Screening system for mild cognitive impairment

Abstract

We provide a system to screen for the possibility of MCI (Mild Cognitive Impairment). [Solution] A system for screening for mild cognitive impairment, which enables the execution of a single test application that can be performed in place without the subject moving their position, based on images using virtual reality, augmented reality, or mixed reality. The test application presents the subject with a problem in which they sequentially select objects from among multiple numbers or multiple selection objects displayed in an image according to a given rule. The system acquires the test results of the test application based on the subject's operation or response, records the acquired test results in a recording unit, compares the test results with pre-stored criteria for determining the possibility of mild cognitive impairment, generates a judgment result regarding the possibility of mild cognitive impairment in the subject, displays the judgment result and test results to the user via an output unit, and provides the user with information to judge the subject's cognitive state by viewing the displayed information.

Description

Technical Field

[0001] The present invention relates to a system for screening persons who may have mild cognitive impairment (hereinafter referred to as "MCI").

Background Art

[0002] MCI refers to an intermediate stage between healthy persons and dementia, in a state where daily life can be carried out without problems, but cognitive function decline has occurred. If left untreated, the symptoms will progress, and there is a high possibility of transitioning to dementia in the future.

[0003] Although MCI is a pre-stage of dementia, it has recently been found that even if a person has MCI, there is a possibility of recovery to a healthy state by taking appropriate measures. Therefore, in order to prevent the progression from MCI to dementia, it is expected that by detecting MCI at an early stage and taking appropriate measures, the probability of progression to dementia can be reduced.

[0004] To diagnose MCI, diagnosis from various viewpoints by a doctor is required. However, as a pre-stage for receiving a doctor's diagnosis, if persons with a high possibility of having MCI can be screened (sorted out), it is expected that the process can be smoothly connected to a doctor's diagnosis. However, as conventional screening means, since neuropsychological tests such as MoCA-J and MMSE need to be carried out by medical professionals, etc., the opportunities for implementation have been few.

[0005] Patent Document 1 relates to an examination system for higher brain function disorders using mixed reality, and does not describe screening for MCI. Patent Document 2 relates to a rehabilitation system for improving higher brain function disorders (including cases caused by dementia), and does not describe screening for MCI. Patent Document 3 describes a system for rehabilitation using virtual reality, and states that it is also effective in preventing dementia (paragraph 0035 of Patent Document 3), but it does not describe screening for MCI. Patent Document 4 describes an AR device that uses augmented reality and is used for training to reduce the risk of dementia. Paragraph 0021 of Patent Document 4 states that it is possible to assess the level of dementia, but it does not describe screening for mild cognitive impairment (MCI). [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2019-10441 [Patent Document 2] WO2020 / 152779 [Patent Document 3] Patent No. 6200615 [Patent Document 4] Japanese Patent Publication No. 2019-76302 [Overview of the project] [Problems that the invention aims to solve]

[0007] Therefore, the present invention aims to provide a system capable of screening individuals who may have MCI (Mild Cognitive Impairment). [Means for solving the problem]

[0008] To solve the above problems, the present invention has the following features. A system for screening whether or not someone has mild cognitive impairment. Based on images using virtual reality, augmented reality, or mixed reality, a single test application can be executed in place without the subject having to move their location. The test app presents the subject with a problem in which they must sequentially select objects from multiple numbers or objects displayed in an image, according to the given rules. The test results of the test app are obtained based on the subject's actions or responses. The acquired test results are recorded in the recording unit, The test results are compared with pre-memorized criteria for determining the possibility of mild cognitive impairment to generate a determination result regarding the possibility of mild cognitive impairment in the subject. The judgment results and test results are displayed to the user via the output unit. The system provides information that allows users to view displayed information and assess the cognitive state of the subject.

[0009] Preferably, a group of subjects is divided into a first group determined to have mild cognitive impairment based on another determination method and a second group of healthy subjects, and a ROC curve (Receiver Operating Characteristic curve) is obtained based on the test results of the first group and the second group, and the determination criteria are determined based on the cut-off value obtained from the ROC curve.

[0010] Preferably, when the AUC (Area Under the Curve) of the ROC curve is greater than or equal to a predetermined value, the determination criteria are determined based on the cut-off value.

[0011] Preferably, the determination criteria may have a margin within a predetermined range.

[0012] Preferably, whether the subject may have mild cognitive impairment is expressed as a percentage indicating the possibility or a stepwise risk level.

[0013] Preferably, based on the test results when the test application is executed for the second time, it is determined whether the subject may have mild cognitive impairment. In that case, the testing app should ideally be one that presents questions requiring the user to select numbers in order.

[0014] As one embodiment, the inspection application is an application that presents a problem in which the user sequentially selects the objects to be selected according to the aforementioned rules regarding the attributes of the objects to be selected.

[0015] In one embodiment, an image processing device for executing a test application and an information processing device for determining the possibility of mild cognitive impairment are provided.

[0016] In one embodiment, the image processing device is a head-mounted display, smart glasses, or a smartphone mounted on goggles.

[0017] As one embodiment, an image processing apparatus for executing an inspection application is provided, and the image processing apparatus determines the possibility of mild cognitive impairment.

[0018] Furthermore, the present invention relates to a computer system for screening whether or not a person has mild cognitive impairment. Based on images using virtual reality, augmented reality, or mixed reality, a single test application can be executed in place without the subject having to move their location. The test app is run in a state where the subject is presented with a problem in which they must sequentially select objects from multiple numbers or multiple selection items displayed in an image, according to the given rules. The test results of the test app are obtained based on the subject's actions or responses. Record the obtained test results, The test results are compared with pre-memorized criteria for determining the possibility of mild cognitive impairment to generate a determination result regarding the possibility of mild cognitive impairment in the subject. The judgment results and test results are displayed to the user via the output unit. This program is a screening tool for mild cognitive impairment, characterized by its function as a means of providing information for users to view displayed information and assess the cognitive state of the subject.

Advantages of the Invention

[0019] According to the present invention, a system capable of screening persons who may have MCI can be provided. According to the present invention, screening can be easily performed, so that more MCI can be expected to be detected at an early stage. Since neuropsychological tests are basically paper-based tests, they are performed two-dimensionally. According to the present invention, since they can be performed in a three-dimensional space, screening can be easily performed with a small number of tasks.

[0020] By determining the criteria for determining the possibility of MCI based on the cut-off value obtained from the ROC curve, statistically valid screening becomes possible.

