Computer program, brain dysfunction determination assistance device, and brain dysfunction determination assistance method

By determining the dominant hemisphere controlling language and optimizing test order, the system addresses delays in brain dysfunction assessment, enhancing diagnostic efficiency and prognosis.

WO2025164696A1PCT designated stage Publication Date: 2025-08-07TERUMO KK
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2025/002891
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-01-30
Filing Date
2025-01-30
Publication Date
2025-08-07

AI Technical Summary

Technical Problem

Existing methods for assessing brain dysfunction, such as stroke, often require unnecessary testing due to a predetermined order of tests, which can delay diagnosis and worsen prognosis, especially when the dysfunction affects functions controlled by one cerebral hemisphere.

Method used

A computer program and device that determine the dominant hemisphere controlling language and selectively prioritize relevant tests based on the hemisphere affected, optimizing the order of assessments to minimize unnecessary testing.

Benefits of technology

This approach reduces the time required to diagnose brain dysfunction by prioritizing necessary tests and eliminating redundant ones, potentially improving patient prognosis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2025002891_07082025_PF_FP_ABST
    Figure JP2025002891_07082025_PF_FP_ABST
Patent Text Reader

Abstract

The present invention provides a computer program, a brain dysfunction determination assistance device, and a brain dysfunction determination assistance method that can shorten the time to determine brain dysfunction. The computer program causes a computer to execute a process comprising: acquiring dominant hemisphere information about the dominant hemisphere that governs a subject's language; acquiring abnormality information due to brain dysfunction in the subject; determining on the basis of the acquired abnormality information whether or not the abnormality of the subject is due to impairment of a function governed by only the cerebral hemisphere on one side; and upon determining that the abnormality is due to impairment of a function governed by only the cerebral hemisphere on one side, selecting on the basis of the acquired dominant hemisphere information a required test item from among a plurality of test items in a brain dysfunction determination test.
Need to check novelty before this filing date? Find Prior Art

Description

Computer program, brain dysfunction assessment support device, and brain dysfunction assessment support method

[0001] The present disclosure relates to a computer program, a brain dysfunction diagnosis support device, and a brain dysfunction diagnosis support method.

[0002] In cases of stroke and other brain dysfunction, damage often occurs on the opposite side of the body from the damaged area of ​​the brain. It is also known that the type of dysfunction that occurs varies depending on the location of the brain damage. In many cases of brain dysfunction like this, the time it takes for the patient to be treated or transported to a specialist is extremely important, and starting appropriate treatment early is also very important for improving the prognosis.

[0003] Patent Document 1 discloses a computer-assisted triage system that matches patients with emergency conditions such as stroke or LVO (Large Vessel Occlusion) with specialists and can shorten the time it takes to transport them to a secondary medical facility where a specialist is available.

[0004] Furthermore, at primary care locations (e.g., emergency locations such as ambulances) where a patient's brain dysfunction is first assessed, emergency personnel often perform tests to determine LVO in order to detect occlusion of major cerebral arteries.

[0005] Special Publication No. 2020-524064

[0006] However, the order of the tests to be performed for LVO diagnosis is predetermined. For example, if a patient with right hemisphere damage is first tested for a functional disorder that occurs less frequently in right hemisphere damage, unnecessary testing will delay the diagnosis of severe disease, which may worsen the patient's prognosis.

[0007] The present disclosure has been made in consideration of the above circumstances, and aims to provide a computer program, a brain dysfunction assessment support device, and a brain dysfunction assessment support method that can shorten the time required to assess brain dysfunction.

[0008] (1) A computer program according to the present disclosure causes a computer to execute the following processing: acquire dominant hemisphere information of a subject's dominant hemisphere that controls language; acquire abnormality information due to brain dysfunction of the subject; determine whether the subject's abnormality is due to a disorder of a function controlled by only one cerebral hemisphere based on the acquired abnormality information; and, if it is determined that the abnormality is due to a disorder of a function controlled by only one cerebral hemisphere, select a required test item from multiple test items of a brain dysfunction assessment test based on the acquired dominant hemisphere information. Here, an embodiment of the present disclosure is as follows: (2) The computer program according to (1) above causes a computer to execute the following processing: determine that the abnormality is due to a disorder of a function controlled by only one cerebral hemisphere, and select a test item related to language if the hemisphere controlling the impaired function is the dominant hemisphere that controls language. (3) The computer program of (1) or (2) above determines that the abnormality is due to a disorder of a function controlled by only one cerebral hemisphere, and if the hemisphere controlling the impaired function is not the dominant hemisphere controlling language, causes the computer to execute a process of selecting test items other than language-related test items. (4) Any one of the computer programs of (1) to (3) above, if it determines that the abnormality is due to a disorder of a function not controlled by only one cerebral hemisphere, causes the computer to execute a process of determining whether the impaired hemisphere is left-right based on the brain dysfunction assessment test and the dominant hemisphere information. (5) Any one of the computer programs of (1) to (4) above, if it determines that the abnormality is due to a disorder of a function not controlled by only one cerebral hemisphere, causes the computer to execute a process of selecting test items from the brain dysfunction assessment test so that language-related test items are administered before spatial recognition test items. (6) Any one of the computer programs (1) to (5) above causes a computer to execute a process of identifying the dominant hemisphere responsible for language based on diagnostic data or treatment data from the subject's past brain dysfunction, and storing the dominant hemisphere information of the identified dominant hemisphere.(7) Any one of the computer programs described in (1) to (6) above causes a computer to execute a process of identifying a dominant hemisphere responsible for language based on information about the subject's handedness and storing information about the identified dominant hemisphere. (8) In the computer program described in (7) above, the information about the dominant hand is estimated based on an image of the subject acquired. (9) Any one of the computer programs described in (1) to (8) above causes a computer to execute a process of acquiring an image of the subject and acquiring test results of a brain dysfunction assessment test for the subject, and, when the image of the subject and the test results of the brain dysfunction assessment test for the subject are input, inputting the acquired image and test results into a first learning model that outputs information about the dominant hemisphere responsible for language of the subject, and storing the information about the dominant hemisphere responsible for language output by the first learning model. (10) Any one of the computer programs described in (1) to (9) above causes a computer to execute a process of acquiring an image of the subject, determining abnormality information due to brain dysfunction of the subject based on the acquired image, and acquiring the determined abnormality information. (11) Any one of the computer programs described in (1) to (10) above causes a computer to execute a process of accepting input of abnormality information of the subject and acquiring the accepted abnormality information. (12) Any one of the computer programs described in (1) to (11) above causes a computer to execute a process of acquiring an image of the subject, inputting the acquired image into a second learning model that, when the image is input, outputs the presence or absence of an abnormality due to dysfunction of a function controlled only by one cerebral hemisphere, and acquiring the presence or absence of an abnormality due to dysfunction of a function controlled only by one cerebral hemisphere output by the second learning model. (13) Any one of the computer programs described in (1) to (12) above causes a computer to execute a process of selecting a notification means according to the severity of the brain dysfunction assessment result obtained by executing the selected test item.(14) A brain dysfunction assessment support device according to the present disclosure includes a control unit, which acquires dominant hemisphere information of a subject's dominant hemisphere that controls language, acquires abnormality information due to the subject's brain dysfunction, and determines whether the subject's abnormality is due to a disorder of a function controlled by only one cerebral hemisphere based on the acquired abnormality information, and if it is determined that the abnormality is due to a disorder of a function controlled by only one cerebral hemisphere, selects a required test item from among multiple test items in a brain dysfunction assessment test based on the acquired dominant hemisphere information. (15) A brain dysfunction assessment support method according to the present disclosure acquires dominant hemisphere information of a subject's dominant hemisphere that controls language, acquires abnormality information due to the subject's brain dysfunction, and determines whether the subject's abnormality is due to a disorder of a function controlled by only one cerebral hemisphere based on the acquired abnormality information, and if it is determined that the abnormality is due to a disorder of a function controlled by only one cerebral hemisphere, selects a required test item from among multiple test items in a brain dysfunction assessment test based on the acquired dominant hemisphere information.

