Dementia detection device, dementia detection method, and program

Abstract

To create opportunities for the early detection of dementia. [Solution] The dementia detection device communicates with a target terminal used by the target person and with family terminals used by each member of the family. The dementia detection device stores the personal information of the target person and each member, as well as photographic information of the target person, based on the consent of the target person and each member. The dementia detection device transmits and displays the photographic information to the target person's terminal. The dementia detection device uses an AI model to converse with the target person via the target person's terminal, based on the personal information and photographic information. The dementia detection device analyzes the content of the conversation based on the personal information and photographic information to determine whether the target person's memory is unclear. If the dementia detection device determines that the target person's memory is unclear, it notifies the family terminal.

Description

Technical Field

[0001] This disclosure relates to a technology for facilitating early detection of dementia.

Background Art

[0002] In recent years, the aging process has advanced in Japan, and the number of dementia patients has been increasing. In an aging society, the problem of dementia is serious, and early detection has become an issue. Patent Document 1 describes an electronic device utilization system for determining the presence or absence of dementia suspicion from daily conversations.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The risk of dementia tends to increase when the amount of conversation is low. However, due to living alone away from family, the number of elderly people living alone with a low amount of conversation is increasing year by year. Also, even among the elderly living with family or residing in facilities, the daily amount of conversation tends to be low due to problems such as family being busy or a shortage of caregivers.

[0005] One of the objectives of this disclosure is to create an opportunity for early detection of dementia.

Means for Solving the Problems

[0006] To solve the above problems, from one aspect of this disclosure, a dementia detection device includes a subject terminal used by a subject, and communication means for communicating with family terminals respectively used by members constituting the family, A storage means for storing the personal information of the subject and each member, and photographic information relating to the subject, based on the consent of the subject and each member, A photo display means that transmits and displays the photo information on the target person's terminal, A conversation means that uses an AI model to converse with the subject via the subject's terminal, based on the personal information and photographic information, A memory determination means that determines whether or not the subject's memory is unclear by analyzing the content of the conversation based on the aforementioned personal information and photographic information, If the subject's memory is unclear, a notification means for notifying the family terminal, It is equipped with.

[0007] From another perspective of this disclosure, the dementia detection method performed by the dementia detection device is: It communicates with the target person's device and with the family devices used by each member of the family. The personal information of the aforementioned subject and each member, and photographic information relating to the aforementioned subject, will be stored based on the consent of the aforementioned subject and each member. The aforementioned photographic information is transmitted to the subject's terminal and displayed. Using the aforementioned target person's terminal, an AI model is used to conduct a conversation with the target person based on the aforementioned personal information and the aforementioned photographic information. Based on the aforementioned personal information and photographic information, the content of the conversation is analyzed to determine whether or not the subject's memory is unclear. If the subject's memory is unclear, a notification will be sent to the family terminal.

[0008] In yet another aspect of this disclosure, the program is It communicates with the target person's device and with the family devices used by each member of the family. The personal information of the aforementioned subject and each member, and photographic information relating to the aforementioned subject, will be stored based on the consent of the aforementioned subject and each member. The aforementioned photographic information is transmitted to the subject's terminal and displayed. Through the subject terminal, based on the personal information and the photographic information, a conversation is carried out with the subject using an AI model, Based on the personal information and the photographic information, by analyzing the content of the conversation, it is determined whether the memory of the subject is unclear, When the memory of the subject is unclear, the computer is made to execute a process of notifying the family terminal.

Advantages of the Invention

[0009] According to the present disclosure, an opportunity for early detection of dementia can be created.

Brief Description of the Drawings

[0010] [Figure 1] An example of the schematic configuration of a dementia detection system is shown. [Figure 2] An example of the hardware configuration of a server and a subject terminal is shown. [Figure 3] It is a diagram for explaining data registered in various databases. [Figure 4] It is a block diagram showing an example of the functional configuration of a server. [Figure 5] It is a diagram for explaining the registration of information by a card information capture API. [Figure 6] It is a diagram for explaining the function of a photo correction unit. [Figure 7] An example of evaluation data. [Figure 8] It is a diagram for explaining the flow until disclosure information is provided to a medical institution. [Figure 9] It is a flowchart of a memory determination process. [Figure 10] An example of a photo subjected to mosaic processing. [Figure 11] It is a block diagram showing the functional configuration of a dementia detection device according to a second embodiment. [Figure 12] It is a flowchart of a process by a dementia detection device according to a second embodiment.

Mode for Carrying Out the Invention

[0011] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. [First Embodiment] (Overall Configuration) FIG. 1 is an example of the schematic configuration of a dementia detection system 100 to which the dementia detection device of the present disclosure is applied. The dementia detection system 100 is a system that uses an AI model to talk to a subject and creates an opportunity for early detection of dementia by notifying the family of the subject based on the analysis results of the conversation. Specifically, the AI model learns old photos and videos related to the subject in advance, talks to the subject about memories, etc., and notifies of suspected dementia and provides situation information based on the conversation history. In other words, the dementia detection system 100 provides a service for preventing and early detecting dementia to the subject and their family by having the AI model talk to the subject.

[0012] In the dementia detection system 100 of FIG. 1, the server 1, the subject terminal 2, the family terminal 3a, and the family terminal 3b are communicably connected via a network 5 such as the Internet. The family terminals 3a and 3b are also collectively referred to as the family terminal 3.