[0021] By providing a margin in the criteria, false negatives can be eliminated.

[0022] By expressing the possibility of MCI in terms of percentage, risk level, time required for tasks, etc., an impression that the diagnosis is not yet confirmed can be given to the subject.

[0023] Using the test results from the second run improves the accuracy of screening for the possibility of MCI (Mild Cognitive Impairment). Experiments have confirmed that apps that require selecting numbers in order or selecting specified objects are effective in this case.

[0024] Other testing apps can also be used to screen for the possibility of MCI (Mild Cognitive Impairment).

[0025] These and other purposes, features, aspects, and effects of the present invention will become even clearer from the following detailed description in conjunction with the accompanying drawings. [Brief explanation of the drawing]

[0026] [Figure 1] Figure 1 shows the overall configuration of the MCI screening system 3 according to one embodiment of the present invention. [Figure 2] Figure 2 is a block diagram showing the functional configuration of an image processing apparatus 1 according to one embodiment of the present invention. [Figure 3] Figure 3 is a block diagram showing the functional configuration of an information processing device 2 according to one embodiment of the present invention. [Figure 4] Figure 4 is a flowchart illustrating the operation of the MCI screening system 3 according to one embodiment of the present invention. [Figure 5] Figure 5 is a flowchart showing the operation of the inspection application execution process in the image processing device 1. [Figure 6] Figure 6 is a flowchart showing the operation of the image drawing process in the image processing device 1. [Figure 7] Figure 7 is a flowchart showing the operation of the inspection execution process in the image processing device 1. [Figure 8] Figure 8 is a flowchart showing the operation of the inspection and judgment process in the information processing device 2. [Figure 9] Figure 9 is a diagram illustrating the overview of a digit erasure app, which is an example of a testing app. [Figure 10]Figure 10 is a diagram illustrating the overview of a selective deletion app, which is an example of a testing app. [Figure 11] Figure 11 is a diagram illustrating the overview of the Hanamichi app, which is an example of a testing app. [Figure 12] Figure 12 is a diagram illustrating the overview of a maze app, which is an example of a testing app. [Figure 13] Figure 13 is a diagram illustrating the overview of a grid application, which is an example of a testing application. [Figure 14] Figure 14 is a diagram illustrating the overview of a quiz app, which is an example of a testing app. [Figure 15] Figure 15 is a diagram illustrating the overview of an object-finding app, which is an example of an inspection app. [Figure 16] Figure 16 shows the ROC curve, AUC, p-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower when the digit erasure app was performed for the first time under the first condition (20 placements and placement angle of 120 degrees, and so on), using head tracking erasure. [Figure 17] Figure 17 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower when the digit cancellation app was performed for the second time under Condition 1, using head tracking for cancellation. [Figure 18] Figure 18 shows the ROC curve and cutoff value created based on the average time taken for 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower, when the digit cancellation app was performed for the first and second times in Condition 1, using head tracking for cancellation. Note that the average value here refers to the average value for the first and second times for each subject, and not the overall average value for all subjects (the same applies to Figures 18, 21, 24, 27, 30, 33, 36, and 39). [Figure 19]Figure 19 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower when the digit erasure app was performed for the first time under the second condition (20 placements and placement angle of 180 degrees, and so on), using head tracking erasure. [Figure 20] Figure 20 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower when the digit cancellation app was performed for the second time under condition 2, using head tracking for cancellation. [Figure 21] Figure 21 is a graph showing the ROC curve and its cutoff value, created using head tracking erasure, based on the average time taken for 80 subjects with an MMSE score of 28 or higher and the average time taken for 29 subjects with an MMSE score of 27 or lower when the digit erasure app was performed for the first and second times under condition 2. [Figure 22] Figure 22 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower when the Hanado app was run for the first time under the third condition (eliminate two types of flowers, however, selecting two types of flowers is considered an error; the same applies hereafter) using head tracking for elimination. [Figure 23] Figure 23 shows the ROC curve, AUC, P-value, and cutoff value created using head tracking erasure, based on the time taken by 80 subjects with an MMSE score of 28 or higher and the time taken by 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed for the second time under condition 3. [Figure 24] Figure 24 is a graph showing the ROC curve and its cutoff value, created using head tracking erasure, based on the average time taken for 80 subjects with an MMSE score of 28 or higher and the average time taken for 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed once and twice under condition 3. [Figure 25] Figure 25 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower when the Hanado app was run for the first time under the fourth condition (eliminate three types of flowers; however, selecting one type of flower is considered an error; the same applies hereafter) using head tracking for elimination. [Figure 26] Figure 26 shows the ROC curve, AUC, P-value, and cutoff value created using head tracking erasure, based on the time taken by 80 subjects with an MMSE score of 28 or higher and the time taken by 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed for the second time under condition 4. [Figure 27] Figure 27 shows the ROC curve, AUC, P-value, and cutoff value created using head tracking erasure, based on the average time taken for 80 subjects with an MMSE score of 28 or higher and the average time taken for 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed for the first and second times in the fourth condition. [Figure 28] Figure 28 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower when they performed the digit erasure app for the first time under Condition 1, using erasure by touch (tapping by gesture; the same applies hereafter). [Figure 29] Figure 29 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and the time taken by 29 subjects with an MMSE score of 27 or lower when the digit cancellation app was performed for the second time under Condition 1, using touch-based cancellation. [Figure 30] Figure 30 shows the ROC curve and its cutoff value, created based on the average time taken for 80 subjects with an MMSE score of 28 or higher and the average time taken for 29 subjects with an MMSE score of 27 or lower, when the digit cancellation app was performed for the first and second times under Condition 1, using touch-based cancellation. [Figure 31]Figure 31 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower when they performed the digit cancellation app for the first time under condition 2, using touch-based cancellation. [Figure 32] Figure 32 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and 29 subjects with an MMSE score of 27 or lower when the digit cancellation app was performed for the second time under condition 2, using touch-based cancellation. [Figure 33] Figure 33 is a graph showing the ROC curve and its cutoff value, created based on the average time taken for 80 subjects with an MMSE score of 28 or higher and the average time taken for 29 subjects with an MMSE score of 27 or lower, when the digit cancellation app was performed for the first and second times under condition 2, using touch-based cancellation. [Figure 34] Figure 34 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and the time taken by 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed for the first time under condition 3, using touch-based deletion. [Figure 35] Figure 35 is a graph showing the ROC curve and its cutoff value, created based on the time taken by 80 subjects with an MMSE score of 28 or higher and the time taken by 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed for the second time under condition 3, using touch-based erasure. [Figure 36] Figure 36 shows the ROC curves and their cutoff values, created using touch-based cancellation, based on the average time taken for 80 subjects with an MMSE score of 28 or higher and the average time taken for 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed once and twice under condition 3. [Figure 37]Figure 37 shows the ROC curve, AUC, P-value, and cutoff value created based on the time taken by 80 subjects with an MMSE score of 28 or higher and the time taken by 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed for the first time under condition 4, using touch-based cancellation. [Figure 38] Figure 38 shows the ROC curve and its cutoff value, created based on the time taken by 80 subjects with an MMSE score of 28 or higher and the time taken by 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed for the second time under condition 4, using touch-based cancellation. [Figure 39] Figure 39 shows the ROC curves and their cutoff values, created using touch-based cancellation, based on the average time taken for 80 subjects with an MMSE score of 28 or higher and the average time taken for 29 subjects with an MMSE score of 27 or lower when the Hanamichi app was performed for the first and second times under condition 4. [Figure 40] Figure 40 is a table summarizing the AUC, p-value, and cutoff value obtained in Figures 16 to 39. [Modes for carrying out the invention]

