Aortic aneurysm imaging diagnostic support device, aortic aneurysm imaging diagnostic support method, and program

Abstract

To provide an aortic aneurysm image diagnosis support apparatus capable of detecting a candidate of an aortic aneurysm with a determination reference corresponding to a section.SOLUTION: The aortic aneurysm image diagnosis support device comprises: an aorta detection unit which detects an aorta classified into a plurality of sections, from a computer tomographic image of a living body; and an aortic aneurysm candidate detection unit which detects an aortic aneurysm candidate from the detected aorta, by using a condition about a determination size of an aortic aneurysm corresponding to each of the sections.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an aortic aneurysm image diagnosis support apparatus, an aortic aneurysm image diagnosis support method, and a program. [Background Art]

[0002] When determining an aortic aneurysm from CT (Computer Tomography) image data using image processing technology, there is a method in which aortic segmentation is performed on horizontal cross-sections (axial cross-sections) of CT images, and the peak position of the minor axis of the aorta recognized in each cross-section is detected as a candidate for an aortic aneurysm. [Prior Art Documents] [Patent Documents]

[0003] [Patent Document 1] International Publication No. 2017-047819 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] However, in the method for diagnosing an aortic aneurysm from CT image data, since determination criteria (for example, size) differ depending on the site of the aorta (chest, abdomen, iliac, etc.), there is a problem in that it may take time to perform a diagnosis corresponding to the site for each detected aortic aneurysm candidate.

[0005] The present invention has been made in view of such circumstances, and provides an aortic aneurysm image diagnosis support apparatus, an aortic aneurysm image diagnosis support method, and a program capable of detecting aortic aneurysm candidates based on determination criteria corresponding to the site. [Means for Solving the Problem]

[0006] This invention was made to solve the above-mentioned problems, and one aspect of the present invention is an aortic aneurysm image diagnostic support device comprising: an aortic detection unit that detects the aorta classified into multiple parts from an image obtained by computed tomography of a living organism; and an aortic aneurysm candidate detection unit that detects aortic aneurysm candidates from the detected aorta using conditions relating to the determination size of aortic aneurysms corresponding to each of the multiple parts.

[0007] Another aspect of the present invention is the above-described aortic aneurysm image diagnostic support device, comprising a thinning line extraction unit for thinning the detected aorta, wherein the aortic aneurysm candidate detection unit comprises a vertical cross-section extraction unit for extracting a cross-section of the aorta perpendicular to the line obtained by thinning by the thinning line extraction unit, and a condition determination unit for detecting aortic aneurysm candidates by applying the above conditions to the cross-section of the aorta.

[0008] Another aspect of the present invention is an aortic aneurysm image diagnostic support device as described above, comprising a whisker removal unit that performs whisker removal processing on the lines obtained by the thinning by the thinning unit, wherein the determination conditions for whether or not to remove the whiskers in the whisker removal unit depend on the plurality of sites.

[0009] Another aspect of the present invention is the above-described aortic aneurysm image diagnostic support device, comprising a detection result display unit that displays side by side a graph plotting a value representing the thickness of the aorta in the axial direction along the line obtained by thinning by the thinning unit, an image of the aorta projected onto the coronal surface, and an image of a cross-section perpendicular to the centerline of the aorta.

[0010] Another aspect of the present invention is an aortic aneurysm image diagnostic support method comprising the steps of detecting the aorta, which is classified into multiple parts, from computed tomography images of an organism, and detecting candidate aortic aneurysms from the detected aorta using conditions relating to the determination size of aortic aneurysms corresponding to each of the multiple parts.