[0009] According to the present disclosure, the time required to determine brain dysfunction can be shortened.

[0010] 1 is a diagram showing an example of the configuration of a brain dysfunction assessment support system according to this embodiment; FIG. 1 is a diagram showing an example of the functions of the brain's dominant and inferior hemispheres; FIG. 2 is a diagram showing an example of the procedure for identifying the dominant hemisphere responsible for language; FIG. 3 is a diagram showing an example of a method for identifying dominant hemisphere information of the dominant hemisphere responsible for language using a first learning model; FIG. 4 is a diagram showing an example of a method for identifying dominant hemisphere information of the dominant hemisphere responsible for language using a first learning model; FIG. 5 is a diagram showing an example of abnormality information due to brain dysfunction; FIG. 6 is a diagram showing an example of the processing procedure for selecting test items for a brain dysfunction assessment test using a brain dysfunction assessment support device; FIG. 7 is a diagram showing an example of a method for performing LVO assessment; FIG. 8 is a diagram showing an example of a method for performing LVO assessment; FIG. 9 is a diagram showing advantages based on the selection of test items for a brain dysfunction assessment test when an abnormality detected by the brain dysfunction assessment support device is due to a disorder in a function controlled by only one cerebral hemisphere; FIG. 10 is a diagram showing an example of a method for determining whether an abnormality detected by a second learning model is due to a disorder in a function controlled by only one cerebral hemisphere; 10A and 10B are diagrams illustrating merits based on the determination of the affected hemisphere by the cerebral dysfunction determination support device. 10B are diagrams illustrating examples of notifications according to the cerebral dysfunction determination results by the cerebral dysfunction determination support device.

[0011] An embodiment of the present disclosure will be described below. Fig. 1 is a diagram showing an example of the configuration of a brain dysfunction assessment support system according to this embodiment. The brain dysfunction assessment support system includes a brain dysfunction assessment support device 50, a camera 10, a first terminal device 20, a third terminal device 30, and a second terminal device 40. The camera 10, the first terminal device 20, the third terminal device 30, and the second terminal device 40 are connected to the brain dysfunction assessment support device 50 via a communication network 1. A dominant hemisphere information DB (database) 60 and a brain dysfunction assessment test DB (database) 70 are connected to the brain dysfunction assessment support device 50, and the brain dysfunction assessment support device 50 can access information stored in the dominant hemisphere information DB 60 and the brain dysfunction assessment test DB 70.

[0012] The camera 10 detects a motor function disorder (an abnormal state due to a brain dysfunction) of the subject based on an image obtained by photographing the subject, and outputs the detected disorder as abnormality information to the brain dysfunction assessment support device 50. The image obtained by the camera 10 may be a video or a still image. The image obtained by the camera 10 may include audio corresponding to the image acquired by a microphone (not shown). The microphone (not shown) may only acquire audio. There may be multiple cameras 10.

[0013] The first terminal device 20 can be configured as a personal computer, tablet terminal, smartphone, etc., and is a device used by the subject, the subject's family, or a third party. The first terminal device 20 accepts input of the subject's motor function, sensory, or visual field impairments (abnormal conditions due to brain dysfunction) and outputs the accepted impairments as abnormality information to the brain dysfunction assessment support device 50. The impairments can be input by text input or voice input. There may be multiple first terminal devices 20. By having the subject, the subject's family, or a third party input the abnormal conditions due to brain dysfunction, it becomes possible to input sensory or visual field impairments (abnormal conditions) that cannot be detected by the camera 10.

[0014] Disabilities detected by the camera 10 and the first terminal device 20 include motor function abnormalities, sensory abnormalities, visual field abnormalities, etc. Motor function abnormalities include, for example, quadriplegia on one side, facial paralysis on one side, entanglement of one leg, and loss of strength in one limb. Sensory abnormalities include, for example, not noticing touch on one side, not feeling pain on one side, and numbness on one side. Visual field abnormalities include loss of vision on one side, etc.