[0027] (Overall configuration of MCI screening system 3) In Figure 1, the MCI screening system 3 comprises an image processing device 1 and an information processing device 2.

[0028] Image processing device 1 is a device such as a head-mounted display (HMD), smart glasses, or a device that is realized by attaching a smartphone to goggles, and is a device that can realize mixed reality (MR). However, the image processing device in the present invention may also be a device that can realize augmented reality (AR) or virtual reality (VR).

[0029] While various methods exist for classifying virtual reality, augmented reality, and mixed reality, this specification will describe them as having the following technical significance.

[0030] When virtual reality is used, the image processing device 1 displays images related to the virtual space, giving the user the perception that they are actually present in the virtual space. Generally, when virtual reality is used, the displayed images change in accordance with the user's movements, giving the user the perception that they are moving within the virtual space. When using virtual reality, 3D information of objects in the virtual space, as well as tracking information such as the position and tilt of the image processing device 1 and the user's gaze, are required. The position and orientation of the user viewing the virtual space are calculated, and based on this calculation, images of the virtual space are displayed on the image processing device 1.

[0031] In the case of so-called video see-through augmented reality, the image processing device 1 obtains a real image, and an image of a virtual object is superimposed on the real image, and the combined image of both is displayed. In augmented reality, a real image is recognized for superimposing a virtual object, and the image of the virtual object is superimposed on the recognized real image. Therefore, in this specification, it is assumed that the image processing device 1 does not obtain tracking information such as the position, tilt, or gaze of the user, as in virtual reality or mixed reality. However, this definition is merely a definition used in this specification to make the explanation easier to understand, and it goes without saying that in the present invention, the present invention may be realized by augmented reality using an image processing device that can obtain tracking information.

[0032] In mixed reality, spatial recognition is achieved by recognizing the real space as a three-dimensional shape using image recognition, depth cameras, infrared sensors, laser irradiation, and various other sensors and detectors. When using mixed reality, the image processing device 1 displays an image that makes it appear as if an object exists in the recognized real space, giving the user the perception that a virtual object exists in the real space. Furthermore, when using mixed reality, the image processing device 1 can recognize the position and tilt of the image processing device 1 in the real space, as well as the user's gaze, face, and head orientation (tracking information), and in conjunction with the user's movement in the real space, it can display an image that gives the user the perception that the virtual object remains placed in the real space. In mixed reality, the image processing device 1 can detect the user's gaze, face, and head orientation, and can understand which direction the user is facing. Therefore, the image processing device 1 can change the displayed image according to the user's orientation. Furthermore, in mixed reality, the user's finger and hand movements can be detected, allowing for selection of virtual objects, etc.

[0033] Various methods for spatial recognition in mixed reality have been proposed. For example, one method involves extracting singularities from an image, converting the extracted singularities into 3D coordinates, thereby understanding the surrounding three-dimensional shape, and further determining one's own position. In augmented reality and mixed reality, it is sufficient for virtual objects to be placed in real space, so the way the real space is presented can be either optical see-through or video see-through.

[0034] In the following explanation, the case where the user's gaze is detected by eye-tracking or the like will be used as an example, but it is sufficient to know at least the direction the user is facing. Therefore, all instances of "user's gaze" can be replaced with "user's direction," and in that case, the invention should be understood as using a direction detection unit 14 instead of a gaze detection unit 14 in the image processing device 1. Note that the user's gaze is a concept encompassed by the user's direction. The gaze detection unit 14 is a concept encompassed by the direction detection unit 14. As a specific means for detecting the user's direction other than gaze, for example, cervical movement sensing can be used, but the specific means are not limited to the present invention.

[0035] Although this embodiment assumes the use of mixed reality, embodiments using virtual reality and augmented reality will also be described as appropriate. The MCI screening system of the present invention can be implemented not only using mixed reality, but also using virtual reality and augmented reality.

[0036] The position, tilt, and gaze information of the image processing device 1 will be collectively referred to as tracking information. Since the user's position and tilt are detected by detecting the position and tilt of the image processing device 1, the tracking information of the image processing device 1 and the user's tracking information will be considered synonymous in this explanation.

[0037] Image processing device 1 is attached to the subject and runs an application for MCI screening. The subject performs the test according to the instructions in the application (execution process of the test application: see Figure 5). The test results are transmitted to information processing device 1.

[0038] Information processing device 2 is a device such as a personal computer (PC), tablet, or smartphone, and is capable of communicating with image processing device 1. The communication method may be short-range communication such as Bluetooth®, network communication such as Wi-Fi®, mobile phone networks, or the Internet, or wired communication such as USB connection. In this invention, the communication method is not limited as long as data can be exchanged between image processing device 1 and information processing device 2.

[0039] The information processing device 2 performs a process (inspection and determination process: see Figure 8) to determine the possibility of MCI based on the data transmitted from the image processing device 1, and saves and displays the result.

[0040] In Figure 1, the MCI screening system 3 is depicted as having separate devices for the image processing unit 1 and the information processing unit 2. This is so that the subject can use the image processing unit 1, and the results can be displayed separately on the information processing unit 2.

