Information processing device, method, and program, learning device, method, and program, and discrimination model

Abstract

A processor according to the present invention acquires a non-contrast CT image of a patient's head and first information representing one of an infarcted region and a major artery occlusion in the non-contrast CT image, and derives second information representing the other of the infarcted region and major artery occlusion in the non-contrast CT image on the basis of the non-contrast CT image and the first information.

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, a method, and a program, a learning device, a method, and a program, and a discrimination model. [Background technology]

[0002] In recent years, advances in medical equipment such as CT (Computed Tomography) and MRI (Magnetic Resonance Imaging) have made it possible to perform diagnostic imaging using higher quality, higher resolution medical images. In particular, when the target area is the brain, diagnostic imaging using CT and MRI images can identify areas of cerebral vascular disorders such as cerebral infarction and cerebral hemorrhage. For this reason, various methods have been proposed to support diagnostic imaging.

[0003] Cerebral infarction is a disease in which brain tissue is damaged by cerebral blood vessel blockage and is known to have a poor prognosis. Once a cerebral infarction develops, irreversible cell death progresses over time, making shortening the time until treatment can begin a key issue. When applying thrombectomy, a typical treatment for cerebral infarction, two pieces of information are required: the extent of the infarction and the presence or absence of large vessel occlusion (LVO) (see Guidelines for Proper Use of Percutaneous Transluminal Cerebral Thrombectomy Devices, 4th Edition, March 2020, p. 12-(1)).

[0004] On the other hand, when diagnosing patients suspected of having a brain disease, the presence or absence of intracerebral hemorrhage is often confirmed before confirming cerebral infarction. Because intracerebral hemorrhage can be clearly identified on non-contrast CT images, patients suspected of having a brain disease are first diagnosed using non-contrast CT images. However, in non-contrast CT images, the pixel value difference between the area of ​​cerebral infarction and other areas is not significant. Furthermore, although it is possible to visually identify a high-absorption structure (hyperdense artery sign (HAS)) that reflects a thrombus causing major artery occlusion, the image is not very clear, making it difficult to identify the location of the occlusion in the major arteries. Thus, using non-contrast CT images, it is often difficult to identify the area of ​​infarction and the location of the occlusion in the major arteries. Therefore, after diagnosing using non-contrast CT images, MRI or contrast CT images are acquired to diagnose whether or not cerebral infarction has occurred, identify the location of the occlusion in the major arteries, and, if cerebral infarction has occurred, confirm the extent of the occlusion.

[0005] However, if MRI and contrast CT images are obtained after a diagnosis using CT images to determine whether or not a cerebral infarction has occurred, the time that has passed since the onset of the infarction will be long, which will delay the start of treatment and result in a higher likelihood of a poor prognosis.

[0006] For this reason, methods have been proposed for automatically extracting infarct regions and occlusion sites of major arteries from non-contrast CT images. For example, Japanese Patent Application Laid-Open No. 2020-054580 proposes a method for identifying infarct regions and thrombus regions using a classifier trained to extract infarct regions from non-contrast CT images and a classifier trained to extract thrombus regions from non-contrast CT images. Summary of the Invention [Problem to be solved by the invention]

[0007] On the other hand, the HAS, which indicates the location of the occlusion of a major artery, appears in different locations depending on which blood vessel is blocked, and its appearance also varies depending on the angle of the cross-section of the CT image relative to the brain, the characteristics of the thrombus, and the degree of occlusion. It can also be difficult to distinguish it from similar structures in the vicinity, such as calcification. Furthermore, the infarcted area occurs in the vascular area controlled by the blood vessel in which the HAS occurred. Therefore, if the location of the occlusion of a major artery can be identified, it will also be easier to identify the infarcted area.

[0008] The present disclosure has been made in consideration of the above circumstances, and aims to enable accurate identification of a major artery occlusion site or an infarcted region using non-contrast CT images of the head. [Means for solving the problem]

[0009] An information processing device according to the present disclosure includes at least one processor, The processor acquires a non-contrast CT image of the patient's head and first information representing either an infarct region or a major artery occlusion location in the non-contrast CT image; Second information representing the other of the infarct region and the occlusion location of the major artery in the non-contrast CT image is derived based on the non-contrast CT image and the first information.

[0010] In addition, in the information processing device according to the present disclosure, the processor may derive the second information using a discriminant model that has been trained to output the second information when a non-contrast CT image and the first information are input.

[0011] In addition, in the information processing device according to the present disclosure, the processor Non The second information may be derived by further using information on regions symmetrical with respect to the midline of the brain in the contrast CT image and at least the non-contrast CT image of the first information.

[0012] In the information processing device according to the present disclosure, the information on the symmetric region may be inverted information obtained by inverting at least the non-contrast CT image of the first information and the non-contrast CT image with respect to the midline of the brain.

[0013] In the information processing device according to the present disclosure, the processor may further derive second information based on at least one of information representing an anatomical region of the brain and clinical information.

[0014] In the information processing device according to the present disclosure, the processor may acquire the first information by extracting either an infarct region or a major artery occlusion site from a non-contrast CT image.

[0015] In addition, in the information processing device according to the present disclosure, the processor derives quantitative information about at least one of the first information and the second information; It may also display quantitative information.