[0015] The third terminal device 30 includes a camera 31, a microphone, and a speaker (not shown), and may be configured as a smart display, a personal computer, a tablet device, a smartphone, or the like. The third terminal device 30 outputs test instructions to the subject to perform the LVO assessment test, including the test items selected by the brain dysfunction assessment support device 50. The test instructions may be displayed on a display screen and output as audio. The output of test instructions includes both "instruction output that allows the subject to perform the test according to the instructions" and "instruction output that allows a third party to view the instructions and present the test to the subject." The display screen of the third terminal device 30 may be configured as a touch panel or the like, allowing for operation of displayed icons, cursor movement, cursor operation, character input, etc. The third terminal device 30 can accept input of the subject's motor function, sensory, and visual field impairments and output the accepted impairments as abnormality information to the brain dysfunction assessment support device 50. The impairments may be input via a touch panel, a mouse, or a keyboard. The camera 31 captures the subject's test results by photographing them. The third terminal device 30 outputs the test results acquired by the camera 31 to the brain dysfunction assessment support device 50. A microphone (not shown) acquires the test results of the subject by acquiring the sound (speech) emitted by the subject. The microphone may also acquire sound resulting from the subject's actions, such as falling or dropping an object.

[0016] The second terminal device 40 can be configured as a personal computer, tablet terminal, smartphone, or the like. The subject's family or a third party observes the actions, etc., performed by the subject in accordance with the test instructions output by the third terminal device 30. The subject's family or a third party inputs the test results performed by the subject, and the second terminal device 40 accepts the test results and outputs the accepted test results to the brain dysfunction assessment support device 50.

[0017] The brain dysfunction assessment support device 50 includes a control unit 51 that controls the entire device, a communication unit 52, a memory 53, a notification unit 54, a storage unit 55, and a recording medium reading unit 59.

[0018] The control unit 51 may be configured by incorporating a required number of central processing units (CPUs), micro-processing units (MPUs), graphics processing units (GPUs), etc. The control unit 51 may also be configured by combining digital signal processors (DSPs), field-programmable gate arrays (FPGAs), etc.

[0019] The communication unit 52 includes a communication module and has the function of communicating with the camera 10 , the first terminal device 20 , the third terminal device 30 , and the second terminal device 40 via the communication network 1 .

[0020] The notification unit 54 makes the required notification using a notification means selected according to the severity of the brain dysfunction assessment result obtained by executing the test items selected by the brain dysfunction assessment support device 50. If the brain dysfunction is severe, the notification unit 54 makes an emergency call to arrange for an ambulance and to notify a secondary medical location (such as an emergency medical center) where a specialist is located, if the brain dysfunction is moderate to severe, it notifies a hospital or clinic, and if the brain dysfunction is mild, it can notify the subject's family, etc.

[0021] The storage unit 55 can be configured with a semiconductor memory, a hard disk, or the like, and stores a computer program 56 (program product), a first learning model 57, a second learning model 58, and required information. Details of the first learning model 57 and the second learning model 58 will be described later.

[0022] The computer program 56 can be stored in the storage unit 55 by reading the computer program 56 recorded on a storage medium (for example, an optically readable disk storage medium such as a CD-ROM) M using the storage medium reading unit 59. The computer program 56 may also be downloaded from an external device via the communication unit 52 and stored in the storage unit 55.

[0023] The memory 53 can be configured with semiconductor memory such as SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), flash memory, etc. A computer program 56 can be loaded into the memory 53, and the control unit 51 can execute the computer program 56. The control unit 51 can execute processing defined by the computer program 56. In other words, processing by the control unit 51 is also processing by the computer program 56.

[0024] The dominant hemisphere information DB 60 stores, for each subject (including a patient), dominant hemisphere information about the dominant hemisphere that controls the subject's language. When the dominant hemisphere information for multiple subjects is stored in the dominant hemisphere information DB 60, the dominant hemisphere information may be linked to the subject's facial image, name, etc., and registered to identify the subject appearing in a captured image.

[0025] FIG. 2 is a diagram showing an example of the functions of the left and right hemispheres of the brain. The left hemisphere is the hemisphere that controls the motor functions, sensations, visual field, etc. of the right side of the body, and the right hemisphere is the hemisphere that controls the motor functions, sensations, visual field, etc. of the left side of the body. In this example, the left hemisphere is the dominant hemisphere, and therefore the left hemisphere controls language functions. FIG. 2 shows an example of the functions of the left and right hemispheres, and in some cases the right hemisphere controls language functions. In other words, which hemisphere controls what functions may vary from person to person. The dominant hemisphere is the hemisphere that mainly controls language functions, and the inferior hemisphere is the hemisphere that does not mainly control language. For many people, including most right-handed people and approximately two-thirds of left-handed people, the left hemisphere is dominant.

[0026] Next, a method for collecting the dominant hemisphere information to be stored in the dominant hemisphere information DB 60 will be described.

[0027] The control unit 51 can identify the dominant hemisphere that controls language based on diagnostic data or treatment data from the subject's past brain dysfunction, and store dominant hemisphere information about the identified dominant hemisphere in the dominant hemisphere information DB 60. The control unit 51 can also identify the dominant hemisphere that controls language based on information about the subject's dominant hand, and store dominant hemisphere information about the identified dominant hemisphere in the dominant hemisphere information DB 60.

[0028] 3 is a diagram showing an example of the procedure for processing to identify the dominant hemisphere responsible for language. The processing shown in FIG. 3 can be realized by a computer program 56. For convenience, the processing will be described below as being performed by the control unit 51. The control unit 51 determines whether it is possible to identify the dominant hemisphere responsible for language from information about the patient's (subject's) previous brain dysfunction (S11). Information about the patient's previous brain dysfunction may be collected in advance and stored in the storage unit 55 or in an external data server (not shown).

[0029] If it is not possible to identify the dominant hemisphere responsible for language (NO in S11), the control unit 51 determines whether the patient is right-handed (S12). The determination of whether the patient is right-handed can be made by outputting a question (whether the patient is left-handed or right-handed) to the patient or their family member, etc., via a terminal device or display device, using voice or text, and receiving a response to the question. Alternatively, the determination of whether the patient is right-handed, i.e., whether the patient is left-handed or right-handed (handedness information), can be made using a machine learning model, or by recognizing an image of the patient.