[0011] Another aspect of the present invention is a program that causes a computer to function as an aortic detection unit that detects the aorta classified into multiple parts from images obtained by computed tomography of a living organism, and an aortic aneurysm candidate detection unit that detects aortic aneurysm candidates from the detected aorta using conditions relating to the determination size of aortic aneurysms corresponding to each of the multiple parts. [Effects of the Invention]

[0012] According to this invention, it is possible to detect potential aortic aneurysms based on criteria specific to the location. [Brief explanation of the drawing]

[0013] [Figure 1] This is a schematic block diagram showing the configuration of an aortic aneurysm image diagnostic support device 100 according to one embodiment of the present invention. [Figure 2] This is a schematic block diagram showing an example of the configuration of the aortic detection unit 101 in the same embodiment. [Figure 3] This figure illustrates an example of aortic data in the same embodiment. [Figure 4] This figure illustrates an example of wire thinning by the wire thinning unit 113 in the same embodiment. [Figure 5] This figure illustrates the processing of the whiskers removal section 114 in the same embodiment. [Figure 6] This table shows examples of the determination conditions in the whisker removal section 114 in the same embodiment. [Figure 7] This is a schematic block diagram showing an example of the configuration of the aortic aneurysm candidate detection unit 102 in the same embodiment. [Figure 8] This is a schematic diagram illustrating the operation of the vertical cross-sectional extraction unit 121 in the same embodiment. [Figure 9] This figure shows an example of an axial cross-sectional CT image. [Figure 10] This figure shows an example of a CT image of a cross-section perpendicular to the centerline of the aorta in the same embodiment. [Figure 11] This figure shows an example of a CT image illustrating the operation of the condition determination unit 122 in the same embodiment. [Figure 12] It is a graph explaining the operation of the condition determination unit 122 in the same embodiment. [Figure 13] It is a table showing the condition (threshold) regarding the determination size of an aortic aneurysm used by the condition determination unit 122 in the same embodiment. [Figure 14] It is a diagram showing display examples G1 and G2 by the detection result display unit 103 in the same embodiment. [Figure 15] It is a diagram showing a display example G3 by the detection result display unit 103 in the same embodiment. [Figure 16] It is a diagram showing a display example G4 by the detection result display unit 103 in the same embodiment. [Figure 17] It is a schematic block diagram showing the configuration of the trained model generation device 200 in the same embodiment. MODE FOR CARRYING OUT THE INVENTION

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic block diagram showing the configuration of an aortic aneurysm image diagnosis support apparatus 100 according to an embodiment of the present invention. The aortic aneurysm image diagnosis support apparatus 100 is an apparatus that detects candidate positions of an aortic aneurysm from CT image data of a living organism including a human body. A physician can diagnose an aortic aneurysm by checking the candidate position of the aortic aneurysm detected by the aortic aneurysm image diagnosis support apparatus 100 and the CT image in the vicinity thereof. The aortic aneurysm image diagnosis support apparatus 100 is realized by one or more computers such as a tablet terminal, a notebook computer, a desktop computer, and a server machine executing software.

[0015] As shown in Figure 1, the aortic aneurysm image diagnostic support device 100 comprises an aortic detection unit 101, an aortic aneurysm candidate detection unit 102, and a detection result display unit 103. The aortic detection unit 101 detects aortas classified into multiple locations from computed tomography images of living organisms. Computed tomography images of living organisms are CT image data in which voxels with brightness values ​​are arranged in three dimensions, and are, for example, a collection of images of cross-sections (axial sections) perpendicular to the body axis. The aortic detection unit 101 detects voxels from among the voxels contained in the CT image data that are classified into aortas of each of multiple locations.

[0016] The aortic aneurysm candidate detection unit 102 detects aortic aneurysm candidates from the aorta detected by the aortic detection unit 101 using conditions appropriate to each of several locations. The detection result display unit 103 displays the aortic aneurysm candidates detected by the aortic aneurysm candidate detection unit 102.