[0041] If the subject is to be able to directly view the screening test results, the test determination processing in the information processing device 2 may be performed in the image processing device 1. In that case, the inspection results may be displayed using only the image processing device 1, or they may also be displayed using the information processing device 2.

[0042] Therefore, the MCI screening system 3 only needs to be configured to include means for executing the test application and means for determining the test result, and the configuration of the device itself is not limited in this invention.

[0043] The MCI screening system 3 is realized by installing a program on each device that functions as a means for causing a computer system consisting of an image processing device 1 and an information processing device 2, or a computer system consisting of an image processing device 1, to run a test application for presenting a subject with test problems using images of virtual reality, augmented reality, or mixed reality, and as a means for determining whether or not a subject may have mild cognitive impairment by comparing the test results obtained as a result of the subject solving the problems with a judgment criterion for determining whether or not the subject may have mild cognitive impairment.

[0044] (Situations in which the MCI screening system 3 is used) Here are some examples of situations in which the MCI screening system 3 might be used.

[0045] For example, the MCI screening system 3 is used at health checkup centers and local government health fairs. The subject runs the test application using the image processing device 1. After the test, the personal data of the test results is transmitted from the image processing device 1 to the information processing device 2. The information processing device 2 interprets the test results. The organizer either informs the subject of the results on the spot or at a later date.

[0046] Furthermore, the subject or their family member downloads and installs the testing app on a smartphone attached to the HMD or goggles. The subject then uses the HMD or goggles to run the testing app and complete the test. Personal data from the test results is transmitted via the network to an information processing device 2. The information processing device 2 notifies the subject of the test results immediately or at a later date.

[0047] Furthermore, the MCI screening system 3 of the present invention also includes cases where the test results are determined using only the HMD or smartphone. In that case, as described above, the configuration in which personal data is transmitted to the information processing device 2 and the test results are notified separately is not adopted. Instead, the image processing device 1 determines the possibility of MCI after the test and notifies the subject directly.

[0048] While various other usage scenarios are conceivable, these scenarios do not limit the present invention.

[0049] (Configuration of Image Processing Device 1) In Figure 2, the image processing device 1 comprises a communication unit 10, a control unit 11, a storage unit 12, an image processing unit 13, a gaze detection unit 14, a tracking unit 15, a spatial recognition unit 16, an audio output unit 17, a display unit 18, and an input unit 19.

[0050] The communication unit 10 is at least a device capable of communicating with the information processing device 2. The communication method is not limited in this invention and may be any communication method.

[0051] The control unit 11 controls the operation of the entire image processing device 5.

[0052] The storage unit 12 is a recording medium such as memory or an SSD. The storage unit 12 stores an application execution program that controls the operation of the examination application, at least one examination application, and personal data that stores the execution results of the examination application. Note that it is sufficient for at least one examination application to be stored, and it is not necessary for all of the rehabilitation applications described later to be stored.

[0053] The application execution program is a program that controls the overall operation of the image processing device 1. It loads the instructed inspection application, executes the inspection application, and stores the inspection results from the inspection application as personal data. The application execution program is also responsible for sending and receiving data with the information processing device 2.

[0054] The testing application performs tasks such as displaying objects during the test, selecting or deleting objects, and counting the score over time, before passing the test results to the application execution program.

[0055] Personal data is stored by associating information for identifying the subject with test results (such as the time taken to complete the test application and the accuracy rate). The data format of personal data is not limited in this invention.

[0056] The input unit 19 is a device for operating the image processing device 1, such as a wired or wireless switch or touch panel. Since the image processing device 1 can also be operated by recognizing user gestures, it is also possible to consider a camera (not shown) and an image recognition processing unit as functioning as the input unit 19. Furthermore, since a selection operation can be said to have occurred when the gaze remains fixed in a certain place for a certain period of time (i.e., when the user is gazing), the process by which the control unit 11 recognizes an operation based on information from the gaze detection unit 14 can also be understood as a process performed by the input unit 19.

[0057] The display unit 18 is a device for displaying the generated image. In the case of a VR HMD, the display unit 18 is a small display for the left and right eyes. In a VR HMD, the background may be transparent or opaque. In the case of an MR HMD, it is a small display for the left and right eyes. For example, if Microsoft's HOLOLENS® is used as an MR HMD, the background is transparent. When the information processing device 1 is constructed using a device such as a smartphone and goggles to wear it, the display screen of the device becomes the display unit 18.

[0058] For the example of MR, Microsoft's HOLOLENS® is used here, but it goes without saying that other MR devices are also acceptable. In the case of MR, the real world may be perceived as being transmitted through the lens, or real-time images of the real world captured by a camera may be displayed on a small display, and MR may be realized by combining the real-time images with virtual objects. In addition, devices that realize MR using any known method are included in the information processing device 1.

[0059] The image processing device 1 can add new inspection applications to the storage unit 12, for example, by downloading inspection applications that are not currently installed.

[0060] By loading the application execution program into the control unit 11 and executing it, it becomes possible to issue start instructions to each inspection application, set the operating conditions for each inspection application (referred to as "application setting conditions"), store personal data, and transmit personal data to the information processing device 2.

[0061] The audio output unit 17 is a speaker or earphone, etc. In response to instructions from the control unit 11, the audio output unit 17 outputs sounds such as correct and incorrect answers, the sound of the test application starting, and sounds while the test application is running.

[0062] The spatial recognition unit 16 includes a camera and recognizes the three-dimensional shape of the surrounding space of the image processing device 1 using image recognition technology. Since various technologies for recognizing the three-dimensional shape of space by camera photography have already been developed, spatial recognition will be performed using one of these technologies. Note that the spatial recognition unit 16 may be omitted when using VR and AR.

[0063] Furthermore, spatial recognition by the spatial recognition unit 16 may not be necessary depending on the application, and is not essential in this invention.

[0064] The tracking unit 15 recognizes the position and tilt of the image processing device 1. Since various tracking technologies for VR and MR have already been developed, one of these technologies will be used here to recognize the position and tilt.

[0065] The tracking unit 15 is not necessarily structurally included inside the housing of the image processing device 1, but may also detect position, tilt, etc., using sensors attached to the outside of the housing. Therefore, the image processing device 1 may implement not only inside-out tracking but also outside-in tracking. In that case, the tracking unit 15 will be located outside the housing of the image processing device 1, but the image processing device 1 will be considered to include the tracking unit 15 located outside the housing.