[0016] A learning device according to the present disclosure includes at least one processor, The processor acquires learning data including input data consisting of a non-contrast CT image of the head of a patient suffering from cerebral infarction and first information representing either an infarct region or a major artery occlusion location in the non-contrast CT image, and correct answer data consisting of second information representing the other of either the infarct region or the major artery occlusion location in the non-contrast CT image; By machine learning a neural network using learning data, a discriminant model is constructed that outputs second information when a non-contrast CT image and first information are input.

[0017] The discrimination model according to the present disclosure, when input with a non-contrast CT image of a patient's head and first information representing either the infarct area or the occlusion location of a major artery in the non-contrast CT image, outputs second information representing the other of either the infarct area or the occlusion location of a major artery in the non-contrast CT image.

[0018] The information processing method according to the present disclosure includes acquiring a non-contrast CT image of a patient's head and first information representing either an infarct region or a major artery occlusion location in the non-contrast CT image; Second information representing the other of the infarct region and the occlusion location of the major artery in the non-contrast CT image is derived based on the non-contrast CT image and the first information.

[0019] The learning method according to the present disclosure includes acquiring learning data including input data consisting of a non-contrast CT image of the head of a patient suffering from cerebral infarction and first information representing either an infarct region or a major artery occlusion location in the non-contrast CT image, and correct answer data consisting of second information representing the other of either the infarct region or the major artery occlusion location in the non-contrast CT image; By machine learning a neural network using learning data, a discriminant model is constructed that outputs second information when a non-contrast CT image and first information are input.

[0020] The information processing method and learning method according to the present disclosure may be provided as a program for causing a computer to execute the method. [Effects of the Invention]

[0021] According to the present disclosure, the location of major artery occlusion or infarction region can be identified with high accuracy using non-contrast CT images of the head. [Brief explanation of the drawings]

[0022] [Figure 1] FIG. 1 is a diagram showing a schematic configuration of a medical information system to which an information processing device and a learning device according to a first embodiment of the present disclosure are applied. [Figure 2] FIG. 1 is a diagram showing a schematic configuration of an information processing device and a learning device according to a first embodiment; [Figure 3] Functional configuration diagram of an information processing device and a learning device according to a first embodiment [Figure 4] FIG. 1 is a schematic block diagram showing a configuration of an information derivation unit according to a first embodiment. [Figure 5] FIG. 1 is a diagram illustrating a schematic configuration of U-Net. [Figure 6] Diagram to explain the inversion of feature maps [Figure 7] FIG. 1 is a diagram showing training data for training U-Net in the first embodiment. [Figure 8] Diagram explaining arteries and their territories in the brain [Figure 9] Diagram showing the display screen [Figure 10] 1 is a flowchart showing a learning process performed in the first embodiment. [Figure 11] 1 is a flowchart showing information processing performed in the first embodiment. [Figure 12] FIG. 10 is a schematic block diagram showing the configuration of an information derivation unit according to a second embodiment. [Figure 13] FIG. 10 is a diagram showing training data for training U-Net in the second embodiment. [Figure 14] 10 is a flowchart showing a learning process performed in the second embodiment. [Figure 15] 10 is a flowchart showing information processing performed in the second embodiment. [Figure 16] FIG. 10 is a schematic block diagram showing the configuration of an information derivation unit according to a third embodiment. [Figure 17] FIG. 10 is a diagram showing training data for training U-Net in the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0023] A first embodiment of the present disclosure will be described below with reference to the drawings. Fig. 1 is a hardware configuration diagram showing an overview of a diagnosis support system to which an information processing device and a learning device according to the first embodiment of the present disclosure are applied. As shown in Fig. 1, in the diagnosis support system, an information processing device 1 according to the first embodiment, a three-dimensional image capturing device 2, and an image storage server 3 are connected in a communicable state via a network 4. Note that the information processing device 1 includes a learning device according to this embodiment.

[0024] The three-dimensional imaging device 2 is a device that captures an image of a diagnostic target region of a subject to generate a three-dimensional image representing the region, and is specifically a CT device, an MRI device, a PET device, or the like. The medical images generated by the three-dimensional imaging device 2 are transmitted to and stored in an image storage server 3. In this embodiment, the diagnostic target region of the subject is the brain, and the three-dimensional imaging device 2 is a CT device that generates a three-dimensional CT image G0 of the head of the subject. In this embodiment, the CT image G0 is a non-contrast CT image acquired by performing imaging without using a contrast agent.

[0025] The image storage server 3 is a computer that stores and manages various data, and is equipped with a large-capacity external storage device and database management software. The image storage server 3 communicates with other devices via a wired or wireless network 4, sending and receiving image data and the like. Specifically, it acquires various data, including image data of CT images generated by the 3D imaging device 2, via the network, and stores and manages them on recording media such as a large-capacity external storage device. The image storage server 3 also stores training data for building a discrimination model, as will be described later. The storage format of the image data and communication between devices via the network 4 are based on protocols such as DICOM (Digital Imaging and Communication in Medicine). It is based on Col.