[0013] In this embodiment, as an example, the subject is an elderly person who does not have sufficient daily communication. Also, the family is composed of two members, the eldest son and the second son of the subject. The subject uses the subject terminal 2, the eldest son uses the family terminal 3a, and the second son uses the family terminal 3b. Note that the members constituting the family are not limited to two people with a blood relationship with the subject, and are composed of one or more arbitrary members such as family members and relatives who are closely connected to the subject.

[0014] The subject terminal 2 is a smartphone, tablet, PC, etc. used by the subject, and receives voice from the server 1 for the AI model to talk to the subject, or transmits voice from the subject to talk to the AI model to the server 1. The subject terminal 2 is an example of the subject terminal of the present disclosure. <​ Family terminal 3 is a smartphone, tablet, PC, etc., used by each member, and receives notifications from server 1 regarding the subject's memory being unclear, and sends and registers photos of memories with the subject to server 1. Family terminal 3 is an example of a family terminal in this disclosure.

[0016] Server 1 is an information processing device that processes, stores, and transmits various types of data, and has an AI model. Server 1 is also connected to the personal information database (hereinafter referred to as "DB") 31, photo information DB 32, conversation information DB 33, and disclosure information DB 34, which will be described later. The AI ​​model is a Large Language Model (LLM) capable of understanding multimodal information and converses with the subject via the subject terminal 2. Specifically, the AI ​​model exchanges information with the personal information DB 31, photo information DB 32, conversation information DB 33, and disclosure information DB 34 to create conversation content with the subject and select photos related to the conversation.

[0017] Server 1 stores personal information and photographic information about the subject and their family in various databases. It sends audio and related photographs to the subject's terminal 2 for the AI ​​model to converse with the subject, and receives audio and personal information from the subject's terminal 2 for the subject to converse with the AI ​​model. Server 1 also analyzes the conversation between the AI ​​model and the subject to determine whether the subject's memory is unclear. If the subject's memory is unclear, Server 1 notifies the family terminal 3 that the subject may have dementia and recommends that they seek medical attention.

[0018] Server 1 may be a virtual server located in a cloud environment. Server 1 is an example of the dementia detection device described herein.

[0019] (Hardware configuration) Figure 2(a) is a block diagram showing an example of the hardware configuration of Server 1. As shown in the figure, Server 1 comprises an interface 11, a processor 12, memory 13, a recording medium 14, a display unit 15, and an input unit 16. These components are interconnected via a bus with the personal information DB 31, the photo information DB 32, the conversation information DB 33, and the disclosure information DB 34.

[0020] Interface 11 exchanges data with the subject terminal 2 and family terminal 3. Interface 11 is used to receive voice and personal information from the subject terminal 2 for the subject to converse with the AI ​​model, and to send voice and photos related to the conversation from the AI ​​model to the subject terminal 2 for the AI ​​model to converse with the subject. Interface 11 also receives personal information and photo information from family terminal 3, and sends notifications to family terminal 3 if the subject's memory is unclear.

[0021] Processor 12 is a computer such as a CPU (Central Processing Unit) that controls the entire server 1 by executing pre-prepared programs. Processor 12 can be a CPU, GPU (Graphics Processing Unit), DSP (Digital Signal Processor), MPU (Micro Processing Unit), FPU (Floating Point Number Processing Unit), PPU (Physics Processing Unit), TPU (Tensor Processing Unit), quantum processor, microcontroller, or a combination of these.

[0022] Memory 13 consists of ROM (Read Only Memory), RAM (Random Access Memory), and other components. Memory 13 stores programs executed by the processor 12. Memory 13 is also used as working memory while the processor 12 is executing various processes.

[0023] The recording medium 14 is a non-volatile, non-temporary recording medium such as a disk-shaped recording medium or semiconductor memory, and is configured to be detachable from the server 1. The recording medium 14 stores various programs that the processor 12 executes. When the server 1 performs a memory determination process, the programs stored on the recording medium 14 are loaded into the memory 13 and executed by the processor 12.

[0024] The display unit 15 displays a predetermined image, for example, using an LCD (Liquid Crystal Display). The input unit 16 is used by the operator managing server 1, and can be a keyboard, mouse, touch panel, etc.

[0025] Figure 3 is a diagram illustrating the data registered in various databases. As shown in Figure 3, the personal information database 31 stores the following as personal information: basic four pieces of information, family structure data, the subject's medical claim data, consent data, and facial image data. The basic four pieces of information are the name, address, date of birth, and gender of the subject and each member of the family. Family structure data is data indicating the family structure, such as the relationship between each member and the subject, for example, eldest son, second son, etc. Medical claim data is the medical fee statement submitted by the medical institution to the insurer, and information on the illness currently being treated and medication can be obtained from the subject's medical claim data. Consent data is data indicating consent to services provided by the dementia detection system 100, such as the use of facial image data of the subject and family, the import of the subject's medical claim data, and the receipt of notifications that the subject's memory is unclear. Facial image data is current facial image data of the subject and each member.

[0026] The consent data may be a single consent data for the entire service, or it may be consent data for each item necessary to provide the service, such as the use of facial images or the import of medical claim data. For convenience, this embodiment describes an example in which each member of the target person and their family has given consent for the entire service.

