Care support system, server device, server device control method, and storage medium

Abstract

The present invention provides a care support system that contributes to enabling sharing of care know-how according to characteristics of individual care recipients. The care support system comprises an input means, an analysis means, and an output means. The input means acquires data in a plurality of formats at a care site. The analysis means analyzes the data in the plurality of formats and extracts support information. The output means outputs information to a caregiver or a care recipient on the basis of the support information. The support information includes care procedures and precautions specific to the care recipient.

Description

Nursing care support system, server device, control method for server device, and storage medium

[0001] The present invention relates to a nursing care support system, a server device, a control method for the server device, and a storage medium.

[0002] There is a technology related to a nursing care support system that records the state of care recipients and the content of nursing care.

[0003] For example, Patent Document 1 discloses a system related to input support for medical and nursing care records, which morphologically analyzes the text input by voice to extract classification information, and detects and presents the omission of necessary information by comparison with patterns.

[0004] Japanese Patent No. 6594981

[0005] In the nursing care chart system as described in Patent Document 1, there are problems such as difficulty in transmitting the nursing care procedures and precautions specific to the care recipient from the sender of information to the recipient. For example, the detailed nursing care know-how optimized for the care recipient, such as the standing position of the caregiver during meal assistance, the angle at which the spoon is held out, and the timing of speaking, remains accumulated as the tacit knowledge of veteran caregivers, and systematic sharing has not been achieved.

[0006] The main object of the present invention is to provide a nursing care support system, a server device, a control method for the server device, and a storage medium that contribute to making it possible to share nursing care know-how according to the characteristics of each care recipient.

[0007] According to a first aspect of the present invention, there is provided a nursing care support system including an input means for acquiring a plurality of types of data at the nursing care site, an analysis means for analyzing the plurality of types of data to extract support information, and an output means for outputting information to a caregiver or a care recipient based on the support information, wherein the support information includes nursing care procedures and precautions specific to the care recipient.

[0008] According to a second aspect of the present invention, a server device is provided, comprising: an input means for acquiring data in a care setting, which includes data in multiple formats, including voice data, text data, video data, and sensor data; an analysis means for analyzing the data in multiple formats to extract support information, including care procedures and points to note specific to the person being cared for; and an output means for providing personalized information to the caregiver or the person being cared for based on the support information.

[0009] A third aspect of the present invention provides a control method for a server device, which acquires data in a care setting in multiple formats, including voice data, text data, video data, and sensor data; analyzes the data in multiple formats to extract support information including care procedures and points to note specific to the person being cared for; and provides personalized information to the caregiver or the person being cared for based on the support information.

[0010] According to a fourth aspect of the present invention, a computer-readable storage medium is provided that stores a program for a computer mounted on a server device to perform the following: a process of acquiring data in a care setting, which includes data in multiple formats, including voice data, text data, video data, and sensor data; a process of analyzing the data in multiple formats to extract support information, including care procedures and points to note specific to the person being cared for; and a process of providing personalized information to the caregiver or the person being cared for based on the support information.

[0011] From each perspective of the present invention, a care support system, a server device, a control method for the server device, and a storage medium are provided that contribute to enabling the sharing of care know-how tailored to the individual characteristics of each person receiving care. However, the effects of the present invention are not limited to those described above. The present invention may also produce other effects in lieu of or in conjunction with these effects.

[0012] Figure 1 is a diagram illustrating the outline of one embodiment. Figure 2 is a flowchart showing the operation of one embodiment. Figure 3 is a diagram showing an example of the overall configuration of a care support system according to one embodiment of the present disclosure. Figure 4 is a block diagram showing an example of the system configuration of the present disclosure. Figure 5 is a processing flowchart showing the overall processing procedure of the care support system of the present disclosure. Figure 6 is a diagram illustrating the data management of the present disclosure. Figure 7 is a diagram showing an example of the display of a terminal of the present disclosure. Figure 8 is a flowchart relating to anomaly detection through analysis of vital data and behavioral information. Figure 9 is a diagram showing an example of the display of a terminal of the present disclosure. Figure 10 is a diagram showing an example of the hardware configuration of a server device related to the present disclosure.