[0066] The gaze detection unit 14 is a device that detects the gaze of the user using the image processing device 1. Since various gaze detection technologies for VR and MR have already been developed, gaze recognition will be performed using one of these technologies. If gaze detection is considered as part of tracking technology, then the gaze detection unit 14 may be considered as being included in the tracking unit 15.

[0067] The image processing unit 13 generates a three-dimensional image to be displayed on the display unit 18. Here, the data that forms the basis of the three-dimensional image will be referred to as three-dimensional data. While each examination application is running, the image processing unit 13 stores the three-dimensional structure of the real space recognized by the spatial recognition unit 16 as three-dimensional data in the storage unit 12. The image processing unit 13 also stores the three-dimensional data of the virtual object to be placed in the storage unit 12 using the same coordinate axes as the three-dimensional data of the three-dimensional structure in the real space. The image processing unit 13 also stores the three-dimensional data of tracking information (position, tilt, and line of sight) in the storage unit 12 using the same coordinate axes as the three-dimensional data of the three-dimensional structure in the real space. By managing these three types of three-dimensional data using the same coordinate axes, the image processing unit 13 can generate an image of the virtual object visible from the patient's line of sight and display it on the display unit 53.

[0068] Since processing 3D data is computationally intensive, the image processing unit 13 performs the 3D data calculations separately from the control unit 11. However, it is also possible to perform these calculations in the control unit 11.

[0069] The subjects complete the assigned tasks while looking at the images displayed on the display unit 18.

[0070] (Configuration of Information Processing Device 2) In Figure 3, the information processing device 2 includes a communication unit 20, a control unit 21, an input unit 22, a display unit 23, and a storage unit 24.

[0071] The communication unit 20 is at least a device capable of communicating with the image processing device 1. The communication method is not limited in this invention and may be any communication method.

[0072] The control unit 21 controls the overall operation of the information processing device 2. The memory unit 24 is a recording medium such as memory, SSD, or HDD. The memory unit 24 stores the examination program, the criteria for determining the possibility of MCI, and each individual's data. The memory unit 24 may also store past data, such as previous result data. The input unit 22 is a device for operating the information processing device 2. The display unit 23 is a liquid crystal screen or the like.

[0073] The information processing device 2 determines the possibility of MCI (Mild Cognitive Impairment) based on the personal data transmitted from the image processing device 1 after the examination, and outputs the result.

[0074] (Overall operation of MCI screening system 3) The overall operation of the MCI screening system 3 will be explained with reference to Figure 4. First, the information processing device 2 prompts the user to specify a test application (S201) and receives a test start instruction from the user (S202). Once a test start instruction is received, the information processing device 2 instructs the image processing device 1 to execute the test application.

[0075] The application execution program of the image processing device 1 loads the specified inspection application and starts the inspection application (S101). While the inspection application is running, the image processing device 1 saves the inspection results (time taken for the inspection, number of correct answers, number of incorrect answers, etc.) (S102).

[0076] When the examination application is completed (S103), the image processing device 1 transmits the examination results, associated with information that identifies the subject, to the information processing device 2 as personal data. The information used to identify the subjects is assumed to have been initially set through a process not shown in the diagram. It should be noted that identifying the subjects is not mandatory; the information processing device 2 may associate the data with the subjects after receiving it. It goes without saying that the information used to identify the subjects does not need to be in a format that identifies individuals.

[0077] The information processing device 2 receives the transmitted personal data (S203), determines the test result based on the criteria for determining the possibility of MCI (S204), and displays and saves the test result (S205).

[0078] Furthermore, the present invention also includes the capability for the image processing device 1 to determine the possibility of MCI (Mild Cognitive Impairment). In this case, the selection of the examination application shown in Figure 4 is performed by the image processing device 1, and the image processing device 1 determines the examination result without transmitting or receiving personal data.

[0079] (Execution of the inspection application on the image processing device 1) Next, referring to Figure 5, we will explain the execution process of the inspection application in the image processing device 1.

[0080] As a premise, the control unit 11 of the image processing device 1 reads and executes the application execution program, and executes the application instructed by the information processing device 2 according to the instructed application setting conditions. Application setting conditions include, for example, the number and type of objects to be deleted, and the range in which objects are displayed (angles such as left / right and up / down). Also, as a premise, the operation shown in Figure 5 is assumed to be for MR, but explanations for VR and AR will be added as appropriate.

[0081] In the case of MR, the spatial recognition unit 16 uses its built-in camera to recognize the surrounding three-dimensional space and convert it into three-dimensional data (S110). Next, the tracking unit 15 recognizes the position and tilt of the image processing device 1 (S111).

[0082] The control unit 11, through processing S110 and S111, determines the position of the image processing device 1 in three-dimensional space and recognizes the location of the image processing device 1 in three-dimensional space. Note that the processing for spatial recognition and tracking may also be performed by the OS of the image processing device 1, as is included in this invention. Furthermore, the spatial recognition processing is not limited to image recognition using the built-in camera; it may also be performed using various other sensors and detectors such as depth cameras, infrared sensors, and laser irradiation, as is included in this invention.

[0083] In the case of VR, the control unit 11 generates 3D data of the virtual space in S110, and recognizes the position and inclination of the image processing device 1 in S111 to recognize where the image processing device 1 is located in the virtual space.

[0084] In the case of AR, the control unit 11 captures images of the surrounding area in S110 and S111 and determines the region in which the problem will be synthesized.

[0085] Next, the control unit 11 generates a problem to be used in the specified test application (S112).

[0086] Here, we will explain the problem generation process in detail. Based on the application settings, the control unit 11 generates problems. Since this is a test for screening MCI, the control unit 11 generates problems so that the difficulty level of the problems remains constant. For example, when running a number erasure application, if the information processing device 2 specifies that the number of numbers to be erased is between 1 and 20, the application must not generate a problem that lowers the difficulty by, for example, arranging the numbers 1 through 20 in a single line. Furthermore, if the arrangement angle is specified as 180 degrees, the numbers must not be placed unevenly on one side. Therefore, the problems need to be generated according to predetermined criteria (including criteria for randomness). Additionally, problems are generated according to app settings such as object color and size.

[0087] The control unit 11 places the generated object in the virtual space onto the 3D data of the virtual space (S113). Subsequently, the execution of the inspection application proceeds, and in parallel with the execution of the inspection application (S115), the image processing unit 13 performs drawing processing (S114).