[0027] As shown in Figure 3, the photo information DB32 stores image data and attribute data of photographs related to the subject as photo information. Photographs related to the subject include memorable photos of the subject or their family. Attribute data includes the date the photograph was taken, the people in the photograph, the location where the photograph was taken, and anecdotes related to the photograph.

[0028] As shown in Figure 3, conversation data and evaluation data are stored as conversation information in the conversation information DB33. Specifically, Server 1 stores in the conversation information DB33 the audio data presented by the AI ​​model to the target terminal 2, and the audio data acquired from the target terminal 2, as conversation data including the date and time of the conversation. In addition, Server 1 creates evaluation data that assesses the target's memory and stores it in the conversation information DB33.

[0029] As shown in Figure 3, the Disclosure Information DB34 stores information to be disclosed to medical institutions. Specifically, when a subject visits a medical institution due to suspected dementia, Server 1 stores the information to be disclosed to that medical institution in the Disclosure Information DB34. The disclosed information includes, for example, the subject's medical claim data, evaluation data, and conversation data from when the subject's memory was unclear.

[0030] Figure 2(b) is a block diagram showing an example of the hardware configuration of the target terminal 2. As shown in the figure, the target terminal 2 includes an interface 21, a processor 22, a memory 23, a recording medium 24, a display unit 25, and an input unit 26.

[0031] Interface 21 exchanges data with Server 1 via Network 5. Interface 21 is used to send voice and personal information from the subject to Server 1 for conversation with the AI ​​model, and to receive voice and photos related to the conversation from Server 1 for the AI ​​model to converse with the subject.

[0032] The processor 22 is a computer such as a CPU, and controls the entire target terminal by executing a pre-prepared program. The processor 22 can be a CPU, GPU, DSP, MPU, FPU, PPU, TPU, quantum processor, microcontroller, or a combination thereof.

[0033] Memory 23 is composed of ROM, RAM, etc. Memory 23 stores programs executed by the processor 22. Memory 23 is also used as working memory while the processor 22 is executing various processes.

[0034] The recording medium 24 is a non-volatile, non-temporary recording medium such as a disk-shaped recording medium or semiconductor memory, and is configured to be detachable from the user terminal 2. The recording medium 24 stores various programs executed by the processor 22. The display unit 25 displays a predetermined image, for example, an LCD. The input unit 26 is a touch panel or the like, and is used when the user performs a predetermined operation.

[0035] Since the hardware configuration of family terminal 3 is the same as that of target terminal 2, we will omit the explanation for convenience.

[0036] (Functional Configuration) Figure 4 is a block diagram showing an example of the functional configuration of Server 1. Server 1 includes a personal information DB 31, a photo information DB 32, a conversation information DB 33, and a disclosure information DB 34. Functionally, Server 1 also includes a family information registration unit 41, a medical claim data registration unit 42, a photo information linkage unit 43, a photo information registration unit 44, a family information acquisition unit 45, a medical claim data acquisition unit 46, a photo information acquisition unit 47, a conversation data registration unit 48, a conversation data acquisition unit 49, an AI model, a disclosure information registration unit 60, a photo display unit 61, a conversation image acquisition unit 62, an audio presentation unit 63, an audio acquisition unit 64, a family contact unit 65, and a disclosure information presentation unit 66. The AI ​​model functions as a photo correction unit 51, a photo selection unit 52, a conversation content creation unit 53, and a memory determination unit 54.

[0037] The family information registration unit 41, the medical claim data registration unit 42, the photo information linkage unit 43, the photo information registration unit 44, the family information acquisition unit 45, the medical claim data acquisition unit 46, the photo information acquisition unit 47, the conversation data registration unit 48, the conversation data acquisition unit 49, the AI ​​model, the disclosure information registration unit 60, the photo display unit 61, the conversation image acquisition unit 62, the voice presentation unit 63, the voice acquisition unit 64, the family contact unit 65, and the disclosure information presentation unit 66 are realized by the processor 12 executing a program.

[0038] When starting the services provided by the Dementia Detection System 100, each member of the target person and their family installs the prescribed application on the target person's terminal 2 and family terminal 3. When registering personal information and photographic information, the target person's terminal 2 and family terminal 3 launch the application through a prescribed operation and select a registration mode for registering various types of information.

[0039] The family information registration unit 41 uses the card information import API (Application Programming Interface) from the subject's My Number Card, via the subject's terminal 2, to register the subject's basic four pieces of information and facial image data in the personal information database 31. The family information registration unit 41 also obtains the subject's consent data and registers it in the personal information database 31.

[0040] The family information registration unit 41 retrieves card information from each family member's My Number Card via a family terminal 3 or the like, and registers each member's basic information and facial image data in the personal information database 31 using an API. The family information registration unit 41 also obtains each member's consent data and family structure data and registers them in the personal information database 31.

[0041] Figure 5 illustrates the registration of information using the card information import API. As shown in Figure 5, the subject uses the subject terminal 2 to perform a predetermined operation to indicate consent to the services provided by the dementia detection system 100, and then reads their My Number Card. Server 1 uses the card information import API to obtain consent data, the subject's basic four pieces of information held in the My Number Card, and facial image data from the subject terminal 2, and registers them in the personal information DB 31.

