in the scanning angle θ direction. The smoothed boundary 516 is an example of a smoothed boundary line obtained by smoothing the first label boundary 515.

[0067] The control unit 21 determines that an angular range in which the difference between the smoothing boundary 516 and the first label boundary 515 is greater than a predetermined threshold, that is, an angular range in which the smoothing boundary 516 and the first label boundary 515 are far apart, is a low-reliability angular range 57.

[0068] The control unit 21 may also determine that the angle range in which the difference between the smoothing boundary 516 and the first label boundary 515 is less than or equal to a predetermined threshold, that is, the angle range in which the smoothing boundary 516 and the first label boundary 515 are not far apart, is the high-reliability angle range 58.

[0069] When defining a high-reliability angle range 58, the control unit 21 may determine that any region other than the high-reliability angle range 58 is a low-reliability angle range 57. The control unit 21 may also preferentially place the first interpolation point 517 in the high-reliability angle range 58 compared to any region that is neither a low-reliability angle range 57 nor a high-reliability angle range 58.

[0070] Figure 6 is an explanatory diagram illustrating an example of a procedure for determining a low-confidence angle range 57. In Figure 6, information about the first label is extracted from the label data 50 and shown in grayscale. White represents the portion with a high probability of being the first label, and black represents the portion with a low probability of being the first label. The portion indicated by symbol A is a state where black and white portions are mixed together, with a large amount of gray, which is an intermediate color. Therefore, it is clear that a highly reliable first label boundary 515 cannot be created.

[0071] The control unit 21 determines that a scanning angle with a large number of sigmoid values ​​near the threshold for determining whether it is the first label region 511 or a non-first label region 512 is a low-confidence angle range 57. The control unit 21 may also determine that a scanning angle θ range in which the probability of being in the first label region 511 increases or decreases multiple times is a low-confidence angle range 57.

[0072] The control unit 21 may, after binarizing the grayscale image shown in Figure 6 into a first label region 511 and a non-first label region 512 to extract the first label boundary 515, determine the low-confidence angle range 57 using the procedure described with reference to Figure 5.

[0073] The control unit 21 may, after applying a smoothing process to the grayscale image shown in Figure 6, determine that the angular range including the region where the difference from the original grayscale image exceeds a threshold is the low-reliability angular range 57.

[0074] Figure 7 is an explanatory diagram illustrating an example of a procedure for determining the low confidence angle range 57. In Figure 7, the first label boundary 515 and the second label boundary 525, created based on the label data 50, are shown along with the guidewire region 551, the lateral branch region 552, and the calcification region 553 extracted from the label data 50, as well as the first label boundary 515 and the second label boundary 525.

[0075] In Figure 7, the guidewire region 551 indicates an area where the probability that a pixel is a guidewire or a shaded area formed by a guidewire exceeds a threshold, such as 50 percent. Similarly, the side branch region 552 indicates an area where the probability that a pixel is a side branch region exceeds a threshold, such as 50 percent. The calcification region 553 indicates an area where the probability that a pixel is a calcified portion or a shaded area formed by a calcified portion exceeds a threshold, such as 50 percent.

[0076] As described above, the first label boundary 515 is the boundary between the first label region 511, where the probability of being the first label is greater than or equal to a threshold, and the non-first label region 512. Similarly, the second label boundary 525 is the boundary between the second label region 521, where the probability of being the second label is greater than or equal to a threshold, and the non-second label region 522.

[0077] In Figure 7, the first label represents the luminal region, and the second label represents the sum of the luminal region and the region within the intimal boundary. Therefore, the first label boundary 515 indicates the internal elastic lamina, and the second label boundary 525 indicates the external elastic lamina.

[0078] The guidewire region 551 extends medially to the first label boundary 515, i.e., from the lumen region toward the outside of the vessel. The collateral branch region 552 extends on the first label boundary 515, i.e., from the edge of the lumen region toward the outside of the vessel. The calcified region 553 extends between the first label boundary 515 and the second label boundary 525, i.e., from the region within the medial boundary of the vessel toward the outside of the vessel.

[0079] The guide wire region 551 and the side branch region 552 obscure both the first label boundary 515 and the second label boundary 525. Therefore, the control unit 21 determines that the angular range including the guide wire region 551 and the angular range including the side branch region 552 are low-confidence angular ranges 57 for both the first label boundary 515 and the second label boundary 525.