[0030] If the patient's dominant hand is right (YES in S12), the control unit 51 determines that the dominant hemisphere responsible for language is the left hemisphere (S13), stores dominant hemisphere information indicating that the dominant hemisphere responsible for language is the left hemisphere in the dominant hemisphere information DB 60 (S15), and terminates the processing.

[0031] If the patient's dominant hand is not right, i.e., if the patient's dominant hand is left (NO in S12), the control unit 51 determines that the dominant hemisphere responsible for language is unknown (S14) and ends the process. If the dominant hemisphere responsible for language can be identified (YES in S11), the control unit 51 stores dominant hemisphere information indicating that the dominant hemisphere responsible for language is the left hemisphere or the right hemisphere in the dominant hemisphere information DB 60 (S15), and ends the process.

[0032] As described above, if the dominant hemisphere responsible for language can be identified through fMRI or other methods used during diagnosis or treatment of a past brain dysfunction, or if a functional disorder with a bilateral difference occurred during a past brain dysfunction and it is known which hemisphere the disorder occurred in, this information can be used to store in advance dominant hemisphere information indicating which hemisphere is dominant for language.

[0033] 4A and 4B are diagrams illustrating an example of a method for identifying dominant hemisphere information of the dominant hemisphere responsible for language using the first learning model 57. The process illustrated in FIG. 4 can be performed when a subject (patient) is suffering from a brain dysfunction. In the example of FIG. 4A, images of the patient (subject), test results of a brain dysfunction assessment test for the patient, and brain images taken at a medical institution to which the patient is transported are input to the learning model 57. The learning model 57 outputs dominant hemisphere information of the dominant hemisphere responsible for language. Brain dysfunction assessment tests include, for example, LVO (Large Vessel Occlusion) assessment methods and screen methods such as ELVO, RACE, FAST-ED, and VAN. That is, the control unit 51 acquires images of the subject. The images may be video or still images. The images may be acquired using cameras 10, 31, etc. The control unit 51 acquires test results of the brain dysfunction assessment test for the subject. The control unit 51 also acquires brain images of the subject. The control unit 51 inputs the acquired images of the subject, the test results of the brain dysfunction assessment test, and the brain images of the subject into a first learning model 57 that outputs dominant hemisphere information of the dominant hemisphere that controls language when images of the subject, the test results of the brain dysfunction assessment test, and the brain images of the subject are input, and can store the dominant hemisphere information output by the first learning model 57 in a dominant hemisphere information DB 60. Note that the images of the patient (subject) include images of the patient when they are suffering from a brain dysfunction and images of the patient when they are not suffering from a brain dysfunction.

[0034] In the example of FIG. 4B , images of a patient (subject) and test results of a brain dysfunction assessment test for the patient are input to the learning model 57. The learning model 57 outputs dominant hemisphere information about the dominant hemisphere that controls language. For example, when hemibody disability information (right or left paralysis) such as upper limb paralysis, lower limb paralysis, or facial paralysis is identified from an image of the patient and a language disorder such as aphasia is identified from the test results, the learning model 57 is able to output dominant hemisphere information. That is, the control unit 51 acquires images of the subject. The images may be video or still images. The images may be acquired using cameras 10, 31, etc. The control unit 51 acquires test results of a brain dysfunction assessment test for the subject. The control unit 51 inputs the acquired images of the subject and the test results of the brain dysfunction assessment test to a first learning model 57, which outputs dominant hemisphere information of the dominant hemisphere that controls language when the images of the subject and the test results of the brain dysfunction assessment test are input, and can store the dominant hemisphere information output by the first learning model 57 in a dominant hemisphere information DB 60. Furthermore, if the subject's handedness information is estimated, for example, from an image of a patient, the learning model 57 can output dominant hemisphere information. That is, the control unit 51 acquires images of the subject. The images may be video or still images. The images may be acquired using cameras 10, 31, etc. The control unit 51 inputs the acquired image of the subject to the first learning model 57, which outputs dominant hemisphere information of the dominant hemisphere that controls language when an image of the subject is input, and can store the dominant hemisphere information based on information about the subject's handedness output by the first learning model 57 in the dominant hemisphere information DB 60. Note that the learning model 57 may output the handedness information itself.

[0035] When the dominant hemisphere information of a patient is identified using input data (including images of the patient, test results of a brain dysfunction assessment test, and brain images of the patient) at the time of the patient's first onset using the method shown in Figure 4A or 4B, the dominant hemisphere responsible for language will be known at the time of the patient's second or subsequent onset. Note that if the dominant hemisphere information of the patient is estimated using information about the patient's handedness, the dominant hemisphere responsible for language can be known even when the patient first develops a brain dysfunction.

[0036] The first learning model 57 is composed of, for example, a CNN (Convolutional Neural Network) or the like, and can use algorithms such as AlexNet, GoogleNet, Residual Networks (ResNet), and ConvNeXt. Training data for generating (learning) the first learning model 57 can be collected as follows. In the case of FIG. 4A , images of many (multiple) patients, test results of brain dysfunction assessment tests of the many patients, brain images of the many patients, and dominant hemisphere information indicating which hemisphere is dominant for language in the many patients are collected as training data. Then, when the images of the patients, the test results of the patients, and the brain images of the patients are input into the first learning model 57, learning is repeated until the dominant hemisphere information output by the first learning model 57 matches the dominant hemisphere information as training data. The generated first learning model 57 is stored in the memory unit 55. The first learning model 57 may be generated by the brain dysfunction assessment support device 50 or another learning processing device.

[0037] In the case of FIG. 4B , images of many patients, test results of brain dysfunction assessment tests of those patients, and dominant hemisphere information indicating which hemisphere is the left or right hemisphere that controls language of those patients are collected as training data. Note that instead of test results, information on whether a patient is left-handed or right-handed (e.g., a question (whether the patient is left-handed or right-handed) is output to the patient or their family via a terminal device or display device, and the answer to the question is provided) may be collected as training data. Then, when the images of the patient and the test results of that patient are input to the first learning model 57, learning is repeated until the dominant hemisphere information output by the first learning model 57 matches the dominant hemisphere information as training data. The generated first learning model 57 is stored in the memory unit 55. Note that the first learning model 57 may be generated by the brain dysfunction assessment support device 50 or another learning processing device. It should be noted that the images taken of many patients include images of patients when they are suffering from brain dysfunction and images of patients when they are not suffering from brain dysfunction.