[0017] Figure 2 is a schematic block diagram showing an example configuration of the aortic detection unit 101 in this embodiment. In the example shown in Figure 2, the aortic detection unit 101 comprises a whole aortic detection unit 111, an iliac aorta detection unit 112, a thin line extraction unit 113, and a whisker removal unit 114. The whole aortic detection unit 111 uses a trained neural network to detect the ascending aorta, the aortic arch, the descending aorta, the abdominal aorta, and the iliac aorta from CT image data. The iliac aorta detection unit 112 uses a trained neural network to detect the iliac aorta from the axial cross-sectional data of the CT image data in which the whole aortic detection unit 111 has detected the iliac aorta. The iliac aorta has a more complex structure than the aortas in other parts of the body, but by using a neural network that detects only the iliac aorta, it can be detected with high accuracy. Furthermore, detection in the whole aorta detection unit 111 and the iliac aorta detection unit 112 can be performed using, for example, the method described in Ozgun Cicek et al.'s 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation (https: / / arxiv.org / pdf / 1606.06650.pdf).

[0018] The data representing the ascending aorta, aortic arch, descending aorta, and abdominal aorta detected by the whole aorta detection unit 111, and the data representing the iliac aorta detected by the iliac aorta detection unit 112, are hereafter referred to as aortic data. In other words, aortic data is data representing the voxels in the CT image data that correspond to the ascending aorta, aortic arch, descending aorta, abdominal aorta, and iliac aorta, respectively. For example, the pixel value of each voxel may be a value that represents one of the following: the ascending aorta, aortic arch, descending aorta, abdominal aorta, iliac aorta, or an aorta other than the aorta.

[0019] The thinning unit 113 thins the aorta represented by the aortic data. This thinning is a process that converts a binary image into a line image with a width of 1 pixel. Here, a 3D binary image representing whether or not it is an aorta is converted into a line image with a width of 1 pixel. Known methods such as those by Hilditch, Tamura, and Zhang Suen can be used for thinning. The whisker removal unit 114 removes the parts called whiskers included in the thinning result by the thinning unit 113 to generate thinning data. The thinning data is voxel data representing the centerline of the aorta. The whisker removal unit 114 also detects the whisker parts by converting the thinning result by the thinning unit 113 into a graph structure.

[0020] Figure 3 illustrates an example of aortic data in this embodiment. Figure 3 is a projection of aortic data onto a coronal plane and includes the ascending aorta P1, the aortic arch P2, the descending aorta P3, the abdominal aorta P4, and the iliac aorta P5. The aortic data consists of voxels with pixel values ​​corresponding to each of the ascending aorta P1, the aortic arch P2, the descending aorta P3, the abdominal aorta P4, and the iliac aorta P5.

[0021] Figure 4 illustrates an example of thinning by the thinning unit 113 in this embodiment. Figure 4 shows the aortic data from Figure 3 projected onto the coronal plane as a result of thinning by the thinning unit 113. As shown in Figure 4, the thinning results in obtaining the centerline of the aorta, along with the parts called whiskers that branch off from the centerline.

[0022] Figure 5 is a diagram illustrating the processing of the whisker removal unit 114 in this embodiment. Figure 5 shows the result of converting a portion of the thinning result by the thinning unit 113 into a graph structure. In the graph structure, the thinning result is represented by a set of nodes (vertices) indicated by circles in Figure 5 and edges connecting the nodes. As shown in Figure 5, the whisker removal unit 114 removes the whiskers, which are the targets for removal D, that branch off from the center line. The whisker removal unit 114 may have determination conditions for whether or not a whisker should be removed, depending on the area. Figure 6 is a table showing an example of determination conditions in the whisker removal unit 114 in this embodiment. In the example in Figure 6, the whisker removal unit 114 determines that a whisker is a target for removal D if the length of the branched portion is less than or equal to a threshold, with the threshold being value BTh1 outside the iliac region and value BTh2 in the iliac region. Note that value BTh2 is a smaller value than value BTh1. The aorta in the iliac region has a more complex structure, including branching, compared to aortas in other parts of the body. However, by making the threshold in the iliac region smaller than that in other parts of the body, whisker removal can be performed in the iliac region as well. In the example in Figure 6, the threshold is set differently for the iliac region and other parts of the body, but the threshold may be set differently for each part of the body. In addition, the whisker removal section 114 may determine the shorter of the branches as the target for removal D in parts other than the iliac region.