[0080] The presence of the calcified region 553 does not affect the first label boundary 515. Therefore, the control unit 21 determines that the angular range including the first label boundary 515 is a low-confidence angular range 57 only with respect to the second label boundary 525.

[0081] The procedure for determining the low-reliability angle range 57 is not limited to that described using Figures 3 to 7. For example, the low-reliability angle range 57 can also be determined for the plaque region and the stent region using appropriate procedures. The threshold used when determining the low-reliability angle range 57 should preferably be determined appropriately according to the specifications of the image acquisition catheter 32, etc.

[0082] Figure 8 is an explanatory diagram illustrating an example of the procedure for setting the first interpolation point 517. Figure 8 shows an enlarged view of the relationship between the first label region 511 and the non-first label region 512, expressed in RT format, and the first label boundary 515. The top of Figure 8 indicates the position of the sensor 322. The horizontal direction in Figure 8 corresponds to the scanning angle θ, and the vertical direction of the transverse layer image 59 corresponds to the distance R from the sensor 322.

[0083] In Figure 8, the eight scan lines from the nth scan line to the (n+7th)th scan line are schematically shown as vertically elongated rectangles. The scan angle of the nth scan line is θn, and the scan angle of the (n+1)th scan line is θn+1. For the nth scan line, the region from sensor 322 to distance Rn is determined to be the first label region 511, and the region beyond distance Rn is determined to be the non-first label region 512. For the (n+1)th scan line, the region from sensor 322 to distance Rn+1 is determined to be the first label region 511, and the region beyond distance Rn+1 is determined to be the non-first label region 512.

[0084] The first interpolation points 517 may be placed at regular scanning angle intervals θ. However, if they are placed at regular scanning angle intervals θ, the greater the distance R to the first label boundary 515, the wider the spacing between the first interpolation points 517 along the first label boundary 515 becomes. As a result, the accuracy of the first interpolation line 518 decreases in areas with a large distance R compared to areas with a small distance R.

[0085] Using Figure 8, a method for arranging the first interpolation line 518 to create a first interpolation line 518 with constant accuracy regardless of distance R will be explained.

[0086] In Figure 8, the vertical dashed line schematically indicates the scanning angle θ at which the first interpolation point 517 is positioned. The control unit 21 positions the first interpolation point 517 such that the area of ​​the first label region 511, enclosed by the two dashed lines, is approximately equal. In Figure 8, the dashed lines are positioned at the boundaries of the scan lines, but the control unit 21 may also position the dashed lines to divide the scan lines into two.

[0087] As a result, in the region where the first label boundary 515 is close to the sensor 322, the scanning angle between the first interpolation points 517 is relatively wide, and in the region where the first label boundary 515 is far from the sensor 322, the scanning angle between the first interpolation points 517 is relatively narrow. As a result, the spacing between the first interpolation points 517 on the first label boundary 515 becomes approximately equal. As a result, the control unit 21 can create a highly accurate first interpolation line 518.

[0088] The control unit 21 may also arrange the first interpolation points 517 such that they divide the length of the first label boundary 515 into approximately equal parts. In the XY format, the control unit 21 may also arrange the first interpolation points 517 such that the straight lines connecting the center of the radial scan and two adjacent first interpolation points 517 divide the area inside the first label boundary 515 into approximately equal parts.

[0089] The control unit 21 may arrange the first interpolation points 517 in an XY format such that the straight lines connecting the centroid of the first label region 511 and two adjacent first interpolation points 517 divide the area inside the first label boundary 515 into approximately equal parts.

[0090] Figure 9 is a flowchart illustrating the program's processing flow. The control unit 21 acquires catheter images from the catheter system 30 (step S501). The catheter images include RT-format transverse images 59 and XY-format transverse images 59. If the image acquisition catheter 32 is for three-dimensional scanning, the catheter images may also include longitudinal images and three-dimensional models, etc.

[0091] The control unit 21 acquires label data 50 from the catheter system 30 (step S502). The label data 50 is data that associates label information with each part of the transverse layer image 59.

[0092] The control unit 21 acquires information about boundary lines, such as the first label boundary 515 (step S503). For example, if the label data 50 acquired in step S502 already contains information about boundary lines, such as the first label boundary 515, the control unit 21 extracts the boundary line information from the label data 50.

[0093] If the label data 50 acquired in step S502 does not contain information about boundary lines such as the first label boundary 515, the control unit 21 creates boundary lines such as the first label boundary 515 based on information about each part of the cross-sectional image 59.