[0088] Figure 6 is a flowchart of the image rendering process in the image processing unit 13. Since the image rendering process requires high-speed and large-scale calculations, it is assumed here that the processing is performed by the image processing unit 13, which is separate from the control unit 11. However, it goes without saying that, depending on the performance of the control unit 11, the control unit 11 may also perform the processing.

[0089] The image processing unit 13 detects the position and tilt of the image processing device 1 on the 3D data based on information from the tracking unit 15 (S120). Next, it detects the direction of the gaze on the 3D data based on information from the gaze detection unit 14 (S121). Then, the image processing unit 13 places the object in the virtual space on the 3D data according to the current position and tilt of the image processing device 1 and the direction of the gaze, face, head, etc. (S122), determines the image to be displayed on the display unit 18, and displays the image (S123). By performing this processing for each frame, it becomes possible to display on the display unit 18 a display that makes it appear as if a virtual image exists in real space, in accordance with the movement of the image processing device 1.

[0090] In the case of VR, the image processing unit 13 is the same as in the case of MR described above, except that it displays an image on the display unit 18 in which the object is placed on the image of the virtual space.

[0091] In the case of AR, the image processing unit 1 recognizes the image of the real world being captured by the image processing unit 1, and the image processing unit 13 synthesizes the object in the virtual world.

[0092] Figure 7 is a flowchart showing the operation during the test execution process in Figure 5. The control unit 11 presents the subject with the assigned problem in S113, using text, audio, or other means (S130).

[0093] Here, we will explain each testing application with reference to Figures 8 through 15. As shown in Figure 9, when the digit erasure application is executed, in S113, the specified number of digits are placed in space.

[0094] The outer dashed line shown in Figure 9 represents the maximum area in the space where the numbers are placed. The information processing device 2 receives instructions, such as 120 degrees or 180 degrees, as application setting conditions, and the numbers are placed at intervals of 120 degrees or 180 degrees from the subject's perspective.

[0095] The number of numbers to be placed is also specified by the information processing device 2 as an application setting condition. In the example in Figure 9, the numbers 1 through 20 are placed. The dotted line inside Figure 9 represents the area visible to the subject. As the subject moves their head, the visible area changes accordingly (the same applies hereafter).

[0096] As shown in Figure 9, the instruction "Please select in ascending order" is displayed. This is the presentation of the task to be completed (S130). The number erasure app allows you to select and erase numbers in a specified order. Numbers can be erased either by touching them with a gesture (hereinafter referred to as "tap" or "touch") or by head tracking, where the number is erased when the user's gaze is fixed on the corresponding number (hereinafter the same applies).

[0097] If the wrong numbers are selected, they will not be erased, and the process will continue until the numbers are erased in the correct order. In the case of a number erasure app, the evaluation criterion is the time taken to erase all the numbers (total time).

[0098] Figures 10 and beyond will be explained in the same manner. Figure 10 shows an example of what happens when you run the Select Delete application. The task is to select and eliminate stars. In a selection / elimination app, points should be added for selecting the correct image and deducted for selecting the wrong image, but various evaluation methods can be adopted.

[0099] Figure 11 shows an example of what happens when the Hanamichi app is run. In the Hanamichi app, virtual flowers and virtual stands on which they are placed are displayed in a virtual space. As the user walks, the flowers appear to be nearby as they approach them. When the user gets close enough to touch the flowers, they can make them disappear. For example, one of the instructions might say, "Please remove the black flowers." In this case, it is good to evaluate the results using the number of correct / incorrect answers, similar to how you would evaluate selection / elimination, but other methods are also acceptable.

[0100] Figure 12 shows an example of what happens when you run the maze app. In maze apps, a 3D maze is displayed in a virtual space, along with a 2D maze that serves as a hint. Along the way, items are located at points marked with stars, and the goal is to reach the end by touching and eliminating these items. In maze apps, it is good to evaluate the time taken to reach the goal.

[0101] Figure 13 shows an example of what happens when a grid app is run. The grid app displays instructions to move around the grid, such as, "Move 3 spaces west, 4 spaces south, and 1 space east." The subject moves around the grid according to the instructions. It is a good idea to evaluate the time taken to correctly reach the goal, or to evaluate whether the position after movement is correct.

[0102] Figure 14 shows an example of running a quiz app. In the quiz app, a question such as "What month is it today?" is displayed, and answer choices are shown in a virtual space. The participant selects and eliminates the displayed choices. In the quiz app, multiple questions are presented, and evaluation criteria should include the time taken to answer correctly and the number of correct and incorrect answers.

[0103] The quiz app could also include questions about dates, days of the week, or addresses. Additionally, questions about animal and fruit names, arithmetic questions, questions about abstract concepts (for example, questions about the concept of "fruit" if the subject is bananas and oranges), questions about language ability, questions about attention, and questions about memory may be asked. The questions can be determined by referencing questions used in exams such as MoCaA-J (The Montreal Cognitive Assessment) and MMSE (Mini Mental State Examination).

[0104] Figure 15 shows an example of what happens when you run a hidden object app. Items such as a laptop or telephone are displayed in a three-dimensional space, and a banknote is hidden underneath them. When given the task, "Find the banknote," the participant makes a gesture of touching the item, lifts it up, and checks if the banknote is underneath. The time it takes to find the banknote should be used as the evaluation criterion.

[0105] Furthermore, any and all other testing applications can be considered, and any such testing application is included in the present invention.

[0106] Let's return to the explanation of Figure 7. After the task is presented, the subjects select the object that will be the subject of the task (in the example above, the smallest number, the star, the black flower, etc.). The subject can choose between head tracking or gesture-based tapping, but the choice of method will be specified in the app settings beforehand.

[0107] Assume that the image processing device 1 has detected that the subject has selected an object in the specified manner (S131).

[0108] The image processing device 1 determines whether the selected object was correct or incorrect (S132). Each testing app handles correct and incorrect answers differently. For example, in a number erasure app, if the numbers are entered in the wrong order, the app will not erase them but will leave them as they are, play an audio message indicating the mistake, and prompt the user to select the correct numbers. In the case of a selection / deletion application, if the wrong object is selected, points will be deducted. In this way, the image processing device 1 performs predetermined correct / incorrect processing for each application (S133).

[0109] The image processing device 1 then records the test results (number of correct answers, number of incorrect answers, etc.) in association with the subject's information (S134).