[0038] Next, abnormality information due to brain dysfunction of the subject will be described.

[0039] FIG. 5 shows an example of abnormal information due to brain dysfunction. Abnormal information (symptoms), which indicate an abnormal state due to brain dysfunction, includes impairments in the subject's motor function, sensation, and visual field, such as ocular deviation, aphasia, hemispatial neglect, facial paralysis, upper and lower limb paralysis, and dysarthria. Ocular deviation refers to a condition in which both eyes are turned to one side, or a condition in which a subject is forced to follow a finger but cannot turn it away. Aphasia refers to an inability to understand or speak words. Hemispatial neglect refers to a condition in which one side of the face is ignored, such as being unable to recognize two or three fingers even when four are shown. Facial paralysis refers to a distortion of the face or a condition in which one side does not move as well as the other. Upper and lower limb paralysis refers to a condition in which one arm or leg cannot move, or is less mobile than the other side, or is numb, or is unable to maintain a stable posture. Dysarthria refers to slurred speech and unclear speech.

[0040] Regarding whether abnormalities (symptoms) are due to impairment of functions controlled only by one cerebral hemisphere, aphasia, hemispatial neglect, facial paralysis, and upper and lower limb paralysis are caused by impairment of functions controlled only by one cerebral hemisphere, while ocular deviation and dysarthria are caused by impairment of functions not controlled only by one cerebral hemisphere. Furthermore, ocular deviation, aphasia, and hemispatial neglect are often caused by severe brain dysfunction, while facial paralysis, upper and lower limb paralysis, and dysarthria are often caused by moderate to mild brain dysfunction.

[0041] The control unit 51 can acquire images of the subject from the cameras 10 and 31, determine abnormality information due to brain dysfunction of the subject based on the acquired images, and acquire the determined abnormality information. The control unit 51 can also accept input of abnormality information of the subject from the first terminal device 20 and acquire the accepted abnormality information.

[0042] The brain dysfunction assessment test DB70 stores the contents of brain dysfunction assessment tests. The brain dysfunction assessment tests include a method for assessing LVO (Large Vessel Occlusion). The assessment method specifies the multiple test items to be performed and the order in which the test items are performed. The assessment methods include, for example, screening methods such as ELVO, RACE, FAST-ED, and VAN. In the ELVO screen, the test items are specified to be performed in the following order: ocular deviation, aphasia, and hemispatial neglect. In the ELVO screen, the ocular deviation test checks whether the subject has an eye deviation. In the aphasia test, the subject is shown an object such as glasses or a watch and asked, "What is this?" to check whether the subject can name the object. In the hemispatial neglect test, the subject is shown four fingers and asked how many fingers they have, and the subject's answer is checked. Note that the hemispatial neglect test is not limited to the above. For example, if the dominant hemisphere responsible for language has been identified, it is possible to determine whether or not there is hemispatial neglect due to damage to the inferior hemisphere by checking whether or not the visual field controlled by the hemisphere opposite to the dominant hemisphere responsible for language (the inferior hemisphere) is narrowed.

[0043] Fig. 6 is a diagram showing an example of the processing procedure for selecting test items for a brain dysfunction assessment test by the brain dysfunction assessment support device 50. The processing shown in Fig. 6 shows processing when an abnormality is due to a disorder of a function controlled only by one cerebral hemisphere, and can be realized by a computer program 56. The control unit 51 acquires abnormality information of the patient (subject) (S21) and determines whether or not ocular deviation is present as a brain dysfunction (S22). If ocular deviation is not present (NO in S22), the control unit 51 determines whether or not the abnormality is due to a disorder of a function controlled only by one cerebral hemisphere (S23).

[0044] If the abnormality is due to a disorder in a function controlled by only one cerebral hemisphere (YES in S23), the control unit 51 determines whether or not the disorder is in a function controlled by the left hemisphere (S24), and if the disorder is in a function controlled by the left hemisphere (YES in S24), it determines that the disorder is in the left hemisphere (S25). The control unit 51 determines whether or not the dominant hemisphere controlling language is the left hemisphere (S26). The determination of whether or not the dominant hemisphere controlling language is the left hemisphere can be made based on the dominant hemisphere information of the patient stored in the dominant hemisphere information DB 60. When selecting the patient's dominant hemisphere information from the dominant hemisphere information of multiple subjects registered in the dominant hemisphere information DB 60, the facial image of the patient in the captured image is compared with a facial image linked to the dominant hemisphere information registered in the dominant hemisphere information DB 60, or the subject, a family member of the subject, or a third party inputs the facial image or name of the patient who will undergo the brain dysfunction assessment test using the second terminal device 40. Note that determining whether the dominant hemisphere controlling language is the left hemisphere in step S26 is equivalent to determining whether the dominant hemisphere controlling language is the right hemisphere.

[0045] If the dominant hemisphere responsible for language is the left hemisphere (YES in S26), the control unit 51 selects an aphasia test (S27), performs an LVO assessment (S28), and ends the process. The LVO assessment in step S28 determines the presence or absence of cerebral artery occlusion by conducting a cerebral dysfunction assessment test using the selected test items. If the dominant hemisphere responsible for language is not the left hemisphere, i.e., if the dominant hemisphere responsible for language is the right hemisphere (NO in S26), the control unit 51 performs the process of step S31, which will be described later.

[0046] If the disorder is not a disorder of a function controlled by the left hemisphere, i.e., if the disorder is a disorder of a function controlled by the right hemisphere (NO in S24), the control unit 51 determines that the disorder is a disorder of the right hemisphere (S29). The control unit 51 determines whether the dominant hemisphere controlling language is the right hemisphere (S30). Note that determining whether the dominant hemisphere controlling language is the right hemisphere in step S30 is equivalent to determining whether the dominant hemisphere controlling language is the left hemisphere.