[0023] Figure 7 is a schematic block diagram showing an example configuration of the aortic aneurysm candidate detection unit 102 in this embodiment. In the example shown in Figure 7, the aortic aneurysm candidate detection unit 102 comprises a vertical cross-section extraction unit 121 and a condition determination unit 122. The vertical cross-section extraction unit 121 extracts cross-sections of the aorta shown in the aortic data, which are perpendicular to the center line shown in the thin line data, at regular intervals along the center line shown in the thin line data. The condition determination unit 122 detects aortic aneurysm candidates by applying conditions (determination criteria) related to the determination size of aortic aneurysms corresponding to each of a plurality of locations to the cross-sections of the aorta extracted by the vertical cross-section extraction unit 121.

[0024] Figure 8 is a schematic diagram illustrating the operation of the vertical section extraction unit 121 in this embodiment. The vertical section extraction unit 121 extracts a plane Cs perpendicular to the centerline Cl of the aorta shown by the thin line data. Figure 9 shows an example of an axial section CT image. Figure 10 shows an example of a CT image of a section perpendicular to the centerline of the aorta. Because the aorta is not always aligned with the body axis, axial section CT images can sometimes appear as if the aorta has been cut diagonally, as in aorta A1 in Figure 9. However, the vertical section extraction unit 121 extracts a plane perpendicular to the centerline, so it can capture a vertical section of the aorta, as in aorta A2 in Figure 10.

[0025] Figure 11 is a diagram illustrating an example of a CT image illustrating the operation of the condition determination unit 122 in this embodiment. The condition determination unit 122 performs ellipse fitting on the aorta in each cross-section extracted by the vertical cross-section extraction unit 121, as shown by ellipse E in Figure 11. Known methods such as the least squares method can be used for ellipse fitting. The condition determination unit 122 obtains the minor axis Dm of the fitted ellipse E as a value representing the diameter of the aorta. The condition determination unit 122 may also use a value other than the minor axis, such as the diameter of the fitted circle, as a value representing the diameter of the aorta.

[0026] Figure 12 is a graph illustrating the operation of the condition determination unit 122 in this embodiment. The graph in Figure 12 is a graph plotting the minor axis acquired by the condition determination unit 122 in the direction of the centerline. In Figure 12, graph Gp1 is a graph plotting the minor axis of the ascending portion. Graph Gp2 is a graph plotting the minor axis of the arch portion. Graph Gp3 is a graph plotting the minor axis of the descending portion. Graph Gp4 is a graph plotting the minor axis of the abdomen. Graph Gp5 is a graph plotting the minor axis of the iliac portion. The values ​​Th1, Th2, Th3, Th4, and Th5 are thresholds for the ascending portion, arch portion, descending portion, abdomen, and iliac portion, respectively. The condition determination unit 122 determines the position of the peak in the direction of the centerline as a candidate aortic aneurysm AC when the peak is the peak when the minor axis is plotted in the direction of the centerline of the aorta as shown in Figure 12, and the value of the peak exceeds the threshold corresponding to the region to which the peak belongs. The condition determination unit 122 may also determine multiple locations as candidates for aortic aneurysms.

[0027] Figure 13 is a table showing the conditions (thresholds) for determining the size of an aortic aneurysm used by the condition determination unit 122 in this embodiment. The conditions used by the condition determination unit 122 are thresholds corresponding to the ascending portion, arch, descending portion, abdomen, and iliac portion. In the example shown in Figure 12, the threshold Th1 for the ascending portion is greater than the threshold Th2 for the arch. Also, the threshold Th2 for the arch is the same as the threshold Th3 for the descending portion. The threshold Th3 for the descending portion is greater than the threshold Th4 for the abdomen. The threshold Th4 for the abdomen is greater than the threshold Th5 for the iliac portion.