[0013] First, an overview of one embodiment will be described. The reference numerals in the drawings attached to this overview are provided for convenience as examples to aid understanding, and this overview is not intended to be limiting in any way. Furthermore, unless otherwise specified, the blocks shown in each drawing represent functional units, not hardware units. The connecting lines between blocks in each drawing include both bidirectional and unidirectional lines. Unidirectional arrows schematically indicate the flow of the main signal (data) and do not exclude bidirectional flow. In this specification and in the drawings, elements that can be similarly described are given the same reference numerals to avoid redundant explanation.

[0014] A care support system according to one embodiment comprises an input means 101, an analysis means 102, and an output means 103 (see Figure 1). The input means 101 acquires data in multiple formats at the care site (step S1 in Figure 2). The analysis means 102 analyzes the data in multiple formats and extracts support information (step S2). The output means 103 outputs information to the caregiver or the person receiving care based on the support information (step S3). The support information includes care procedures and points to note specific to the person receiving care.

[0015] The above-mentioned care support system extracts local knowledge (also referred to as support information) from data acquired from caregiving sites. Local knowledge includes information such as the individual characteristics and preferences of the person receiving care, situations requiring special attention, and subtle know-how based on the caregiver's experience. Local knowledge is a general term for the characteristics, preferences, and know-how of each person receiving care. For example, the care support system may extract information about the person receiving care as local knowledge, such as their specific characteristics, preferences, care procedures, and points to note. Local knowledge may also include information about the environment in which care is provided, such as the condition of the floor, steps, the location of furniture and tools, and points to note regarding barrier-free design.

[0016] The care support system uses the extracted local knowledge to provide information to caregivers. For example, the care support system provides caregivers with videos (care manuals) that reflect care know-how tailored to the individual characteristics of the person receiving care. By watching these videos, caregivers can visually share care know-how tailored to the individual characteristics of the person receiving care.

[0017] Existing care support systems lack a means of visually communicating care procedures. Relying solely on textual information makes it difficult to accurately convey detailed explanations of care actions and appropriate responses to the care recipient's reactions from the information provider to the recipient. The care support system disclosed in this application solves this problem by providing caregivers with videos that reflect care know-how.

[0018] Furthermore, existing care support systems lack sufficient mechanisms to dynamically update care procedures in response to changes in the care recipient's condition, resulting in a problem where they cannot respond quickly to the needs of the field. The care support system disclosed in this application solves the above problem by extracting local knowledge using multimodal data acquired from the care site, and reflecting this extracted local knowledge in the manual.

[0019] Specific embodiments will be described in more detail below with reference to the drawings.

[0020] [First Embodiment] The first embodiment will be described in more detail with reference to the drawings.

[0021] <Outline> First, in order to facilitate understanding of the embodiments disclosed in this application, an overview of the embodiments will be provided.

[0022] In care settings, caregivers (care workers, care staff) carry devices such as smartphones and tablets to perform their daily caregiving duties. For example, a caregiver might say, "Mr. / Ms. A, let's take a bath." The care support system disclosed in this application automatically recognizes the bathing assistance scene in the care setting in response to this voice command and takes a picture of the scene using a smartphone or similar device.

[0023] While the following describes an example in a caregiving setting, the application of this embodiment is not limited to caregiving. For example, it can achieve similar effects in user support in hospitals, medical facilities, daycare centers, or facilities for people with disabilities. In these facilities as well, it is important to share support methods tailored to the individual characteristics of each user, and the system disclosed in this application can be used in support settings other than caregiving.

[0024] Furthermore, caregivers input specific points to note and care procedures for the person being cared for into their smartphones. These points are then linked to videos and stored in memory, and a care manual is generated that incorporates the care recipient's own know-how.

[0025] The videos (care manuals) obtained through filming can be viewed by other caregivers. When the care manual is played, individual points to note and procedures for each person receiving care are displayed as text and icons, allowing caregivers to provide care to the person receiving care while checking these points and procedures in real time.

[0026] Furthermore, the care support system disclosed in this application continuously monitors the health status of the person receiving care through wearable devices and IoT (Internet of Things) sensors. If the care support system detects an abnormality in the person receiving care based on input data from wearable devices, etc., it automatically displays an emergency response manual on a smartphone or other device carried by the caregiver.

[0027] Furthermore, caregivers can access necessary care information from the care support system with simple voice commands. For example, if a caregiver says to their smartphone, "Play the manual for assisting with toileting for Ms. B," the manual will play on the smartphone. Caregivers can then access necessary care information (care manuals) via their smartphones.