[0047] If the dominant hemisphere responsible for language is the right hemisphere (YES in S30), the control unit 51 performs the process of step S27. If the dominant hemisphere responsible for language is not the right hemisphere, i.e., if the dominant hemisphere responsible for language is the left hemisphere (NO in S30), the control unit 51 selects a test for hemispatial neglect (S31) and performs the process of step S28. If there is eye deviation (YES in S22), the control unit 51 performs the process of step S28. If the abnormality is due to a disorder of a function not exclusively controlled by one cerebral hemisphere (NO in S23), the control unit 51 ends the process. In this case, the control unit 51 can perform the process for the case where the abnormality is due to a disorder of a function not exclusively controlled by one cerebral hemisphere, as described below.

[0048] 7A and 7B are diagrams illustrating an example of a method for performing LVO assessment. In FIG. 7A, under the control of the control unit 51 of the brain dysfunction assessment support device 50, test instructions for a brain dysfunction assessment test with selected test items are displayed on the screen of the third terminal device 30 or output as audio, and the test results are acquired by the camera 31. The third terminal device 30 then outputs the test results to the brain dysfunction assessment support device 50. The test results may also be acquired by the camera 10 or a microphone (not shown). By displaying the test instructions on the screen of the third terminal device 30 or outputting them as audio, LVO assessment can be performed even when there is no person to assist the subject. In this example, the test instructions may be, for example, an illustration of four fingers displayed on the screen of the third terminal device 30 and the question "How many fingers do you see?" displayed on the screen of the third terminal device 30 or output as audio.

[0049] 7B shows that, under the control of the control unit 51 of the brain dysfunction assessment support device 50, test instructions for the selected brain dysfunction assessment test are displayed on the screen of the third terminal device 30 or output by voice. In this case, a person assisting the subject may provide the test instructions to the subject. The person assisting the subject inputs the test results into the third terminal device 30 or the second terminal device 40, and the third terminal device 30 or the second terminal device 40 outputs the test results to the brain dysfunction assessment support device 50. In this example, the test instructions may be, for example, an instruction displayed on the screen of the third terminal device 30, such as "Press four fingers in front of the subject and ask how many fingers they see."

[0050] 8 is a diagram showing the merits of selecting test items for a brain dysfunction assessment test using the brain dysfunction assessment support device 50 when the abnormality is due to a disorder of a function controlled by only one cerebral hemisphere. The comparative example shows a case where it is known whether the impaired hemisphere is left or right, but it is unclear whether the disorder is in the dominant hemisphere that controls language. In the comparative example, each test item in the brain dysfunction assessment test is performed in a predetermined order. For example, the order of the ELVO screening test is ocular deviation (conjugate deviation), aphasia, and unilateral spatial neglect.

[0051] In the comparative example, it is unclear which hemisphere is dominant in language. If the affected hemisphere is the one not responsible for language (usually the right hemisphere), and questions about aphasia are administered before questions about hemispatial neglect, tests of functions specific to the unaffected hemisphere are administered first, resulting in a delay in determining the severity of the condition due to unnecessary testing. This may worsen the patient's prognosis.

[0052] On the other hand, in this embodiment, it is possible to identify whether the affected hemisphere is the dominant hemisphere that controls language. For example, in the case of a hemisphere that controls language, a test for aphasia, which is a test of functions specific to the affected hemisphere, can be selected, and the test for hemispatial neglect can be deleted. Also, in the case of a hemisphere that does not control language, a test for hemispatial neglect, which is a test of functions specific to the affected hemisphere, can be selected, and the test for aphasia can be deleted.

[0053] As described above, the control unit 51 acquires dominant hemisphere information about the subject's dominant hemisphere, which controls language, from the dominant hemisphere information DB 60, and acquires abnormality information due to the subject's brain dysfunction from the camera 10, the third terminal device 30, or the first terminal device 20. Based on the acquired abnormality information, the control unit 51 determines whether the subject's abnormality is due to a dysfunction of a function controlled only by one cerebral hemisphere. For example, as shown in FIG. 5 , if the symptom (abnormality information) is unilateral facial paralysis, it can be determined that the abnormality is due to a dysfunction of a function controlled only by one cerebral hemisphere. If the control unit 51 determines that the abnormality is due to a dysfunction of a function controlled only by one cerebral hemisphere, it selects a required test item from multiple test items of the brain dysfunction assessment test based on the acquired dominant hemisphere information.

[0054] With the above-mentioned configuration, the time required to diagnose brain dysfunction can be shortened by prioritizing the implementation of necessary test items and deleting unnecessary test items. Note that the deleted test items may be implemented after the implementation of the necessary test items.

[0055] Furthermore, if the control unit 51 determines that the abnormality is due to a disorder of a function controlled only by one cerebral hemisphere, and if the hemisphere to which the disordered function is distributed is the dominant hemisphere that controls language, it can select test items related to language. For example, as shown in Figure 8, in the case of a disorder of the hemisphere that controls language, it selects an aphasia test, which is a test of a function specific to the hemisphere in which the disorder occurred, so that necessary test items can be performed with priority.

[0056] Furthermore, when the control unit 51 determines that the abnormality is due to a disorder of a function controlled only by one cerebral hemisphere and the hemisphere in which the disordered function is distributed is not the dominant hemisphere controlling language, the control unit 51 can select test items other than those related to language. For example, as illustrated in Figure 8, in the case of a disorder of a hemisphere not controlling language, a test of hemispatial neglect, which is a test of a function specific to the hemisphere in which the disorder occurred, is selected, so that necessary test items can be performed with priority.

[0057] FIG. 9 is a diagram showing an example of a method for determining whether an abnormality detected by the second learning model 58 is due to a disorder in a function controlled only by one cerebral hemisphere. The process shown in FIG. 9 can be performed when the subject (patient) has a brain dysfunction. The control unit 51 can perform the process when the subject (patient) has a brain dysfunction. The control unit 51 acquires an image of the subject. The image may be a video or a still image. The image may be acquired using a camera 10, 31, or the like. When an image is input, the control unit 51 inputs the acquired image to the second learning model 58, which outputs the presence or absence of an abnormality due to a disorder in a function controlled only by one cerebral hemisphere, and can obtain the presence or absence of an abnormality due to a disorder in a function controlled only by one cerebral hemisphere output by the second learning model 58.