[0028] Figures 14, 15, and 16 show display examples G1, G2, G3, and G4 by the detection result display unit 103 in this embodiment. Display example G1 is a graph plotting values ​​representing the diameter of the aorta along the body axis. Display example G2 is a graph plotting values ​​representing the diameter of the aorta along the centerline of the aorta. Display example G3 is an image of the aorta projected onto the coronal plane. Display example G4 is an image of a cross-section perpendicular to the centerline of the aorta. The detection result display unit 103 displays some or all of these side by side, for example, display examples G2, G3, and G4.

[0029] In display example G1 of Figure 14, the vertical axis represents the slice number along the body axis, and the horizontal axis represents the minor axis acquired by the condition determination unit 122. Similarly, in display example G2 of Figure 14, the vertical axis represents the slice number in the axial direction along the centerline, and the horizontal axis represents the minor axis acquired by the condition determination unit 122. The circles in display examples G1 and G2 indicate locations detected as candidate aortic aneurysms, and the numbers attached to the circles represent the slice number and minor axis. Lines Sl1 and Sl2 indicate the slice positions of the CT images, as shown in display example G4, which are displayed together with display examples G1 and G2. Lines Sl1 and Sl2 can be moved up and down using input means such as a mouse or keyboard, and the detection result display unit 103 displays the CT images of the slice numbers corresponding to the moved lines Sl1 and Sl2.

[0030] Display example G3 in Figure 15 is an image projected onto the coronal plane, showing the aorta and its location as indicated by the aortic data. Line Sl3 in display example G3 is the same as lines Sl1 and Sl2 in Figure 14. Display example G4 in Figure 16 is a CT image at the slice numbers corresponding to lines Sl1, Sl2, and Sl3. This CT image is a cross-sectional CT image perpendicular to the centerline, but the detection result display unit 103 may also display an axial CT image. Ellipse E4 in Figure 16 is an ellipse fitted to the aorta by the condition determination unit 122, and its minor axis Dm4 and its value (58 mm) are also displayed in display example G4.

[0031] Figure 17 is a schematic block diagram showing the configuration of the trained model generation device 200 in this embodiment. The trained model generation device 200 generates trained models for the neural networks used in the whole aorta detection unit 111 and the iliac aorta detection unit 112, respectively. The trained model generation device 200 is realized by running software on one or more computers, such as a tablet terminal, notebook computer, desktop computer, or server machine.

[0032] The trained model generation device 200 comprises an aorta designation unit 201, a region boundary designation unit 202, a training data generation unit 203, a whole aorta trained model generation unit 204, and an iliac aorta trained model generation unit 205. The aorta designation unit 201 accepts the designation of the aorta region in each axial cross-section of CT image data. This designation is made for each axial cross-sectional image using an input means such as a mouse or touchpad. The region boundary designation unit 202 accepts the designation of region boundaries of the aorta in the CT image data. Here, the region boundaries are the boundary between the ascending portion and the arch, the boundary between the arch and the descending portion, the boundary between the descending portion and the abdomen, and the boundary between the abdomen and the iliac portion. This boundary designation is made by specifying which slice number's axial cross-section is the boundary using an input means such as a mouse or keyboard. As an example, the boundary position between the descending portion and the abdomen is designated as the celiac artery.