[0028] Furthermore, the care support system disclosed in this application includes a function to learn from feedback from caregivers and continuously improve the content of the manual. This function enables more user-friendly, personalized care support for caregivers and others.

[0029] <Device Configuration> The embodiment disclosed in this application will be described in more detail with reference to the attached drawings.

[0030] The care support system disclosed in this application includes three components: an "input component," an "analysis component," and an "output component." These three components work together in the care support system. This collaboration reduces the workload of caregivers and enables the creation of individualized care manuals (care manuals tailored to each individual receiving care), anomaly detection, and emergency response.

[0031] The care support system disclosed in this application includes a terminal such as a smartphone or tablet that can be carried by the caregiver, various sensors including wearable devices and IoT devices, and a server device that analyzes the data transmitted from the smartphone and various sensors.

[0032] Figure 3 shows an example of the overall configuration of a care support system according to one embodiment of the present invention.

[0033] The input unit 201, analysis unit 202, and output unit 203 modules are implemented, for example, on a server device 10 in the cloud. The database 204 is configured on the server device 10 or an external server.

[0034] The input unit 201 is a means for acquiring multimodal data in a caregiving setting. Multimodal data includes voice data, text data, video data, and sensor data. For example, the input unit 201 has the function of acquiring voice, text, video, etc. from a smartphone, etc., and acquiring output data (sensor data) from various sensors. As shown in Figure 3, the input unit 201 includes voice input function, text input function, video input function, sensor data input function, etc.

[0035] In this specification, "multimodal data" can also be rephrased as data in multiple formats, diverse types of data, combinations of data with different characteristics, or data from multiple sources.

[0036] The analysis unit 202 is a means for analyzing multimodal data and extracting local knowledge. Local knowledge includes care procedures and points to note specific to the person being cared for. A more detailed explanation of local knowledge will be given later. For example, the analysis unit 202 has the function of analyzing data acquired by the input unit 201. The analysis unit 202 is equipped with an AI (Artificial Intelligence) module and a machine learning engine. The analysis unit 202 has, for example, a scenario generation function and an anomaly detection function.

[0037] The output unit 203 is a means of providing personalized information to the caregiver or care recipient based on local knowledge. For example, the output unit 203 has functions for creating care manuals and sending emergency notifications based on analysis results. The output unit 203 provides a manual presentation UI (User Interface). In addition, the output unit 203 has an emergency notification function and a family / doctor collaboration interface function.

[0038] Database 204 stores the care recipient's personal profile and local knowledge.

[0039] Further details regarding the various functions implemented by the analysis unit 202 and the local knowledge stored in the database 204 will be described later.

[0040] FIG. 4 is a block diagram showing an example of a device configuration (system configuration) including a server device 10 and a terminal 20 such as a smartphone or a tablet.

[0041] A dedicated application is installed on the terminal 20. The server device 10 acquires information (such as the operation or voice of the caregiver) from the terminal 20 or outputs information to the terminal 20 via the dedicated application.

[0042] The terminal 20 such as a smartphone includes a CPU (Central Processing Unit), a memory, a camera, a microphone, a communication module, a display screen (display), etc. The camera functions as a video shooting unit. Also, the microphone functions as an audio input unit. That is, the terminal 20 realizes input means for acquiring an instruction (command) by voice uttered by the caregiver or acquiring a video.

[0043] The server device 10 includes an arithmetic device such as a CPU and a GPU (Graphics Processing Unit), a main memory, a storage (a large-capacity database), a network IF (Interface), an AI model execution environment, etc. The server device 10 realizes analysis means for analyzing the audio data, video data, and sensor data acquired by the terminal 20 and various sensors.

[0044] Furthermore, the server device 10 realizes output means for outputting various manual screens, warning messages, etc. by transmitting the analysis result to the terminal 20 possessed by the caregiver.

[0045] In addition, the care support system includes a group of sensor devices including a wearable terminal 30, an environmental sensor 40, a medical device 50, etc.

[0046] The care recipient or caregiver wears the wearable terminal 30. The wearable terminal 30 includes, for example, an acceleration sensor and a heart rate sensor (heart rate monitor). The environmental sensor 40 is installed at various locations in the care facility. Examples of the environmental sensor 40 include a thermometer, a hygrometer, an illuminometer, a noise meter, etc. Examples of the medical device 50 include a sphygmomanometer, a thermometer, etc. The care support system disclosed in the present application has a mechanism for acquiring the biometric information of the care recipient at any time by means of the wearable terminal 30 or the like.