[0042] Furthermore, each member of the family uses their own family terminal 3 to indicate their consent to the services provided by the dementia detection system 100 through a prescribed operation, and then reads their My Number Card. At this time, each member inputs their relationship to the person concerned as family composition data. Server 1 uses a card information import API to obtain each member's consent data and family composition data, as well as the basic four pieces of information and facial image data held by the My Number Card, from the family terminal 3 used by each member, and registers them in the personal information DB 31.

[0043] For the sake of explanation, it is assumed that the target individuals and each member will read their own My Number Cards using their own devices. However, this is not the only option; the devices used by the target individuals and each member are arbitrary, as long as they can read their respective My Number Cards and register the information in various databases using the card information import API.

[0044] The claims data registration unit 42 registers claims data in the personal information database 31 using a card information import API via the target person's My Number Card and the target person's terminal 2. The claims data registration unit 42 may also link with an API that imports information related to medications, automatically import medication information for the target person as it becomes available, and store it in the personal information database 31.

[0045] The photo information linkage unit 43 acquires image data of photos related to the subject, as well as attribute data such as the time and location the photos were taken, by linking with the photo folders of the subject terminal 2 and family terminal 3. The photo information linkage unit 43 is not limited to linking with photo folders; it may also acquire photo information by linking with SNS (Social Networking Service) used by the subject or family. Furthermore, the photo information linkage unit 43 may acquire memories related to photos entered by the subject or family as attribute data from the subject terminal 2 and family terminal 3.

[0046] The photo information registration unit 44 registers the image data and attribute data of the photograph as photo information in the photo information DB 32.

[0047] The family information acquisition unit 45 acquires basic information of the subject and each member, family structure data, consent data, and facial image data from the personal information database 31 and provides them to the AI ​​model. The AI ​​model learns by taking in the basic information of the subject and each member, family structure data, consent data, and facial image data.

[0048] The claims data acquisition unit 46 acquires claims data and medication information for the target individual from the personal information database 31 and provides it to the AI ​​model. The AI ​​model learns by incorporating the claims data and medication information.

[0049] The photo information acquisition unit 47 acquires photo information from the photo information DB 32 and provides it to the AI ​​model. The AI ​​model learns by incorporating the photo information.

[0050] The AI ​​model converses with the subject via the subject terminal 2. The content of the conversation mainly consists of reminiscences about photos displayed on the subject terminal 2. The conversation data registration unit 48 registers the content of the conversation between the AI ​​model and the subject as conversation data in the conversation information DB 33. Specifically, the conversation data registration unit 48 acquires the audio data that the AI ​​model presents to the subject terminal 2 for conversation with the subject, as well as the audio data acquired from the subject terminal 2, as conversation data including the date and time of the conversation. The conversation data registration unit 48 registers the conversation data in the conversation information DB 33 as needed.

[0051] The conversation data acquisition unit 49 acquires conversation data from the conversation information DB 33 and provides it to the AI ​​model. The AI ​​model learns by taking in the conversation data.

[0052] The AI ​​model functions as a photo correction unit 51, a photo selection unit 52, a conversation content creation unit 53, and a memory judgment unit 54.

[0053] The photo correction unit 51 links a photograph to the people in it, based on the facial image data of the subject and each member registered in the personal information DB 31 and the image data of the photograph registered in the photo information DB 32. Figure 6 is a diagram illustrating the function of the photo correction unit 51. As shown in Figure 6(a), the photo correction unit 51 identifies person 58 in the photograph as the eldest son and person 59 as the second son, based on the current facial image data of the eldest and second sons and the image data of the photograph, and links them to the photograph. The photo correction unit 51 may store the link data, which links the photograph to the people in it, in the photo information DB 32, or it may classify the photographs registered in the photo information DB 32 based on the link data.

[0054] Next, the photo correction unit 51 displays the photo and the person linked to that photo on the subject terminal 2 and family terminal 3 based on the link data. If there is an error in the link, the subject or each member can correct the link using the subject terminal 2 or family terminal 3 by performing a predetermined operation. For example, as shown in Figure 6(b), if person 58 in the photo is actually the second son and person 59 is the eldest son, the subject or each member can use the subject terminal 2 or family terminal 3 to input the correction of the link error by performing a predetermined operation. The photo correction unit 51 updates the link data based on the input data, and the AI ​​model relearns the photo information based on the updated link data.

[0055] The photo correction unit 51 displays the photo and the person linked to that photo based on the link data, but is not limited to this. Based on attribute data, the date and location of the photo may also be displayed on the subject terminal 2 and family terminal 3. In this case, if there are errors in the date or location of the photo, the subject or each member will use the subject terminal 2 or family terminal 3 to input the information to correct the date and location of the photo through a predetermined operation. The photo correction unit 51 updates the attribute data based on the input data, and the AI ​​model relearns the updated attribute data.

[0056] The photo selection unit 52 selects a photo to send to the target terminal 2 and display. In the first pattern, the photo selection unit 52 selects a photo that corresponds to the conversation content by comparing the conversation content with personal information and photo information. Specifically, the photo selection unit 52 identifies the time period, the child's age at the time, and the event from the conversation content and selects a corresponding photo.

[0057] In the second pattern, the photo selection unit 52 randomly selects a photo from the photos registered in the photo information DB 32 in order to initiate a conversation with the AI ​​model.