[0110] The image processing device 1 determines whether the conditions for the inspection to end have been met (S135). For a digit erasure application, the termination condition is that all digits have been erased. For a selection erasure application, the termination condition is that all specified objects have been erased.

[0111] If the termination conditions have not been met, the image processing device 1 returns to the operation of S131. If the termination conditions have been met, the image processing device 1 transmits personal data of the test results, such as the time taken, the number of correct answers, and the number of incorrect answers, to the information processing device 2 (S136) and terminates the process.

[0112] (Inspection and judgment processing in information processing device 2) Next, the inspection and judgment process in the information processing device 2 will be explained with reference to Figure 8. The information processing device 2 references the personal data of the received test results (S210). This reference does not need to be in real time and may be referenced later.

[0113] The information processing device 2 refers to the criteria for determining the possibility of MCI for the test application used in the test (S211). For example, in the case of a number erasure application, if the time required to erase all numbers exceeds a certain amount, the information processing device 2 will issue a determination that there is a possibility of MCI (Mild Cognitive Impairment).

[0114] Here, we will explain an example of how to determine the criteria for assessing the possibility of MCI (Mild Cognitive Impairment). For example, subjects selected for testing are asked to run a testing app, and the evaluation results (such as time taken and accuracy rate) are compiled. Then, using other assessment methods such as MoCA-J, these subjects are divided into two groups: Group 1, who have MCI (Mild Cognitive Impairment), and Group 2, who are healthy individuals. For these two groups, we calculate the Receiver Operating Characteristic curve (ROC curve). We pre-determine that if the Area Under the Curve (AUC) of the ROC curve is above a certain value, it may be possible to identify subjects who potentially have MCI (Mild Cognitive Impairment). Then, when the AUC exceeds a certain value, the cutoff value of the ROC curve is determined. This cutoff value serves as an indicator for identifying subjects who may have MCI when tested using the test app.

[0115] These criteria are determined using statistical methods for each test application and each application setting condition.

[0116] The information processing device 2 determines from the personal data to be examined whether or not there is a possibility of MCI (S212), records the determination result, and displays it (S213).

[0117] (Example of criteria for determining the possibility of MCI) The following provides specific examples of how to determine the criteria for assessing the possibility of MCI (Mild Cognitive Impairment), referring to Figures 16 through 39. Figures 16 through 39 show the ROC curves, cutoff values, and p-values ​​created based on the time taken by 80 subjects with an MMSE score of 28 or higher and the time taken by 29 subjects with an MMSE score of 27 or lower. In Figures 16 through 39, the term "touch" indicates the condition where an object is selected by tapping it using a gesture. Similarly, the term "head tracking" indicates a condition where the selection is made by directing one's gaze towards the target object and staring at it.

[0118] In the number erasure app, the 120 degrees and 180 degrees refer to the maximum angle at which the numbers are placed, and are conditions specified in the app settings.

[0119] In the Hanado app, the options "2 types" and "3 types" refer to the number of flower types you can select from the displayed flowers. If there are a total of 4 types of flowers, selecting 2 types means the remaining 2 types are not allowed to be selected. Similarly, selecting 3 types means the remaining 1 type is not allowed to be selected.

[0120] The cutoff values ​​obtained in Figures 16 through 39 are summarized in a table in Figure 40. The cutoff values ​​obtained for each application setting condition serve as the criteria for determining the possibility of MCI (Mild Cognitive Impairment). However, the present invention is not limited to using the cutoff value directly as the judgment criterion. As described later, a value with a margin added to the cutoff value may be used as the criterion for judgment. Furthermore, as will be discussed later, the probability of having MCI can be indicated as a percentage or as a risk level, centered around a cutoff value. In other words, the criteria for judgment only need to be determined based on the cutoff value.

[0121] For example, if the cutoff value is used directly as the judgment criterion, then under the conditions shown in Figure 40, if the time required to run the test application exceeds the cutoff value, the information processing device 2 will determine that the subject may have MCI. In this case, the table shown in Figure 40 will serve as the criteria for determining the possibility of MCI.

[0122] According to the examples shown in Figures 16 to 38, the AUC value exceeds 0.7 when the time taken to run the touch-to-erase digit erasure app for the second time, and the time taken to run the three types of touch-to-erase flower path apps for the second time, are used as the screening criteria. In statistical methods using ROC curves, an AUC of 0.7 or higher is considered to indicate moderate accuracy, and it is believed that using the cutoff value at that point will yield test results with reasonable accuracy.

[0123] Therefore, the most preferred screening methods would be the time taken to run the touch-to-erase digit erasure app (app settings: 20 digits arranged at a 180-degree angle) for the second time, and the time taken to run the three types of touch-to-erase flower path apps for the second time.

[0124] However, regarding other examples, the AUC values ​​do not necessarily mean that they are completely unreliable. Also, from a screening perspective, it may be better to broadly identify subjects who may have MCI and connect them to a medical diagnosis. Therefore, in this invention, the cutoff value for determining the possibility of MCI is not limited to only the case where the AUC is 0.7 or higher.

[0125] The cutoff value for AUC (area under which microscopic cognitive impairment) should be used as a criterion for determining the possibility of MCI (Mild Cognitive Impairment) should be determined on an appropriate basis by medical professionals or other experts.

[0126] Furthermore, using the test results from the second time the testing app is run may be preferable for screening purposes. This is because even if the test results are not good the first time due to unfamiliarity with the testing app, if the person has the ability to learn from the first test and reflect that in the second test, the likelihood of having MCI (Mild Cognitive Impairment) is considered low. Conversely, if there is no difference between the first and second test results, and the second test result is also poor, it may indicate MCI (Mild Cognitive Impairment).

[0127] Figures 18, 21, 24, 27, 30, and 33, which average the results of the first and second tests, use an average obtained by equally dividing the results of the first and second tests. However, as mentioned above, if the goal is to reflect the presence or absence of learning ability, the test results may be evaluated using a weighted average that assigns weight to the results of the second test.

[0128] Therefore, the possibility of MCI can be determined based on the results of the second test, or it can be determined using a weighted average that incorporates the results of the second test. In other words, it is preferable that the possibility of MCI be determined based on the results of the second test. Needless to say, in this invention, the possibility of MCI may be determined based on the results of the first test.

[0129] In this invention, as a method for determining screening criteria, the test application is run on healthy individuals and individuals with MCI under the same application settings, and an ROC curve is calculated from the evaluation results such as the time required and the accuracy rate. This ROC curve is then used as a cutoff value to determine the likelihood of MCI, making it possible to screen individuals who are statistically and medically likely to have MCI.