[0058] Like the first learning model 57, the second learning model 58 is configured with a CNN (Convolutional Neural Network) or the like, and can use algorithms such as AlexNet, GoogleNet, ResNet (Residual Networks), and ConvNeXt. Training data for generating (learning) the second learning model 58 can be collected as follows. Images of a subject (patient) in a state where no new brain dysfunction has occurred and information indicating the presence or absence of abnormalities due to dysfunction of functions controlled only by one cerebral hemisphere of the patient are collected as training data. A subject in a state where no new brain dysfunction has occurred is a subject in a state where no new abnormalities due to dysfunction of functions controlled only by one cerebral hemisphere have occurred at the time of image capture. Note that in the case of a subject who still has residual symptoms from a previous brain dysfunction, the subject in that state may also be defined as a subject in a state where no new brain dysfunction has occurred. Then, when an image of a patient is input to the second learning model 58, learning is repeated until the information output by the second learning model 58 indicating the presence or absence of an abnormality due to a disorder of a function controlled only by one cerebral hemisphere matches the information indicating the presence or absence of an abnormality due to a disorder of a function controlled only by one cerebral hemisphere as training data. The generated second learning model 58 is stored in the memory unit 55. The second learning model 58 may be generated by the cerebral dysfunction assessment support device 50 or by another learning processing device. Furthermore, images of many (multiple) patients who have not yet developed a new cerebral dysfunction and information on whether the abnormality in the patient is due to a disorder of a function with a difference in the distribution of cerebral function between the left and right sides may be used as training data.

[0059] 10 is a diagram showing an example of a processing procedure for determining the affected hemisphere of a cerebral dysfunction by the cerebral dysfunction determination support device 50. The processing shown in FIG. 10 illustrates a case where the abnormality is due to a dysfunction of a function not exclusively controlled by one cerebral hemisphere, and can be realized by a computer program 56. The control unit 51 acquires abnormality information of the patient (subject) (S41) and determines whether or not there is eye deviation as a cerebral dysfunction (S42). If there is no eye deviation (NO in S42), the control unit 51 determines whether or not the abnormality is due to a dysfunction of a function not exclusively controlled by one cerebral hemisphere (S43).

[0060] If the disorder is due to a function not controlled by only one cerebral hemisphere (YES in S43), the control unit 51 selects test items so that the aphasia test is conducted first and the hemispatial neglect test is conducted later (S44), and performs LVO assessment (S45). The LVO assessment is the same as in the case of FIG. 6.

[0061] The control unit 51 determines whether the affected area is the hemisphere that controls language based on the test results (S46), and if the affected area is the hemisphere that controls language (YES in S46), determines whether the dominant hemisphere that controls language is the left hemisphere based on the dominant hemisphere information DB 60 (S47). If the dominant hemisphere that controls language is the left hemisphere (YES in S47), the control unit 51 determines that the left hemisphere is impaired (S48), and ends the process.

[0062] If the dominant hemisphere responsible for language is not the left hemisphere, i.e., if the dominant hemisphere responsible for language is the right hemisphere (NO in S47), the control unit 51 determines that the right hemisphere is impaired (S49) and ends the process. If the impaired area is not the hemisphere responsible for language (NO in S46), the control unit 51 determines whether the dominant hemisphere responsible for language is the left hemisphere (S50). If the dominant hemisphere responsible for language is the left hemisphere (YES in S50), the control unit 51 determines that the right hemisphere is impaired (S51) and ends the process.

[0063] If the dominant hemisphere responsible for language is not the left hemisphere, i.e., if the dominant hemisphere responsible for language is the right hemisphere (NO in S50), the control unit 51 determines that there is a disorder of the left hemisphere (S52) and ends the process. If the abnormality is due to a disorder of a function controlled by only one cerebral hemisphere (NO in S43), the control unit 51 ends the process. If there is eye deviation (YES in S42), the control unit 51 makes an LVO determination (S53) and ends the process.

[0064] FIG. 11 is a diagram showing the benefits of determining the affected hemisphere using the brain dysfunction determination support device 50. The comparative example shows a case where it is known whether the affected hemisphere is the dominant hemisphere, but it is unclear which hemisphere it is. In the comparative example, each test item in the brain dysfunction determination test is performed in a predetermined order. For example, the order of the ELVO screening test is ocular deviation (conjugate deviation), aphasia, and unilateral spatial neglect. From the test results, it is known whether the affected hemisphere is the dominant hemisphere that controls language, but it remains unclear which hemisphere is affected.

[0065] On the other hand, in this embodiment, dominant hemisphere information indicating whether the left or right hemisphere is the dominant hemisphere responsible for language is stored in advance, so by conducting an aphasia test and a hemispatial neglect test, it is possible to determine whether the affected hemisphere is the dominant hemisphere responsible for language, and as a result, it is possible to identify whether the affected hemisphere is left or right.

[0066] As described above, when the control unit 51 determines that the abnormality is due to a disorder of a function that is not exclusively controlled by one cerebral hemisphere, it can select test items in the brain dysfunction assessment test so that test items related to language are administered before test items related to spatial recognition. Furthermore, when the control unit 51 determines that the abnormality is due to a disorder of a function that is not exclusively controlled by one cerebral hemisphere, it can determine whether the impaired hemisphere is left or right based on the brain dysfunction assessment test and dominant hemisphere information.

[0067] The above-described configuration allows the doctor to determine whether the affected hemisphere is the left or right hemisphere, which can be used as reference information when making a diagnosis, thereby reducing the risk of delays in starting treatment.

[0068] FIG. 12 shows an example of a notification based on the brain dysfunction assessment result obtained by the brain dysfunction assessment support device 50. As shown in FIG. 12, the brain dysfunction assessment support device 50 can provide the required notification using a notification means selected according to the severity of the brain dysfunction assessment result obtained by executing the selected test items. When the control unit 51 determines via the notification unit 54 that the brain dysfunction is severe based on symptoms such as eye deviation, hemispatial neglect, or aphasia, it can issue an emergency call to dispatch an ambulance or notify a secondary medical facility (such as an emergency medical center) where a specialist is located. Furthermore, the control unit 51 can notify a hospital or clinic if the brain dysfunction is moderate to severe, such as unilateral facial paralysis, unilateral upper or lower limb paralysis, or speech disorder, and can notify the subject's family, etc., if the brain dysfunction is mild.