[0058] The third pattern is when the subject is conversing with a designated member. The photo selection unit 52 acquires a photograph of the member the subject is conversing with as conversation image data, and identifies the member the subject is conversing with based on the conversation image data and face image data. Next, the photo selection unit 52 selects a photograph showing the member the subject is conversing with.

[0059] The conversation content creation unit 53 creates conversation content for the AI ​​model to communicate with the target person via the target person terminal 2, based on information registered in various databases.

[0060] The first pattern involves creating conversation content in response to a conversation initiated by the subject to the AI ​​model. In other words, it involves creating new conversation content that corresponds to the content and flow of the conversation with the subject. In this case, the conversation content creation unit 53 creates new conversation content based on the conversation content initiated to the AI ​​model, past conversation content, and photos corresponding to the conversation content. Specifically, the AI ​​model learns conversation data corresponding to the conversation content, attribute data and link data of photos corresponding to the conversation content, past conversation data related to the photos, and personal information of the members pictured in the photos, and then creates appropriate conversation content.

[0061] The second pattern involves creating conversation content when the conversation is initiated by the AI ​​model. In this case, the conversation content creation unit 53 creates conversation content based on randomly selected photos and past conversations related to the photos. Specifically, the AI ​​model learns attribute data and link data of the selected photos, past conversation data related to the photos, and personal information of the members pictured in the photos, and creates appropriate conversation content.

[0062] The third pattern involves creating conversation content when the subject is conversing with a designated member. In this case, the conversation content creation unit 53 creates conversation content based on a photograph of the member the subject is conversing with, and past conversations related to the photograph. Specifically, the AI ​​model learns attribute data and link data of the photograph of the member the subject is conversing with, past conversation data related to the photograph, and personal information of the member the subject is conversing with, and creates appropriate conversation content.

[0063] The memory determination unit 54 analyzes the latest conversation content related to the photograph based on personal information, photographic information, and conversational data to determine whether the subject's memory is unclear or not. Specifically, the AI ​​model compares the latest conversational data related to the photograph with the photograph's attribute data and link data to determine whether the subject's memory is unclear or not.

[0064] If the subject's memory is unclear, the memory determination unit 54 creates evaluation data that assesses the subject's memory based on the most recent conversation data and the previous conversation data related to the photograph. Figure 7 shows an example of evaluation data. Specifically, the AI ​​model creates evaluation data as shown in Figure 7 by evaluating three types of memories: memories related to the time the photograph was taken, memories related to the family members in the photograph, and memories related to the location where the photograph was taken. The created evaluation data is stored in the conversation information DB 33. In the evaluation data, the subject's memory is represented by evaluation indicators of circles, triangles, and crosses. A circle represents "clear memory," a triangle represents "vague memory," and a cross represents "unclear memory." According to the evaluation data shown in Figure 7, it is easy to understand the types of memories that are unclear to the subject and when those memories became unclear. Note that the evaluation data shown in Figure 7 is just an example, and the types of memories and evaluation indicators can be set arbitrarily.

[0065] The Disclosure Information Registration Unit 60 registers the information to be disclosed to the medical institution where the subject is receiving treatment in the Disclosure Information DB 34. Figure 8 is a diagram illustrating the flow of providing disclosure information to a medical institution. As shown in Figure 8, the Disclosure Information Registration Unit 60 registers the claims data and medication information registered in the Personal Information DB 31 and the evaluation data registered in the Conversation Information DB 33 as disclosure information in the Disclosure Information DB 34. The Disclosure Information Registration Unit 60 may also register conversation data related to types of memories evaluated as unclear in the evaluation data, information on the time and location of photography, etc., as supplementary information to the evaluation data in the Disclosure Information DB 34.

[0066] The photo display unit 61 displays the photo by transmitting the image data of the photo selected by the photo selection unit 52 to the target terminal 2.

[0067] The conversation image acquisition unit 62 acquires photographs of the members the subject is talking to as conversation image data when the subject is talking to a designated member. The conversation image data is, for example, image data of photographs taken with cameras on the subject terminal 2 or family terminal 3, and is provided to the AI ​​model.

[0068] The voice presentation unit 63 transmits and presents audio data for the AI ​​model to converse with the target user terminal 2. Specifically, the voice presentation unit 63 converts the conversation content created by the conversation content creation unit 53 into audio data, transmits it to the target user terminal 2, and presents it.

[0069] The voice acquisition unit 64 acquires voice data from the subject terminal 2 for the subject to converse with the AI ​​model.

[0070] If the Family Communication Unit 65 determines that the subject's memory is unclear, it notifies the Family Terminal 3 that dementia is suspected and recommends that the subject visit a medical institution. Furthermore, as shown in Figure 8, the Family Communication Unit 65 outputs the information to be disclosed to the medical institution regarding the subject in the form of a two-dimensional code, such as a QR code (registered trademark), and notifies the Family Terminal 3. The Family Communication Unit 65 may use the application installed when starting the service provided by the Dementia Detection System 100 to make the notification, or it may register the email address of the Family Terminal 3 as personal information in advance and notify via email. By making these notifications, the family can be encouraged to check the condition of the subject suspected of having dementia and to visit a medical institution.