[0026] Next, an information processing device and a learning device according to a first embodiment of the present disclosure will be described. FIG. 2 describes the hardware configuration of the information processing device and the learning device according to the first embodiment. As shown in FIG. 2, the information processing device and the learning device (hereinafter referred to as the information processing device) 1 includes a CPU (Central Processing Unit) 11, non-volatile storage 13, and memory 16 as a temporary storage area. The information processing device 1 also includes a display 14 such as a liquid crystal display, an input device 15 such as a keyboard and a mouse, and a network I / F (Interface) 17 connected to a network 4. The CPU 11, the storage 13, the display 14, the input device 15, the memory 16, and the network I / F 17 are connected to a bus 18. The CPU 11 is an example of a processor in the present disclosure.

[0027] The storage 13 is realized by a hard disk drive (HDD), a solid state drive (SSD), a flash memory, etc. The storage 13 as a storage medium stores an information processing program 12A and a learning program 12B. The CPU 11 reads the information processing program 12A and the learning program 12B from the storage 13, expands them in the memory 16, and executes the expanded information processing program 12A and the learning program 12B.

[0028] Next, the functional configuration of the information processing device according to the first embodiment will be described. Fig. 3 is a diagram showing the functional configuration of the information processing device according to the first embodiment. As shown in Fig. 3, the information processing device 1 includes an information acquisition unit 21, an information derivation unit 22, a learning unit 23, a quantitative value derivation unit 24, and a display control unit 25. The CPU 11 executes an information processing program 12A to function as the information acquisition unit 21, the information derivation unit 22, the quantitative value derivation unit 24, and the display control unit 25. The CPU 11 also executes a learning program 12B to function as the learning unit 23.

[0029] The information acquisition unit 21 acquires a non-contrast CT image G0 of the patient's head from the image storage server 3. The information acquisition unit 21 also acquires, from the image storage server 3, training data for training a neural network to construct a discrimination model, which will be described later.

[0030] The information derivation unit 22 acquires first information representing either the infarct region or the occlusion location of the major artery in the CT image G0, and derives second information representing the other of either the infarct region or the occlusion location of the major artery in the CT image G0 based on the CT image G0 and the first information. In this embodiment, the first information representing the infarct region in the CT image G0 is acquired, and the second information representing the occlusion location of the major artery in the CT image G0 is derived based on the CT image G0 and the first information.

[0031] FIG. 4 is a schematic block diagram showing the configuration of the information derivation unit in the first embodiment. As shown in FIG. 4, the information derivation unit 22 includes a first discriminant model 22A and a second discriminant model 22B. The first discriminant model 22A is constructed by machine learning a convolutional neural network (CNN) so as to extract, as first information, a brain infarct region from a CT image G0 to be processed. The first discriminant model 22A can be constructed, for example, by using the technique described in Japanese Patent Application Laid-Open No. 2020-054580. Specifically, the first discriminant model 22A can be constructed by machine learning a CNN using non-contrast CT images of the head and mask images representing the infarct region in the non-contrast CT images as training data. As a result, the first discriminant model 22A extracts the infarct region in the CT image G0 from the CT image G0 and outputs a mask image M0 representing the infarct region in the CT image G0.

[0032] The second discriminant model 22B is constructed by machine learning a U-Net, a type of convolutional neural network, using a large amount of training data to extract the occluded location of a major artery from the CT image G0 based on the CT image G0 and a mask image M0 representing the infarct region in the CT image G0 as second information. FIG. 5 is a diagram schematically illustrating the configuration of the U-Net. As shown in FIG. 5, the second discriminant model 22B is composed of nine layers, namely, a first layer 31 to a ninth layer 39. In this embodiment, when deriving the second information, information on regions symmetrical about the brain midline in at least the CT image G0 out of the CT image G0 and the mask image M0 representing the infarct region in the CT image G0 is used. The information on regions symmetrical about the brain midline will be described later.

[0033] In this embodiment, the CT image G0 and a mask image M0 representing the infarct region in the CT image G0 are combined and input to the first layer 31. Note that depending on the CT image G0, the brain midline may be tilted relative to the perpendicular bisector of the CT image G0. In such cases, the brain in the CT image G0 is rotated so that the brain midline coincides with the perpendicular bisector of the CT image G0. In this case, a similar rotation process must also be performed on the mask image M0.

[0034] The first layer 31 has two convolution layers and outputs a feature map F1 obtained by integrating two feature maps, the convolved CT image G0 and the mask image M0. The integrated feature map F1 is input to the ninth layer 39, as indicated by the dashed line in FIG. 5 . The integrated feature map F1 is then pooled to reduce its size by half and input to the second layer 32. In FIG. 5 , the pooling is indicated by a downward arrow. In this embodiment, a 3×3 kernel is used for the convolution, for example, but is not limited to this. In addition, the maximum value of four pixels is used for the pooling, but is not limited to this.

[0035] The second layer 32 has two convolutional layers, and the feature map F2 output from the second layer 32 is input to the eighth layer 38, as indicated by the dashed line in Figure 5. The feature map F2 is also pooled to reduce its size by half and then input to the third layer 33.

[0036] The third layer 33 also has two convolutional layers, and the feature map F3 output from the third layer 33 is input to the seventh layer 37, as indicated by the dashed line in Figure 5. The feature map F3 is also pooled to reduce its size by half and then input to the fourth layer 34.