[0130] Note that depending on the statistical software used to calculate the ROC curve, the same cutoff value may be output even if the original data is the same. Therefore, when using screening criteria, it is advisable to include a predetermined margin of error in the cutoff value.

[0131] For example, as shown in the example in Figure 29, when using tap-to-erasure, the cutoff value when the number erasure app is run for the second time under the second condition (20 placements and placement angle of 180 degrees) is 78.99 seconds. However, assuming a margin of, for example, minus 3% (approximately 2 seconds), the decimal part will be rounded. In this case, the criterion for determining the possibility of MCI (Mild Cognitive Impairment) could be set as a time taken of 77 seconds or more, and those who take 77 seconds or more could be identified as potentially having MCI. This margin of error can be determined as appropriate, and the above figures are merely illustrative examples.

[0132] Alternatively, the likelihood of MCI could be expressed as a percentage or risk level, using the degree to which the test results approach the criteria for determining the possibility of MCI as an indicator. For example, if the probability significantly exceeds the probability criteria, the percentage of MCI probability may be increased, and if it is close to the probability criteria, the percentage of MCI probability may be decreased. Alternatively, the risk can be expressed in several stages. Additionally, the possibility of MCI may be expressed by showing changes or trends from previous test results.

[0133] Specifically, in the example shown in Figure 29, the time taken could be divided into 5-second intervals, with a 65% probability of MCI for subjects taking 79 seconds or more but less than 84 seconds, a 90% probability for subjects taking 84 seconds or more, a 50% probability for subjects taking 75 seconds or more but less than 79 seconds, and a 30% or less probability for subjects taking less than 75 seconds. Alternatively, these could be expressed using risk classifications such as A, B, C, and D.

[0134] Furthermore, by providing a program capable of executing the operation process of the present invention and installing and using it in various devices, the present invention can be implemented in a general-purpose image processing device, and widespread adoption of the present invention can be expected.

[0135] Furthermore, by providing a program capable of executing the operation processing of the present invention as a recording medium that is not a temporary storage medium, it becomes possible to execute the present invention on a general-purpose image processing device, and widespread adoption of the present invention can be expected.

[0136] Although the present invention has been described in detail above, the above description is merely illustrative in all respects and is not intended to limit its scope. Needless to say, various improvements and modifications can be made without departing from the scope of the present invention. Each constituent element of the invention disclosed herein shall stand as an independent and standalone invention. Inventions obtained by combining each constituent element in any way shall also be included in the present invention. [Industrial applicability]

[0137] This invention provides a system that can screen individuals who may have MCI (Mild Cognitive Impairment), and is industrially applicable. [Explanation of Symbols]

[0138] 1 Image Processing Device 2. Information Processing Device 3 MCI Screening System 10 Communications Department 11 Control Unit 12 Storage section 13 Image Processing Unit 14 Eye-line detection unit 15 Tracking section 16 Spatial recognition section 17 Audio output section 18 Display 19 Input section 20 Communications Department 21 Control Unit 22 Input section 23 Display section 24 Memory section

Claims

1. A system for screening whether or not someone has mild cognitive impairment. Based on images using virtual reality, augmented reality, or mixed reality, a single test application can be executed in place without the subject having to move their location. The aforementioned testing application presents the subject with a problem in which they sequentially select objects from among multiple numbers or multiple selection objects displayed in the image, according to the given rules. Based on the subject's actions or responses, the test results of the test application are obtained. The acquired inspection results are recorded in the recording unit, The test results are compared with pre-stored criteria for determining the possibility of mild cognitive impairment to generate a determination result regarding the possibility of mild cognitive impairment in the subject. The judgment result and the inspection result are displayed to the user via the output unit. A screening system for mild cognitive impairment, which provides information for the user to view the displayed information and determine the cognitive state of the subject.

2. The screening system according to claim 1, characterized in that the subjects are divided into a first group judged to have mild cognitive impairment based on a different assessment method and a second group of healthy individuals, an ROC curve (Receiver Operating Characteristic curve) is calculated based on the test results of the first and second groups, and the judgment criteria are determined based on the cutoff value obtained from the ROC curve.

3. The screening system according to claim 2, characterized in that the judgment criteria are determined based on the cutoff value when the AUC (Area Under the Curve) of the ROC curve is greater than or equal to a predetermined value.

4. The screening system according to claim 2, characterized in that the judgment criteria have a predetermined margin of error.

5. The screening system according to claim 2, characterized in that whether or not the subject may have mild cognitive impairment is expressed as a percentage indicating the possibility or as a graded risk level.

6. A screening system for mild cognitive impairment according to claim 1, characterized in that it determines whether or not the subject may have mild cognitive impairment based on the test results obtained when the test application is run for the second time.

7. The screening system according to claim 1, characterized in that the aforementioned testing application is an application that presents a problem in which the user selects numbers in order.

8. The screening system according to claim 1, characterized in that the inspection application is an application that presents a problem in which the user sequentially selects the objects to be selected according to the rules relating to the attributes of the objects to be selected.

9. An image processing device for executing the inspection application, The screening system according to claim 1, further comprising an information processing device for generating a determination result of the possibility of mild cognitive impairment.

10. The screening system according to claim 9, characterized in that the image processing device is a head-mounted display, smart glasses, or a smartphone attached to goggles.

11. The system includes an image processing device for executing the aforementioned inspection application, The screening system according to claim 1, characterized in that the image processing device generates a determination result regarding the possibility of mild cognitive impairment.

12. A computer system for screening whether or not someone has mild cognitive impairment. Based on images using virtual reality, augmented reality, or mixed reality, a single test application can be executed in place without the subject having to move their location. The aforementioned test application is run in a state where the subject is presented with a problem in which they sequentially select objects from among multiple numbers or multiple selection objects displayed in the aforementioned image, according to the given rules. Based on the subject's actions or responses, the test results of the test application are obtained. The obtained inspection results are recorded, The test results are compared with pre-stored criteria for determining the possibility of mild cognitive impairment to generate a determination result regarding the possibility of mild cognitive impairment in the subject. The judgment result and the inspection result are displayed to the user via the output unit. A program for screening for mild cognitive impairment, characterized in that it functions as a means of providing information for the user to view the displayed information and determine the cognitive state of the subject.

Images