[0071] The information disclosure unit 66 presents the information to be disclosed upon request from a medical institution. As shown in Figure 8, the family who receives the notification visits a designated medical institution and informs them that the person in question is suspected of having dementia. At this time, as shown in Figure 8, the family presents the two-dimensional code notified to the family terminal 3 at the medical institution. When the medical institution reads the two-dimensional code through a prescribed operation, the information disclosure unit 66 transfers the information to the electronic medical record or other necessary locations as needed.

[0072] In the above configuration, the photo correction unit 51, photo selection unit 52, memory determination unit 54, photo display unit 61, and family contact unit 65 of Server 1 are examples of the photo correction means, photo selection means, memory determination means, photo display means, and notification means, respectively, as disclosed. Furthermore, the conversation content creation unit 53, voice presentation unit 63, and voice acquisition unit 64 are examples of the conversation means, as disclosed. In addition, the personal information DB 31 and photo information DB 32 are examples of the storage means, as disclosed.

[0073] (Memory determination process) Next, we will explain the memory determination process performed by Server 1. Figure 9 is a flowchart of the memory determination process performed by Server 1. This process is achieved by the processor 12 shown in Figure 2 executing a pre-prepared program.

[0074] First, Server 1 retrieves the latest conversation data between the AI ​​model and the subject from the conversation information DB 33 (Step S101). Next, Server 1 uses the AI ​​model to select a photo based on the retrieved conversation data and information registered in various DBs (Step S102). Furthermore, Server 1 uses the AI ​​model to create new conversation content based on the retrieved conversation data and information registered in various DBs (Step S103).

[0075] Server 1 displays the selected photo by sending the image data of the photo to the target terminal 2 (step S104). Server 1 also converts the newly created conversation content into audio data and sends it to the target terminal 2 for presentation (step S105). Next, Server 1 obtains the audio data of the target person conversing with the AI ​​model from the target terminal 2 (step S106). Next, Server 1 registers the new conversation content between the AI ​​model and the target person as conversation data in the conversation information DB 33 (step S107).

[0076] Server 1 determines whether the conversation with the subject has ended (step S108). Specifically, Server 1 determines that the conversation with the subject has ended if it does not acquire any new audio data from the subject terminal 2. If it is determined that the conversation with the subject has not ended (step S108; No), Server 1 returns to the process in step S101 and continues the conversation with the subject via the subject terminal 2. On the other hand, if it is determined that the conversation with the subject has ended (step S108; Yes), Server 1 analyzes the latest conversation content regarding the photograph based on the information registered in various databases to determine whether the subject's memory is unclear (step S109).

[0077] If it is determined that the subject's memory is not unclear (Step S110; No), Server 1 terminates the memory assessment process. On the other hand, if it is determined that the subject's memory is unclear (Step S110; Yes), Server 1 notifies Family Terminal 3 that the subject may have dementia and recommends that they visit a medical institution (Step S111). At this time, Server 1 creates evaluation data and notifies Family Terminal 3 of the disclosure information, including the medical claim data and evaluation data, along with a QR code. Thus, the memory assessment process terminates.

[0078] In this disclosure, the target terminal 2 is assumed to be a smartphone or tablet, but is not limited to these; it may also be a robot equipped with the functions of server 1. In this case, the robot may project photos onto a wall or screen, or it may connect to a smartphone or television for display. By designing the robot in a way that enhances the target's motivation to engage in conversation compared to a smartphone, etc.

[0079] Furthermore, the photographs disclosed herein are not limited to film photographs but also include digital photographs. Additionally, the dementia detection system 100 disclosed herein is not limited to still photographs but may also utilize video.

[0080] Furthermore, if the AI ​​model on Server 1 determines from the medication information registered in the personal information DB31 that the subject was taking dementia-related medication, it will create conversation content assuming the subject has dementia. This allows for conversations that take into account the characteristics of dementia patients, which differ from those of healthy individuals.

[0081] According to this dementia detection system 100, Server 1 uses an AI model to frequently engage in conversations and reminiscences with individuals at high risk of dementia, such as the elderly. Having others empathize with these conversations has a healing effect on the individuals. Furthermore, Server 1 can increase the amount of conversation the individuals have, thus potentially preventing dementia. Server 1 also analyzes the conversation content to determine whether the individual's memory is unclear, and if dementia is suspected, it can notify the family to recommend a visit to a medical institution. Therefore, it can help facilitate the early detection of dementia.

[0082] [First variation] Server 1 may use the AI ​​model to interview the subject based on the medical claim data registered in the personal information DB31 to confirm the medication status after hospital visits and their physical condition after medication. In this case, Server 1 registers the content of the interview as conversation data in the conversation information DB33 and keeps a log. When the subject next visits the hospital, Server 1 provides the log to the hospital using a QR code or the like, allowing doctors and pharmacists to easily understand the medication status and physical condition at home.

[0083] [Second variation] The services provided by the dementia detection system 100 may be offered, for example, as an option for other AI communication services for the elderly.

[0084] [Third variation] When displaying a photograph on the target terminal 2, Server 1 may blur the faces of members who have consented to the use of their facial images and other individuals who have not been included in the photograph by applying a mosaic effect. Figure 10 shows an example of a photograph with mosaic processing applied. In this way, Server 1 can protect privacy by applying a mosaic effect to the faces of members who have not consented to the use of their facial images and to the faces of other people who have been accidentally included in the photograph.