[0037] Furthermore, in this embodiment, when deriving the second information, information on regions symmetrical with respect to the brain midline in the CT image G0 and the mask image M0 representing the infarct region of the CT image G0 is used. Therefore, in the third layer 33 of the second discriminant model 22B, the pooled feature map F3 is flipped horizontally with respect to the brain midline to derive an inverted feature map F3A. The inverted feature map F3A is an example of inverted information of the present disclosure. FIG. 6 is a diagram for explaining the inversion of the feature map. As shown in FIG. 6, the feature map F3 is flipped horizontally with respect to the brain midline C0 to derive the inverted feature map F3A. Note that, although the inverted information is generated within the U-Net in this embodiment, when the CT image G0 and the mask image M0 are input to the first layer 31, an inverted image of at least the CT image G0 may be generated from among the CT image G0 and the mask image M0, and the CT image G0, the inverted image of the CT image G0, and the mask image M0 may be combined and input to the first layer 31. Furthermore, an inverted image of the mask image M0 may be generated in addition to an inverted image of the CT image G0, and the CT image G0, the inverted image of the CT image G0, the mask image M0, and the inverted image of the mask image M0 may be combined and input to the first layer 31. In this case, the inverted image may be generated by rotating the brain in the CT image G0 or by rotating the mask in the mask image M0 so that the midline of the brain coincides with the perpendicular bisector of the CT image G0 and the mask image M0.

[0038] The fourth layer 34 also has two convolutional layers, and the pooled feature map F3 and the inverted feature map F3A are input to the first convolutional layer. The feature map F4 output from the fourth layer 34 is input to the sixth layer 36, as shown by the dashed line in Figure 5. The feature map F4 is also pooled and reduced in size by half, and then input to the fifth layer 35.

[0039] The fifth layer 35 has one convolutional layer, and the feature map F5 output from the fifth layer 35 is upsampled to double its size and input to the sixth layer 36. In FIG. 5, upsampling is indicated by an upward arrow.

[0040] The sixth layer 36 has two convolution layers, and performs a convolution operation by integrating the feature map F4 from the fourth layer 34 and the upsampled feature map F5 from the fifth layer 35. The feature map F6 output from the sixth layer 36 is upsampled to double its size and input to the seventh layer 37.

[0041] The seventh layer 37 has two convolution layers and performs a convolution operation by integrating the feature map F3 from the third layer 33 and the upsampled feature map F6 from the sixth layer 36. The feature map F7 output from the seventh layer 37 is upsampled and input to the eighth layer 38.

[0042] The eighth layer 38 has two convolution layers and performs a convolution operation by integrating the feature map F2 from the second layer 32 and the upsampled feature map F7 from the seventh layer 37. The feature map output from the eighth layer 38 is upsampled and input to the ninth layer 39.

[0043] The ninth layer 39 has three convolution layers and performs a convolution operation by integrating the feature map F1 from the first layer 31 and the upsampled feature map F8 from the eighth layer 38. The feature map F9 output from the ninth layer 39 is an image in which the occlusion areas of the major arteries in the CT image G0 are extracted.

[0044] 7 is a diagram showing training data for training U-Net in the first embodiment. As shown in FIG. 7, training data 40 consists of input data 41 and correct answer data 42. The input data 41 consists of a non-contrast CT image 43 and a mask image 44 representing an infarct region in the non-contrast CT image 43. The correct answer data 42 is a mask image representing the location of occlusion in a major artery in the non-contrast CT image 43.

[0045] In this embodiment, a large amount of training data 40 is stored in the image storage server 3, and the training data 40 is acquired from the image storage server 3 by the information acquisition unit 21 and used for training the U-Net by the training unit 23.

[0046] The learning unit 23 inputs a non-contrast CT image 43 and a mask image 44, which are input data 41, into the U-Net, and causes the U-Net to output an image representing the occlusion site of a major artery in the non-contrast CT image 43. Specifically, the learning unit 23 causes the U-Net to extract the HAS in the non-contrast CT image 43 and output a mask image in which the HAS portion is masked. The learning unit 23 derives the difference between the output image and the ground truth data 42 as a loss and learns the connection weights and kernel coefficients of each layer in the U-Net to minimize the loss. Note that perturbations may be applied to the mask image 44 during learning. Examples of perturbations include applying morphological processing to the mask with random probability or zero-filling the mask. Applying perturbations to the mask image 44 can accommodate patterns seen in hyperacute cerebral infarction cases, where only thrombi appear on the image without significant infarction areas, and further prevents the second discriminant model 22B from relying too heavily on the input mask image during discrimination.

[0047] The learning unit 23 then repeatedly performs learning until the loss becomes equal to or less than a predetermined threshold. As a result, when a non-contrast CT image G0 and a mask image M0 representing an infarct region in the CT image G0 are input, a second discriminant model 22B is constructed that extracts the occluded area of ​​a major artery included in the CT image G0 as second information and outputs a mask image H0 representing the occluded area of ​​a major artery in the CT image G0. Note that the learning unit 23 may construct the second discriminant model 22B by repeatedly performing learning a predetermined number of times.

[0048] The configuration of the U-Net constituting the second discriminant model 22B is not limited to that shown in Fig. 5. For example, in the U-Net shown in Fig. 5, the inverted feature map F3A is derived from the feature map F3 output from the third layer 33, but the inverted feature map may be used in any layer in the U-Net. Furthermore, the number of convolutional layers in each layer in the U-Net is not limited to that shown in Fig. 5.