[0069] As described above, according to this embodiment, the hemisphere in which the brain dysfunction is occurring is identified based on the dominant hemisphere information indicating whether the hemisphere responsible for language is the left or right hemisphere acquired in advance and the patient's symptoms and behavior. Tests of the functions controlled by the hemisphere identified as being impaired are then prioritized, thereby shortening the time required to assess brain dysfunction (time required for LVO assessment). Furthermore, shortening the assessment time improves the patient's prognosis. Furthermore, since the required notification is sent using a notification method selected according to the severity of the brain dysfunction assessment result obtained by executing the selected test items, notification can be sent to the most appropriate notification destination, and an emergency call can be made only when truly necessary, thereby reducing the strain on medical care.

[0070] REFERENCE SIGNS LIST 1 Communication network 10 Camera 20 First terminal device 30 Third terminal device 31 Camera 40 Second terminal device 50 Brain dysfunction assessment support device 51 Control unit 52 Communication unit 53 Memory 54 Notification unit 55 Storage unit 56 Computer program 57 First learning model 58 Second learning model 59 Recording medium reading unit 60 Dominant hemisphere information DB 70 Brain dysfunction assessment test DB

Claims

1. A computer program that causes a computer to perform the following processes: acquire dominant hemisphere information about the subject's dominant hemisphere that controls language; acquire abnormality information due to brain dysfunction of the subject; determine whether the subject's abnormality is due to a disorder in a function controlled by only one cerebral hemisphere based on the acquired abnormality information; and if it is determined that the abnormality is due to a disorder in a function controlled by only one cerebral hemisphere, select a required test item from multiple test items in a brain dysfunction assessment test based on the acquired dominant hemisphere information.

2. The computer program according to claim 1, which causes a computer to execute the following process: determining that the abnormality is due to a disorder in a function controlled by only one cerebral hemisphere, and selecting test items related to language if the hemisphere controlling the disordered function is the dominant hemisphere controlling language.

3. The computer program according to claim 1, which causes a computer to execute the following process: determining that the abnormality is due to a disorder in a function controlled by only one cerebral hemisphere, and selecting test items other than those related to language if the hemisphere controlling the disordered function is not the dominant hemisphere controlling language.

4. A computer program as claimed in any one of claims 1 to 3, which causes a computer to execute a process to determine whether the affected hemisphere is left-handed or right-handed based on the brain function disorder assessment test and the dominant hemisphere information, if it is determined that the abnormality is due to a disorder of a function that is not controlled by only one cerebral hemisphere.

5. A computer program as claimed in any one of claims 1 to 3, which causes a computer to execute a process of selecting test items from the brain dysfunction assessment test so that test items relating to language are administered before test items relating to spatial recognition, if it is determined that the abnormality is due to a disorder of a function not controlled by only one cerebral hemisphere.

6. A computer program according to any one of claims 1 to 3, which causes a computer to execute the following process: identifying the dominant hemisphere responsible for language based on diagnostic data or treatment data from the subject's past brain dysfunction; and storing information about the identified dominant hemisphere.

7. A computer program according to any one of claims 1 to 3, which causes a computer to execute the following process: identifying the dominant hemisphere responsible for language based on information about the subject's handedness; and storing information about the identified dominant hemisphere.

8. The computer program according to claim 7, wherein the information on the dominant hand is estimated based on an image of the subject that has been captured.

9. A computer program as claimed in any one of claims 1 to 3, which causes a computer to execute the following processes: acquiring images of the subject; acquiring test results of a brain function disorder assessment test of the subject; inputting the acquired images and test results into a first learning model that, when the images of the subject and the test results of the brain function disorder assessment test of the subject are input, outputs dominant hemisphere information of the dominant hemisphere that controls language of the subject; and storing the dominant hemisphere information of the dominant hemisphere that controls language output by the first learning model.

10. A computer program as claimed in any one of claims 1 to 3, which causes a computer to execute the following processes: acquiring an image of the subject; determining abnormal information due to brain dysfunction of the subject based on the acquired image; and acquiring the determined abnormal information.

11. A computer program according to any one of claims 1 to 3, which causes a computer to execute the following process: accept input of abnormality information of the subject; and acquire the accepted abnormality information.

12. A computer program as claimed in any one of claims 1 to 3, which causes a computer to execute a process of acquiring an image of the subject, inputting the acquired image into a second learning model which, when inputted, outputs whether or not there is an abnormality due to a disorder of a function controlled only by one cerebral hemisphere, and obtaining the output of the second learning model indicating whether or not there is an abnormality due to a disorder of a function controlled only by one cerebral hemisphere.

13. A computer program as claimed in any one of claims 1 to 3, which causes a computer to execute a process of selecting a notification means according to the severity of the brain dysfunction assessment result obtained by executing the selected test item.

14. A brain dysfunction assessment support device comprising a control unit, which acquires dominant hemisphere information of the subject's dominant hemisphere that controls language, acquires abnormality information due to brain dysfunction of the subject, determines whether the subject's abnormality is due to a disorder in a function controlled by only one cerebral hemisphere based on the acquired abnormality information, and if it is determined that the abnormality is due to a disorder in a function controlled by only one cerebral hemisphere, selects a required test item from multiple test items in a brain dysfunction assessment test based on the acquired dominant hemisphere information.

15. A method for assisting in the assessment of brain dysfunction, comprising: acquiring dominant hemisphere information of the subject's dominant hemisphere that controls language; acquiring abnormality information due to brain dysfunction of the subject; determining whether the subject's abnormality is due to a disorder in a function controlled by only one cerebral hemisphere based on the acquired abnormality information; and, if it is determined that the abnormality is due to a disorder in a function controlled by only one cerebral hemisphere, selecting a required test item from multiple test items in a brain dysfunction assessment test based on the acquired dominant hemisphere information.

Citation Information

Patent Citations

  • Evaluation device, evaluation method, and evaluation program

    JP2020203014A

  • Task determination device during brain activity analysis

    JP2023509988A

  • Program, information processing device, information processing method, and information processing program

    WO2022054677A1