[0085] [Second Embodiment] Figure 11 is a block diagram showing the functional configuration of a dementia detection device according to the second embodiment. The dementia detection device 90 includes a communication means 91, a storage means 92, a photo display means 93, a conversation means 94, a memory determination means 95, and a notification means 96.

[0086] Figure 11 is a flowchart of the processing performed by the dementia detection device 90. The communication means 91 communicates with the subject terminal 2 used by the subject and with family terminals 3 used by each member of the family (step S201). The memory means 92 stores the personal information of the subject and each member, as well as photographic information about the subject, based on the consent of the subject and each member (step S202). The photo display means 93 transmits and displays the photographic information to the subject terminal 2 (step S203). The conversation means 94 uses an AI model to converse with the subject via the subject terminal 2, based on the personal information and photographic information (step S204). The memory determination means 95 analyzes the content of the conversation based on the personal information and photographic information to determine whether the subject's memory is unclear or not (step S205). The notification means 96 notifies the family terminal if the subject's memory is unclear (step S206).

[0087] According to the dementia detection device 90 of the second embodiment, by using AI to converse with the subject and notifying the subject's family based on the analysis results of the conversation, it is possible to create an opportunity for the early detection of dementia.

[0088] In addition, some or all of the above embodiments (including modifications, the same applies hereinafter) may also be described as follows, but are not limited to the following.

[0089] (Note 1) A communication means for communicating with the target person's device, and with the family devices used by each member of the family, A storage means for storing the personal information of the subject and each member, and photographic information relating to the subject, based on the consent of the subject and each member, A photo display means that transmits and displays the photo information on the target person's terminal, A conversation means that uses an AI model to converse with the subject via the subject's terminal, based on the personal information and photographic information, A memory determination means that determines whether or not the subject's memory is unclear by analyzing the content of the conversation based on the aforementioned personal information and photographic information, If the subject's memory is unclear, a notification means for notifying the family terminal, A dementia detection device equipped with the following features.

[0090] (Note 2) The aforementioned photographic information includes image data of a photograph and attribute data indicating the subject's memories related to the photograph. The aforementioned photograph display means transmits the image data of the photograph to the subject terminal and displays it. The dementia detection device described in Appendix 1, wherein the memory determination means determines whether or not the subject's memory is unclear by comparing the content of the conversation with the subject regarding the photograph with the attribute data of the photograph.

[0091] (Note 3) The notification means is a dementia detection device as described in Appendix 1, which creates disclosure information to be disclosed to a medical institution regarding the subject, and notifies the family terminal of a two-dimensional code indicating the disclosure information.

[0092] (Note 4) The memory determination means, if the subject's memory is unclear, creates evaluation data that evaluates the subject's memory at the most recent and previous time, based on the content of the most recent conversation regarding the photograph and the content of the previous conversation regarding the photograph. The aforementioned personal information includes the medical claim data of the aforementioned subject, The disclosed information includes the claims data and the dementia detection device described in Appendix 3, including the evaluation data.

[0093] (Note 5) The dementia detection device described in Appendix 4 is a table that evaluates three types of memories: memories related to the time the photograph was taken, memories related to the family members in the photograph, and memories related to the location where the photograph was taken.

[0094] (Note 6) The aforementioned personal information includes facial image data of the subject and each member, and family structure data indicating family composition. The system includes a photo correction means that identifies a person in the photograph based on the image data and face image data of the aforementioned photograph and links them to the photograph. The dementia detection device described in Appendix 2, wherein the conversation means engages in conversation with the subject based on the attribute data of the photograph, the members pictured in the photograph, and the family structure data.

[0095] (Note 7) A photo selection means that selects a photo corresponding to the content of the conversation by comparing the content of the conversation with the subject, the personal information and the photo information, The aforementioned photo display means is a dementia detection device according to Appendix 6 that transmits and displays image data of a photograph corresponding to the content of the conversation to the target person's terminal.

[0096] (Note 8) The aforementioned photo selection means identifies the member with whom the subject is conversing based on the conversation image data and the facial image data, and selects a photo in which that member is pictured. The aforementioned photo display means is a dementia detection device according to Appendix 7, which transmits and displays image data of a photograph showing the members with whom the subject is conversing to the subject's terminal.

[0097] (Note 9) A dementia detection method performed by a dementia detection device, It communicates with the target person's device and with the family devices used by each member of the family. The personal information of the aforementioned subject and each member, and photographic information relating to the aforementioned subject, will be stored based on the consent of the aforementioned subject and each member. The aforementioned photographic information is transmitted to the subject's terminal and displayed. Using the aforementioned target person's terminal, an AI model is used to conduct a conversation with the target person based on the aforementioned personal information and the aforementioned photographic information. Based on the aforementioned personal information and photographic information, the content of the conversation is analyzed to determine whether or not the subject's memory is unclear. A dementia detection method that notifies the family terminal if the subject's memory is unclear.

[0098] (Note 10) It communicates with the target person's device and with the family devices used by each member of the family. The personal information of the aforementioned subject and each member, and photographic information relating to the aforementioned subject, will be stored based on the consent of the aforementioned subject and each member. The aforementioned photographic information is transmitted to the subject's terminal and displayed. Using the aforementioned target person's terminal, an AI model is used to conduct a conversation with the target person based on the aforementioned personal information and the aforementioned photographic information. Based on the aforementioned personal information and photographic information, the content of the conversation is analyzed to determine whether or not the subject's memory is unclear. A program that causes a computer to execute a process to notify the family terminal if the subject's memory is unclear.