[0049] The quantitative value derivation unit 24 derives a quantitative value for at least one of the infarct region and the occlusion site of the major artery derived by the information derivation unit 22. The quantitative value is an example of quantitative information in the present disclosure. In this embodiment, the quantitative value derivation unit 24 derives quantitative values ​​for both the infarct region and the occlusion site of the major artery, but may derive a quantitative value for either the infarct region or the occlusion site of the major artery. Since the CT image G0 is a three-dimensional image, the quantitative value derivation unit 24 may derive the volume of the infarct region, the volume of the occlusion site of the major artery, and the length of the occlusion site of the major artery as quantitative values. Furthermore, the quantitative value derivation unit 24 may derive the ASPECTS score as a quantitative value.

[0050] "ASPECTS" is an abbreviation for Alberta Stroke Program Early CT Score, and is a scoring method that quantifies early CT signs in plain CT for cerebral infarction in the middle cerebral artery region. Specifically, when the medical image is a CT image, the middle cerebral artery region is divided into 10 regions in two representative cross sections (at the basal ganglia level and the corona radiata level), and the presence or absence of early ischemic changes is evaluated for each region, with positive areas scored using a demerit system. In ASPECTS, a lower score indicates a larger area of ​​infarction. The quantitative value derivation unit 24 can derive a score depending on whether the infarction region is included in the above 10 regions.

[0051] The quantitative value derivation unit 24 may also identify the control area of ​​the occluded blood vessel based on the occlusion location of the major artery, and derive the overlap (volume) between the control area and the infarct area as a quantitative value. FIG. 8 is a diagram for explaining arteries and their control areas in the brain. FIG. 8 shows a slice image S1 at a certain cross section of the CT image G0. As shown in FIG. 8, the brain includes an anterior cerebral artery (ACA) 51, a middle cerebral artery (MCA) 52, and a posterior cerebral artery (PCA) 53. Although not shown, the brain also includes an internal carotid artery (ICA). The brain is divided into left and right anterior cerebral artery-fed regions 61L, 61R, middle cerebral artery-fed regions 62L, 62R, and posterior cerebral artery-fed regions 63L, 63R, each of which has blood flow controlled by an anterior cerebral artery 51, a middle cerebral artery 52, and a posterior cerebral artery 53. In Figure 8, the right side is the left hemisphere of the brain.

[0052] The control region can be specified by aligning the CT image G0 with a standard brain image prepared in advance in which the control region has been specified.

[0053] The quantitative value deriving unit 24 identifies the artery in which the occlusion site of the major artery is located and identifies the area of ​​the brain controlled by the identified artery. For example, if the occlusion site of the major artery is located in the left anterior cerebral artery, the area of ​​control is determined as the anterior cerebral artery. artery The infarcted area is located downstream of the location of the thrombus in the artery. artery Therefore, the quantitative value deriving unit 24 calculates the quantitative value of the anterior cerebral cortex in the CT image G0. artery The volume of the infarct region relative to the volume of the control region 61L may be derived as a quantitative value.

[0054] The display control unit 25 displays the patient's CT image G0 and the quantitative values ​​on the display 14. FIG. 9 is a diagram showing the display screen. As shown in FIG. 9, slice images included in the patient's CT image G0 are displayed on a display screen 70 in a switchable manner based on the operation of the input device 15. A mask 71 of the infarct region is superimposed on the CT image G0. An arrow-shaped mark 72 indicating the occlusion site of the major artery is also superimposed. A quantitative value 73 derived by the quantitative value derivation unit 24 is displayed on the right side of the CT image G0. Specifically, the volume of the infarct region (40 ml), the length of the occlusion site of the major artery (HAS length: 10 mm), and the volume of the occlusion site of the major artery (HAS volume: 0.1 ml) are displayed.

[0055] Next, the processing performed in the first embodiment will be described. Fig. 10 is a flowchart showing the learning processing performed in the first embodiment. It is assumed that the learning data is acquired from the image storage server 3 and stored in the storage 13. First, the learning unit 23 inputs input data 41 included in the learning data 40 to the U-Net (step ST1), and causes the U-Net to extract major artery occlusion sites (step ST2). The learning unit 23 then derives a loss from the extracted major artery occlusion sites and the ground truth data 42 (step ST3), and determines whether the loss is equal to or less than a predetermined threshold (step ST4).

[0056] If step ST4 is negative, the process returns to step ST1, and the learning unit 23 repeats the processes of steps ST1 to ST4. If step ST4 is positive, the process ends. In this way, the second discriminant model 22B is constructed.

[0057] 11 is a flowchart showing information processing performed in the first embodiment. It is assumed that the non-contrast CT image G0 to be processed is acquired from the image storage server 3 and stored in the storage 13. First, the information derivation unit 22 derives the infarct region in the CT image G0 using the first discriminant model 22A (step ST11). Furthermore, the information derivation unit 22 derives the occlusion location of the major artery in the CT image G0 using the second discriminant model 22B based on the CT image G0 and a mask image M0 representing the infarct region (step ST12).

[0058] Next, the quantitative value derivation unit 24 derives quantitative values ​​based on the information on the infarct region and the occlusion site of the main artery (step ST13). Then, the display control unit 25 displays the CT image G0 and the quantitative values ​​(step ST14), and the process ends.