[0033] The training data generation unit 203 generates data indicating which part of the aorta the region of the aorta received by the aorta designation unit 201 belongs to, based on the region boundary designation unit 202. The training data generation unit 203 combines this data with CT image data to create training data for the whole aorta trained model generation unit 204. The training data generation unit 203 also combines data indicating the iliac aorta with axial cross-sectional data from the CT image data that includes the iliac aorta to create training data for the iliac aorta trained model generation unit 205. The whole aorta trained model generation unit 204 uses the training data for the whole aorta trained model generation unit 204 generated by the training data generation unit 203 to train a neural network and generate a trained model that detects the ascending, arch, descending, abdominal, and iliac aorta from the CT image data. The iliac aorta trained model generation unit 205 uses the training data for the iliac aorta trained model generation unit 205 generated by the training data generation unit 203 to train a neural network and generate a trained model that detects the iliac aorta from axial cross-sectional data including the iliac aorta. The trained model generated by the whole aorta trained model generation unit 204 is used by the whole aorta detection unit 111, and the trained model generated by the iliac aorta trained model generation unit 205 is used by the iliac aorta detection unit 112.

[0034] Thus, the aortic aneurysm image diagnostic support device 100 includes an aortic detection unit 101 that detects the aorta classified into multiple parts from computed tomography images of a living organism, and an aortic aneurysm candidate detection unit 102 that detects aortic aneurysm candidates from the detected aorta using conditions appropriate to each of the multiple parts. As a result, the aortic aneurysm image diagnostic support device 100 can detect aortic aneurysm candidates using criteria appropriate to the part.

[0035] The aortic aneurysm image diagnostic support device 100 further includes a thinning line extraction unit 113 that thins the detected aorta, and the aortic aneurysm candidate detection unit 102 includes a vertical cross-section extraction unit 121 that extracts a cross-section of the aorta perpendicular to the line obtained by thinning by the thinning line extraction unit 113, and a condition determination unit 122 that detects aortic aneurysm candidates by applying conditions according to each of a plurality of locations to the cross-section of the aorta. As a result, the aortic aneurysm image diagnostic support device 100 can suppress false detections because it detects aortic aneurysm candidates by applying conditions to a cross-section perpendicular to the centerline of the aorta.

[0036] The aortic aneurysm image diagnostic support device 100 further includes a whisker removal unit 114 that performs whisker removal processing on the lines obtained by thinning by the thinning unit 113, and the determination conditions for whether or not to remove a whisker in the whisker removal unit 114 depend on multiple locations. As a result, the aortic aneurysm image diagnostic support device 100 can apply determination conditions according to the structure of the location and perform whisker removal more accurately.

[0037] The aortic aneurysm image diagnostic support device 100 includes a detection result display unit 103 that displays a graph plotting values ​​representing the aortic diameter along the axial direction of the lines obtained by thinning using the thinning unit 113. This allows the aortic aneurysm image diagnostic support device 100 to intuitively grasp changes in the aortic diameter and support diagnosis.

[0038] Alternatively, the aortic aneurysm image diagnostic support device 100 and the trained model generation device 200 shown in Figures 1 and 17 may be implemented by recording a program for realizing each or part of their functions on a computer-readable recording medium, loading the program recorded on this recording medium into a computer system, and executing it. The term "computer system" here includes hardware such as the operating system and peripheral devices.

[0039] Furthermore, "computer system" shall also include the homepage provisioning environment (or display environment) if a WWW system is being used. Furthermore, "computer-readable recording media" refers to portable media such as flexible disks, magneto-optical disks, ROMs, and CD-ROMs, as well as storage devices such as hard disks built into computer systems. Moreover, "computer-readable recording media" also includes those that dynamically hold programs for a short period of time, such as communication lines used when transmitting programs over networks such as the Internet or communication lines such as telephone lines, and those that hold programs for a certain period of time, such as volatile memory inside computer systems that act as servers or clients in such cases. In addition, the above-mentioned programs may be for the purpose of realizing some of the functions described above, and may also be able to realize the above-mentioned functions in combination with programs already recorded in the computer system.