[0094] The control unit 21 determines the low-reliability angle range 57 based on the methods described using Figures 3 to 7 (step S504). The control unit 21 sequentially determines the low-reliability angle range 57, for example, in a predetermined order. The control unit 21 may also accept instructions from the user regarding which method to use.

[0095] The control unit 21 sets multiple interpolation points in the area of ​​the boundary line other than the low-reliability angle range 57 (step S505). The control unit 21 preferably places the interpolation points relatively coarsely in the area where the boundary line is close to the sensor 322, and relatively densely in the area where the boundary line is far from the sensor 322, as explained using, for example, Figure 8.

[0096] The control unit 21 converts the arrangement of interpolation points set in step S505 into XY format (step S506). The control unit 21 creates interpolation lines that smoothly connect the interpolation points (step S507). The control unit 21 superimposes the interpolation lines onto the XY format cross-sectional image 59 and displays it on the display unit 25 (step S508). The control unit 21 may further superimpose interpolation points and display them on the display unit 25. The control unit 21 then terminates the process.

[0097] Figure 10 shows an example screen. The cross-sectional image 59 displays two interpolation lines, the first interpolation line 518 and the second interpolation line 528, and the first interpolation point 517 and the second interpolation point 527 used to create each interpolation line, superimposed on the image. The guide wire region 551, the lateral branch region 552, and the calcification region 553 are indicated in the cross-sectional image 59 by hatching.

[0098] The user can visually check whether the guidewire region 551, the lateral branch region 552, and the calcification region 553 are properly determined, and whether the first interpolation point 517 and the second interpolation point 527 are properly displayed, in order to determine whether reliable first interpolation lines 518 and second interpolation lines 528 have been created.

[0099] The control unit 21 may switch the display of the second interpolation point 527 and the second interpolation line 528, etc., based on instructions received from the user. The control unit 21 may also display the area, average diameter, minimum diameter, maximum diameter, etc., of the region enclosed by the first interpolation line 518 and the second interpolation line 528, respectively. If the proportion occupied by the low-reliability angle range 57 is greater than a predetermined threshold, the control unit 21 may display a warning about the low reliability of this data, or a message indicating that this data cannot be calculated properly.

[0100] According to this embodiment, even if the cross-sectional image 59 has an unclear boundary in part, it is possible to provide an information processing method that generates and displays an appropriate boundary line. Because the influence of the unclear part of the boundary can be avoided, it is possible to provide an information processing method that can calculate parameters such as area, average diameter, minimum diameter, and maximum diameter with good reproducibility.

[0101] According to this embodiment, processing is consistently performed in RT format until just before the interpolation lines are created, thus providing an information processing method with less error and noise due to conversion processing between RT format and XY format. Furthermore, it provides an information processing method that prevents an increase in computational load due to repeated conversion processing between RT format and XY format.

[0102] [Embodiment 2] This embodiment relates to a configuration in which an information processing system 10 is realized by operating a general-purpose computer 90 and a program 97 in combination. The parts that are common with Embodiment 1 will not be described.

[0103] Figure 11 is an explanatory diagram illustrating the configuration of the information processing system 10 of the second embodiment. The computer 90 includes the control unit 21, main memory 22, auxiliary memory 23, communication unit 24, display unit 25, input unit 26, and bus, as described above, as well as a read unit 29.

[0104] The program 97 is recorded on a portable recording medium 96. The control unit 21 reads the program 97 via the reading unit 29 and saves it to the auxiliary storage device 23. The control unit 21 may also read the program 97 stored in a semiconductor memory 98, such as a flash memory, installed 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 save it to the auxiliary storage device 23.

[0105] Program 97 is installed as a control program for the computer 90, loaded into the main memory 22, and executed. Thus, the information processing system 10 described in Embodiment 1 is realized. Program 97 in this embodiment is an example of a program product.

[0106] A program is an example of a program product. A program may be provided on a recording medium or distributed from an external computer. A computer program can be deployed on a single computer or at a single site, or distributed across multiple sites and interconnected by a communication network to run on multiple computers.

[0107] The technical features (constituent elements) described in each embodiment are combinable with each other, and by combining them, new technical features can be formed. 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 meaning and scope equivalent to the claims are intended to be included.

[0108] The independent and dependent claims described in the claims can be combined with each other in any combination, regardless of the form of reference. Furthermore, while the claims use a multi-claim format in which claims refer to two or more other claims (multi-claim format), this is not the only option. A multi-claim format in which at least one multi-claim is referenced (multi-multi-claim format) is also acceptable.