[0059] In this way, in the first embodiment, the occlusion location of the major artery in the CT image G0 is derived based on the non-contrast CT image G0 of the patient's head and the infarct area in the CT image G0. This allows the infarct area to be taken into consideration, so the occlusion location of the major artery can be identified with high accuracy in the CT image G0.

[0060] Brain diseases such as cerebral infarction rarely occur simultaneously in both the left and right hemispheres of the brain. Therefore, by using an inverted feature map F3A, which is created by inverting the feature map F3 with respect to the midline C0 of the brain, it is possible to identify the location of major artery occlusion while comparing the features of the left and right hemispheres. This allows for accurate identification of the location of major artery occlusion.

[0061] Furthermore, displaying quantitative values ​​makes it easier for doctors to determine treatment strategies based on the quantitative values. For example, displaying the volume or length of the occluded area of ​​a major artery makes it easier to determine the type or length of an instrument to be used when applying thrombectomy.

[0062] Next, a second embodiment of the present disclosure will be described. Note that the configuration of the information processing device in the second embodiment is the same as the configuration of the information processing device in the first embodiment, and only the processing performed is different, so detailed description of the device will be omitted here.

[0063] FIG. 12 is a schematic block diagram showing the configuration of the information derivation unit according to the second embodiment. As shown in FIG. 12, the information derivation unit 82 according to the second embodiment includes a first discriminant model 82A and a second discriminant model 82B. The first discriminant model 82A according to the second embodiment is constructed by machine learning a CNN to extract the occlusion site of a major artery from the CT image G0 as the first information. The construction of the first discriminant model 82A can be performed using, for example, the technique described in Japanese Patent Application Laid-Open No. 2020-054580. Specifically, the first discriminant model 82A can be constructed by machine learning a CNN using non-contrast CT images of the head and the occlusion site of a major artery in the non-contrast CT images as training data.

[0064] The second discriminant model 82B in the second embodiment is constructed by machine learning a U-Net using a large amount of training data so as to extract a cerebral infarction region from the CT image G0 as second information, based on the CT image G0 and a mask image M1 representing the occlusion site of a major artery in the CT image G0. Note that the configuration of the U-Net is the same as that of the first embodiment, and therefore a detailed description thereof will be omitted here.

[0065] Fig. 13 is a diagram showing training data for training U-Net in the second embodiment. As shown in Fig. 13, training data 90 consists of input data 91 and correct answer data 92. The input data 91 consists of a non-contrast CT image 93 and a mask image 94 representing the occlusion site of a major artery in the non-contrast CT image 93. The correct answer data 92 is a mask image representing the infarct region in the non-contrast CT image 93.

[0066] In the second embodiment, the learning unit 23 constructs a second discriminant model 82B by training a U-Net using a large amount of training data 90 shown in Fig. 13. As a result, when a CT image G0 and a mask image M1 representing the occlusion site of a major artery are input, the second discriminant model 82B in the second embodiment extracts an infarct region in the CT image G0 and outputs a mask image K0 representing the infarct region. Note that in the second embodiment, the second discriminant model 82B may further extract the infarct region by using information on regions symmetrical with respect to the midline of the brain in at least the CT image G0 out of the CT image G0 and the mask image M1.

[0067] Next, the processing performed in the second embodiment will be described. Fig. 14 is a flowchart showing the learning processing performed in the second embodiment. It is assumed that the learning data is acquired from the image storage server 3 and stored in the storage 13. First, the learning unit 23 inputs input data 91 included in the learning data 90 to the U-Net (step ST21) and causes the U-Net to extract an infarct region (step ST22). Then, the learning unit 23 derives a loss from the extracted infarct region and the correct answer data 92 (step ST23), and determines whether the loss is equal to or less than a predetermined threshold (step ST24).

[0068] If step ST24 is negative, the process returns to step ST21, and the learning unit 23 repeats the processes of steps ST21 to ST24. If step ST24 is positive, the process ends. In this way, the second discriminant model 82B is constructed.

[0069] 15 is a flowchart showing information processing performed in the second embodiment. It is assumed that the non-contrast CT image G0 to be processed is acquired from the image storage server 3 and stored in the storage 13. First, the information derivation unit 82 derives the occlusion site of the major artery in the CT image G0 using the first discriminant model 82A (step ST31). The information derivation unit 82 also derives the infarct region in the CT image G0 using the second discriminant model 82B based on the CT image G0 and a mask image representing the occlusion site of the major artery (step ST32).

[0070] Next, the quantitative value derivation unit 24 derives quantitative values ​​based on the information on the infarct region and the occlusion site of the main artery (step ST33). Then, the display control unit 25 displays the CT image G0 and the quantitative values ​​(step ST34), and the process ends.

[0071] In this way, in the second embodiment, the infarct region in the CT image G0 is derived based on the non-contrast CT image G0 of the patient's head and the occlusion location of the major artery in the CT image G0. This allows the occlusion location of the major artery to be taken into consideration, making it possible to accurately identify the infarct region in the CT image G0.

[0072] Next, a third embodiment of the present disclosure will be described. Note that the configuration of the information processing device in the third embodiment is the same as the configuration of the information processing device in the first embodiment, and only the processing performed is different, so detailed description of the device will be omitted here.