[0099] (Note 11) The photographic display means is a dementia detection device as described in Appendix 1, which displays a photograph in which the faces of people other than the patient and members who have given their consent are blurred.

[0100] While the present disclosure has been described above with reference to embodiments, the present disclosure is not limited to the embodiments described above. Various modifications to the structure and details of the present disclosure may be understood by those skilled in the art within the scope of the present disclosure. That is, the present disclosure includes the entire disclosure, including the claims, and of course, various modifications and alterations that those skilled in the art may make in accordance with the technical idea. [Explanation of symbols]

[0101] 1 server 2. Target user's device 3, 3a, 3b family terminal 11, 21 Interfaces 12, 22 processors 13, 23 memory 14, 24 Recording media 15, 25 Display section 16, 26 Input section 31 Personal information DB 32 Photo Information Database 33 Conversation Information Database 34 Disclosure Information Database 51 Photo Correction Section 52 Photo Selection Section 53 Conversation Content Creation Department 54 Memory judgment section 65 Family Liaison Department 66 Disclosure Information Presentation Department 100 Dementia Detection Systems

Claims

1. A communication means for communicating with the target person's device, and with the family devices used by each member of the family, A storage means for storing the personal information of the subject and each member, and photographic information relating to the subject, based on the consent of the subject and each member, A photo display means that transmits and displays the photo information on the target person's terminal, A conversation means that uses an AI model to converse with the subject via the subject's terminal, based on the personal information and photographic information, A memory determination means that determines whether or not the subject's memory is unclear by analyzing the content of the conversation based on the aforementioned personal information and photographic information, If the subject's memory is unclear, a notification means for notifying the family terminal, A dementia detection device equipped with the following features.

2. The aforementioned photographic information includes image data of a photograph and attribute data indicating the subject's memories related to the photograph. The aforementioned photograph display means transmits the image data of the photograph to the subject terminal and displays it. The dementia detection device according to claim 1, wherein the memory determination means determines whether or not the subject's memory is unclear by comparing the content of the conversation with the subject regarding the photograph with the attribute data of the photograph.

3. The dementia detection device according to claim 1, wherein the notification means creates disclosure information to be disclosed to a medical institution regarding the subject, and notifies the family terminal of a two-dimensional code indicating the disclosure information.

4. The memory determination means, if the subject's memory is unclear, creates evaluation data that evaluates the subject's memory at the most recent and previous time, based on the content of the most recent conversation regarding the photograph and the content of the previous conversation regarding the photograph. The aforementioned personal information includes the medical claim data of the aforementioned subject, The dementia detection device according to claim 3, wherein the disclosed information includes the claims data and the evaluation data.

5. The dementia detection device according to claim 4, wherein the evaluation data is a table that evaluates three types of memories: memories related to the time the photograph was taken, memories related to the family members in the photograph, and memories related to the location where the photograph was taken.

6. The aforementioned personal information includes facial image data of the subject and each member, and family structure data indicating family composition. The system includes a photo correction means that identifies a person in the photograph based on the image data and face image data of the aforementioned photograph and links them to the photograph. The dementia detection device according to claim 2, wherein the conversation means engages in conversation with the subject based on the attribute data of the photograph, the members pictured in the photograph, and the family structure data.

7. A photo selection means that selects a photo corresponding to the content of the conversation by comparing the content of the conversation with the subject, the personal information and the photo information, The dementia detection device according to claim 6, wherein the photo display means transmits and displays image data of a photograph corresponding to the content of the conversation to the target person's terminal.

8. The aforementioned photo selection means identifies the member with whom the subject is conversing based on the conversation image data and the facial image data, and selects a photo in which that member is pictured. The dementia detection device according to claim 7, wherein the photo display means transmits and displays image data of a photograph showing the members with whom the subject is conversing to the subject's terminal.

9. A dementia detection method performed by a dementia detection device, It communicates with the target person's device and with the family devices used by each member of the family. The personal information of the aforementioned subject and each member, and photographic information relating to the aforementioned subject, will be stored based on the consent of the aforementioned subject and each member. The aforementioned photographic information is transmitted to the subject's terminal and displayed. Using the aforementioned target person's terminal, an AI model is used to conduct a conversation with the target person based on the aforementioned personal information and the aforementioned photographic information. Based on the aforementioned personal information and photographic information, the content of the conversation is analyzed to determine whether or not the subject's memory is unclear. A dementia detection method that notifies the family terminal if the subject's memory is unclear.

10. It communicates with the target person's device and with the family devices used by each member of the family. The personal information of the aforementioned subject and each member, and photographic information relating to the aforementioned subject, will be stored based on the consent of the aforementioned subject and each member. The aforementioned photographic information is transmitted to the subject's terminal and displayed. Using the aforementioned target person's terminal, an AI model is used to conduct a conversation with the target person based on the aforementioned personal information and the aforementioned photographic information. Based on the aforementioned personal information and photographic information, the content of the conversation is analyzed to determine whether or not the subject's memory is unclear. A program that causes a computer to execute a process to notify the family terminal if the subject's memory is unclear.

Images