[0109] 10 Information Processing System 20 Information Processing Device 21 Control Unit 22 Main Memory Unit 23 Auxiliary Memory Unit 24 Communication Unit 25 Display Unit 26 Input Unit 29 Reading Unit 30 Catheter System 31 Catheter Control Device 32 Image Acquisition Catheter 322 Sensor 323 Shaft 324 Sheath 33 MDU 50 Label Data 511 First Label Area 512 Non-First Label Area 515 First Label Boundary 516 Smoothing Boundary 517 First Interpolation Point (Interpolation Point) 518 First Interpolation Line (Interpolation Line) 521 Second Label Area 522 Non-Second Label Area 525 Second Label Boundary 527 Second Interpolation Point 528 Second Interpolation Line 551 Guidewire Area 552 Side Branch Area 553 Calcification Area 57 Low-Confidence Angle Range 58 High-reliability angle range 59 Cross-sectional image 90 Computer 96 Portable recording medium 97 Program 98 Semiconductor memory

Claims

1. An information processing method in which a computer performs the following steps: acquire label data by associating information about multiple labels, including a first label, with each part of a transverse image of a blood vessel; determine a low-confidence angular range in which a straight line passing through the scanning center of the transverse image passes through a region with low confidence in the label data; set multiple interpolation points in a region that is the boundary between a region to which the first label is assigned and a region to which the first label is not assigned, and is outside the low-confidence angular range; and create an interpolation line connecting the interpolation points.

2. The information processing method according to claim 1, wherein the label data includes the probability that the assignment of the first label is correct for each portion, and the region to which the first label is assigned is a region in which the probability of the assignment of the first label is correct is equal to or greater than a predetermined threshold.

3. The information processing method according to claim 1, wherein the label data includes information on whether or not a first label has been assigned to each of the parts, and the reliability of the information.

4. The information processing method according to claim 1, wherein the low-confidence angular range is an angular range in which a straight line passing through the scanning center of the transverse layer image does not pass through the boundary line between the region to which the first label is assigned and the region to which the first label is not assigned.

5. The information processing method according to claim 1, wherein the low-confidence angular range is an angular range in which a straight line passing through the scanning center of the transverse layer image passes multiple times through the boundary line between the region to which the first label is assigned and the region to which the first label is not assigned.

6. The information processing method according to claim 1, wherein the low-confidence angle range is an angle range in which a straight line passing through the scanning center of the transverse layer image passes through a region in which the difference between the boundary line between the region to which the first label is assigned and the region to which the first label is not assigned and the smoothed boundary line obtained by smoothing the boundary line is greater than a predetermined threshold.

7. The information processing method according to claim 1, wherein the low-reliability angular range is the angular range in which a straight line passing through the scanning center of the transverse layer image passes through the guide wire region or the side branch region.

8. The information processing method according to claim 1, wherein the interpolation points are arranged more densely in locations where the distance between the scanning center of the transverse layer image and the boundary is greater.

9. The information processing method according to claim 1, wherein the first label is a label indicating that it is a region within the median boundary of the main pipe.

10. The information processing method according to claim 1, wherein the first label is a label indicating an extravascular region.

11. The information processing method according to claim 1, which creates an interpolation line when the first label is a label indicating an intramedial region of the main vessel, and an interpolation line when the first label is a label indicating an extravascular region.

12. The information processing method according to any one of claims 1 to 11, which superimposes the interpolation line and the transverse layer image.

13. A program that causes a computer to perform the following processes: acquire label data by associating information about multiple labels, including a first label, with each part of a transverse image of a blood vessel; determine a low-confidence angular range in which a straight line passing through the scanning center of the transverse image passes through a region with low confidence in the label data; set multiple interpolation points in a region that is the boundary between a region to which the first label is assigned and a region to which the first label is not assigned, and is outside the low-confidence angular range; and create an interpolation line connecting the interpolation points.

14. An information processing device having a control unit, wherein the control unit acquires label data by associating information about a plurality of labels, including a first label, with each portion of a transverse image of a blood vessel; determines a low-confidence angular range in which a straight line passing through the scanning center of the transverse image passes through a region with low confidence in the label data; sets a plurality of interpolation points in a region that is the boundary between a region to which the first label is assigned and a region to which the first label is not assigned, and is outside the low-confidence angular range; and creates an interpolation line connecting the interpolation points.

Images