[0073] Fig. 16 is a schematic block diagram showing the configuration of an information derivation unit according to the third embodiment. As shown in Fig. 16, the information derivation unit 83 according to the third embodiment has a first discriminant model 83A and a second discriminant model 83B. The first discriminant model 83A in the third embodiment is constructed by machine learning CNN so as to extract an infarct region from the CT image G0 as first information, similar to the first discriminant model 22A in the first embodiment.

[0074] The second discriminant model 83B in the third embodiment is constructed by machine learning a U-Net using a large amount of training data so as to extract the occlusion site of a major artery from the CT image G0 as second information, based on the CT image G0, a mask image M0 representing the infarct region in the CT image G0, and at least one of information representing the anatomical region of the brain and clinical information (hereinafter referred to as additional information A0). Note that the configuration of the U-Net is the same as that of the first embodiment, and therefore a detailed description thereof will be omitted here.

[0075] Fig. 17 is a diagram showing training data for training U-Net in the third embodiment. As shown in Fig. 17, training data 100 consists of input data 101 and supervised data 102. The input data 101 consists of a non-contrast CT image 103, a mask image 104 representing an infarct region in the non-contrast CT image 103, and at least one of information representing an anatomical region and clinical information (hereinafter referred to as additional information) 105. The supervised data 102 is a mask image representing the location of a major artery occlusion in the non-contrast CT image 103.

[0076] Here, as information representing the anatomical region, for example, a mask image of a vascular region where an infarct region exists in the non-contrast CT image 103 can be used. Also, a mask image of an ASPECTS region where an infarct region exists in the non-contrast CT image 103 can be used as information representing the anatomical region. As clinical information, the ASPECTS score for the non-contrast CT image 103 and the National Institutes of Health Stroke Scale (NIHSS) for the patient from whom the non-contrast CT image 103 was acquired can be used. The NIHSS is one of the most widely used evaluation methods worldwide as an evaluation scale for the neurological severity of stroke.

[0077] In the third embodiment, the learning unit 23 constructs a second discriminant model 83B by training a U-Net using a large amount of training data 100 shown in Fig. 17. As a result, when a CT image G0, a mask image M0 representing an infarct region, and additional information A0 are input, the second discriminant model 83B in the third embodiment extracts the occluded area of ​​a major artery from the CT image G0 and outputs a mask image H0 representing the occluded area of ​​a major artery.

[0078] The learning process in the third embodiment differs from the first embodiment only in that additional information A0 is used, and therefore a detailed description of the learning process will be omitted. Also, the information processing in the third embodiment differs from the first embodiment only in that the information input to the second discriminant model 83B includes additional information A0 of the patient in addition to the CT image G0 and the mask image representing the infarct region, and therefore a detailed description of the information processing will be omitted.

[0079] In the third embodiment, the location of the occlusion of the major artery in the CT image G0 is derived based on additional information in addition to the non-contrast CT image G0 of the patient's head and the infarct area in the CT image G0, thereby enabling the location of the occlusion of the major artery in the CT image G0 to be identified with higher accuracy.

[0080] In the third embodiment, when a CT image G0, a mask image M0 representing an infarcted region, and additional information A0 are input, second discriminant model 83B is constructed to extract the occluded area of ​​a major artery in the CT image G0, but this is not limiting. Second discriminant model 83B may be constructed to extract the occluded area of ​​a major artery in the CT image G0 when a CT image G0, a mask image representing an occluded area of ​​a major artery, and additional information are input.

[0081] In each of the above embodiments, the second discriminant model derives the second information (i.e., the infarct area or the location of the occlusion of a major artery) using information on regions symmetrical about the brain midline in the CT image G0 and the first information, but this is not limiting. The second discriminant model may be constructed so as to derive the second information without using information on regions symmetrical about the brain midline in the CT image G0 and the first information.

[0082] In addition, although the second discriminant model is constructed using U-Net in each of the above embodiments, the present invention is not limited to this and the second discriminant model may be constructed using a convolutional neural network other than U-Net.

[0083] In each of the above embodiments, the first information (i.e., the infarct region or the occlusion location of the major artery) is derived from the CT image G0 using CNN in the first discriminant model 22A, 82A, 83A of the information derivation unit 22, 82, 83, but this is not limited to this. The information derivation unit may acquire, as the first information, a mask image generated by a doctor identifying the infarct region or the occlusion location of the major artery by interpreting the CT image G0, without using the first discriminant model, and derive the second information from the mask image.

[0084] In the above embodiments, the information derivation units 22, 82, and 83 derive the infarct region and the occluded portion of the major artery, but this is not limiting. A bounding box surrounding the infarct region and the occluded portion of the major artery may be derived.

[0085] Furthermore, in the above embodiment, the following various processors can be used as the hardware structure of the processing units that perform various processes, such as the information acquisition unit 21, the information derivation unit 22, the learning unit 23, the quantitative value derivation unit 24, and the display control unit 25 in the information processing device 1. As described above, the various processors include a CPU, which is a general-purpose processor that executes software (programs) and functions as various processing units, as well as dedicated electrical circuits that are processors having a circuit configuration specifically designed to perform specific processes, such as a programmable logic device (PLD), which is a processor whose circuit configuration can be changed after manufacture, such as an FPGA (Field Programmable Gate Array), and an ASIC (Application Specific Integrated Circuit).