[0040] Although embodiments of this invention have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments, and design modifications and the like are also included within the scope of the gist of this invention. [Explanation of Symbols]

[0041] 100 Aortic Aneurysm Imaging Support Devices 101 Aortic detection unit 102 Aortic Aneurysm Candidate Detection Unit 103 Detection result display unit 111 Total Aortic Detection Unit 112 Iliac aorta detection section 113 Fine line extraction part 114 Beard removal section 121 Vertical section extraction part 122 Condition judgment section 200 Pre-trained model generator 200 201 Designated Aortic Section 202 Part boundary specification part 203 Training Data Generation Department 204 Whole Aorta Trained Model Generation Unit 205 Iliac Aorta Trained Model Generation Unit

Claims

1. An aortic detection unit that detects the aorta, which is classified into multiple parts, from computed tomography images of living organisms, The aforementioned thinning unit for the detected aorta, A whisker removal unit performs whisker removal on the lines obtained by the thinning process by the aforementioned thinning unit, An aortic aneurysm candidate detection unit detects aortic aneurysm candidate from the detected aorta using conditions for determining the size of an aortic aneurysm corresponding to each of the aforementioned multiple sites. Equipped with, The aforementioned aortic aneurysm candidate detection unit is A cross-section of the aorta, comprising a vertical cross-section extraction unit that extracts a cross-section perpendicular to the line obtained by the thinning by the thinning unit, A condition determination unit that detects a candidate for an aortic aneurysm by applying the above conditions to the cross-section of the aorta. Equipped with, In the aforementioned beard removal section, the criteria for determining whether or not to remove a particular area depend on the multiple areas mentioned above. Aortic aneurysm imaging diagnostic support device.

2. The aortic aneurysm image diagnostic support device according to claim 1, further comprising a detection result display unit that displays side by side a graph plotting a value representing the thickness of the aorta in the axial direction along the line obtained by thinning by the thinning unit, an image of the aorta projected onto the coronal surface, and a cross-sectional image perpendicular to the centerline of the aorta.

3. An aortic aneurysm imaging support device comprising the process of detecting the aorta, which is classified into multiple parts, from images obtained by computed tomography of a living organism, The aortic aneurysm image diagnostic support device includes the process of thinning the detected aorta, The aortic aneurysm image diagnostic support device includes a process of removing whiskers from the lines obtained by the thinning process, The aortic aneurysm image diagnostic support device performs a process of detecting aortic aneurysm candidates from the detected aorta using conditions relating to the size of the aortic aneurysm according to each of the multiple locations. It has, The process for detecting the aforementioned aortic aneurysm candidate is as follows: The aortic aneurysm image diagnostic support device includes a process of extracting a cross-section of the aorta, which is perpendicular to the lines obtained by the thinning process, The aortic aneurysm image diagnostic support device performs the process of detecting a candidate aortic aneurysm by applying the conditions to a cross-section of the aorta. It has, In the process of performing the aforementioned beard removal treatment, the criteria for determining whether or not to remove a particular area depend on the multiple areas mentioned above. Methods for supporting imaging diagnosis of aortic aneurysms.

4. Computers, An aortic detection unit that detects the aorta, which is classified into multiple parts, from computed tomography images of living organisms. A wire extraction unit that thins the detected aorta, A whisker removal unit performs whisker removal on the lines obtained by the thinning process by the aforementioned thinning unit. Aortic aneurysm candidate detection unit detects aortic aneurysm candidates from the detected aorta using conditions related to the size of the aortic aneurysm determined for each of the aforementioned multiple sites. It is a program designed to function as such. The aforementioned aortic aneurysm candidate detection unit is A cross-section of the aorta, comprising a vertical cross-section extraction unit that extracts a cross-section perpendicular to the line obtained by the thinning by the thinning unit, A condition determination unit that detects a candidate for an aortic aneurysm by applying the above conditions to the cross-section of the aorta. Equipped with, In the aforementioned beard removal section, the criteria for determining whether or not to remove a particular area depend on the multiple areas mentioned above. program.

Images