[0086] A single processing unit may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs or a combination of a CPU and an FPGA). Also, multiple processing units may be configured with a single processor. Examples of multiple processing units configured with a single processor include, first, a configuration in which one processor is configured with a combination of one or more CPUs and software, as typified by client and server computers, and this processor functions as multiple processing units. Second, a configuration in which a processor is used to realize the functions of an entire system including multiple processing units on a single IC (Integrated Circuit) chip, as typified by a System on Chip (SoC). In this way, the various processing units are configured with one or more of the above-mentioned various processors as a hardware structure.

[0087] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements. [Explanation of symbols]

[0088] 1. Information processing equipment 2. 3D imaging device 3. Image storage server 4 Network 11 CPU 12A Information Processing Program 12B Study Program 13. Storage 14 Display 15 Input Devices 16 memory 17 Network I / F 18 Bus 21 Information Acquisition Department 22,82,83 Information derivation part 22A, 82A, 83A First discriminant model 22B, 82B, 83B Second discriminant model 23 Learning Department 24 Quantitative value derivation section 25 Display control unit 31 1st layer 32 2nd layer 33 3rd layer 34 4th layer 35 5th layer 36 Layer 6 37 7th layer 38 8th layer 39 9th layer 40, 90, 100 training data 41,91,101 input data 42,92,102 correct data 43,93,103 Non-contrast CT image 44,94,104 mask images 51 Anterior cerebral artery 52 Middle cerebral artery 53 Posterior cerebral artery 61L,61R Anterior cerebral artery innervation area 62L, 62R Middle cerebral artery distribution area 63L,63R Posterior cerebral artery innervation area 70 displays screen 71 Mask 72 marks 73 Quantitative Value 105 Additional Information A0 Additional Information G0 CT image H0, K0, M0, M1 mask images

Claims

1. at least one processor; The processor: acquiring a non-contrast CT image of the patient's head and first information representing either an infarct region or a major artery occlusion site in the non-contrast CT image; An information processing device that derives the second information based on the acquired non-contrast CT image and the first information using a discrimination model that has been trained to output second information representing the other of either the infarction area or the occlusion location of a major artery in the non-contrast CT image when the non-contrast CT image and the first information are input.

2. The information processing device according to claim 1 , wherein the processor further uses information of regions symmetrical with respect to the midline of the brain in at least the non-contrast CT image out of the non-contrast CT image and the first information to derive the second information.

3. The information processing device according to claim 2 , wherein the information on the symmetric region is inverted information obtained by inverting at least the non-contrast CT image of the non-contrast CT image and the first information with respect to the midline of the brain.

4. The information processing device according to claim 1 , wherein the processor further derives the second information based on at least one of information representing an anatomical region of the brain and clinical information.

5. The information processing device according to claim 1 , wherein the processor acquires the first information by extracting either the infarct region or the occlusion site of the major artery from the non-contrast CT image.

6. the processor deriving quantitative information about at least one of the first information and the second information; The information processing device according to claim 1 , wherein the quantitative information is displayed.

7. at least one processor; The processor: acquiring learning data including input data consisting of a non-contrast CT image of the head of a patient suffering from cerebral infarction and first information representing either an infarct region or a major artery occlusion location in the non-contrast CT image, and correct answer data consisting of second information representing the other of either the infarct region or the major artery occlusion location in the non-contrast CT image; A learning device that constructs a discriminant model that outputs the second information when the non-contrast CT image and the first information are input, by machine learning a neural network using the learning data.

8. A discrimination model that causes a computer to function such that, when a non-contrast CT image of a patient's head and first information representing either the infarct area or the occlusion location of a major artery in the non-contrast CT image are input, the computer outputs second information representing the other of either the infarct area or the occlusion location of a major artery in the non-contrast CT image.

9. acquiring a non-contrast CT image of the patient's head and first information representing either an infarct region or a major artery occlusion site in the non-contrast CT image; An information processing method that derives the second information based on the acquired non-contrast CT image and the first information using a discriminant model that has been trained to output second information that represents the other of the infarction area and the occlusion location of a major artery in the non-contrast CT image when the non-contrast CT image and the first information are input.

10. acquiring learning data including input data consisting of a non-contrast CT image of the head of a patient suffering from cerebral infarction and first information representing either an infarct region or a major artery occlusion location in the non-contrast CT image, and correct answer data consisting of second information representing the other of either the infarct region or the major artery occlusion location in the non-contrast CT image; A learning method for constructing a discriminant model that outputs the second information when the non-contrast CT image and the first information are input, by machine learning a neural network using the learning data.

11. acquiring a non-contrast CT image of the patient's head and first information representing either an infarct region or a major artery occlusion site in the non-contrast CT image; and a procedure for deriving the second information based on the acquired non-contrast CT image and the first information using a discrimination model that has been trained to output second information that represents the other of the infarction area and the occlusion location of a major artery in the non-contrast CT image when the non-contrast CT image and the first information are input.

12. a step of acquiring learning data including input data consisting of a non-contrast CT image of the head of a patient suffering from cerebral infarction and first information representing either an infarct region or a major artery occlusion location in the non-contrast CT image, and correct answer data consisting of second information representing the other of either the infarct region or the major artery occlusion location in the non-contrast CT image; and a procedure for constructing a discriminant model that outputs the second information when the non-contrast CT image and the first information are input, by machine learning a neural network using the learning data.

Images