[0047] <Processing Flow> Figure 5 is a processing flowchart showing the main processing procedures of the entire care support system. Referring to Figure 5, the overall operation of the care support system will be described.

[0048] In step S01, data is acquired from the care site. In step S01, voice input and video shooting are performed.

[0049] In step S02, scene estimation is performed by voice analysis and simple natural language processing. For example, when a caregiver's utterance such as "Mr. A, let's take a bath" is detected, the current care scene is estimated to be a bath assistance scene. When a specific scene is estimated, video shooting starts.

[0050] In step S03, the acquired video and text information are associated, and local knowledge (specific care considerations) is tagged. For example, when the caregiver talks about the considerations in caring for the care recipient, local knowledge is extracted from the utterance. The extracted local knowledge (text information) is stored in association with the acquired video (video playback position; time stamp).

[0051] In this way, in step S03, local knowledge is tagged.

[0052] In step S04, the scenario classification and personal profile are referred to, and a video manual is automatically created (a video reflecting the specific considerations of the care recipient, etc. is created). That is, in step S04, information specific to the care recipient is reflected, and individual optimization of the manual for the caregiver to refer to is performed.

[0053] In step S05, the presence or absence of an abnormality in the person receiving care is detected based on vital data and behavioral patterns. If it is determined that an abnormality has occurred in the person receiving care (step S06, Yes branch), emergency response is performed (step S07). Note that in step S05, vital information is analyzed and predictions are made by an AI model. In step S07, an emergency first aid manual is presented and an emergency notification is issued.

[0054] If it is determined that no abnormality has occurred in the person receiving care (Step S06, No. branch), the emergency response in Step S07 will not be performed.

[0055] In step S08, the learning algorithm updates the content of the care manual based on feedback from caregivers, gradually improving the accuracy of the information provided by the system. Specifically, in step S08, evaluation (receiving feedback) and updating of the learning model take place.

[0056] <Data Acquisition Method> Next, we will explain the data (multimodal data) acquired by the input unit 201.

[0057] Input data in caregiving settings can be broadly classified into four types: audio data, video data, sensor data, and text data.

[0058] The audio data includes statements made by caregivers at the start of their duties, such as "I will begin assisting with meals" and "I will now assist with bathing," as well as communication with the person receiving care and handover information between caregivers.

[0059] Video data includes videos recording the implementation of care procedures, videos capturing the facial expressions and movements of the person receiving care, and surveillance camera footage recording the person's behavior patterns within their room.

[0060] The sensor data includes vital data such as heart rate obtained from a wearable device 30 worn by the person receiving care, and blood pressure and body temperature obtained from a medical device 50. Furthermore, the sensor data includes sitting and standing status obtained from pressure sensors installed on the bed and chair, environmental data (temperature, humidity, illuminance, etc.) obtained from IoT sensors (environmental sensors 40) placed in the room, and the movement trajectory of the person receiving care obtained from a position detection sensor.

[0061] The text data includes quantitative information such as food intake, fluid intake, and excretion status entered as care records, as well as medication management information, rehabilitation progress, requests from family members, and instructions from medical institutions.

[0062] Audio data, video data, and text data are input into the system in real time via terminals 20 such as smartphones, tablets, and smart glasses, and are used for automatic recognition of caregiving scenes, anomaly detection, and generation of personalized caregiving manuals.

[0063] Furthermore, sensor data is transmitted in real time to the server device 10 via the wireless LAN (Local Area Network) environment within the nursing care facility and various sensor networks. Upon receiving the sensor data, the server device 10 performs integrated data management and analysis processing.

[0064] Furthermore, feedback from caregivers, such as suggestions for improving care manuals, additions of new points to note, and records of emergency responses, can be entered into the system as needed, and this information may be used for the system's continuous learning and updating.

[0065] Thus, the input unit 201 includes a voice input unit for acquiring the caregiver's speech, a text input unit for acquiring care records, a video recording unit for capturing care scenes, and a sensor input unit for acquiring the care recipient's biometric information.

[0066] <Specific Processing Method of Analysis> Next, the processing content of the analysis unit 202 will be explained in more detail.

[0067] The analysis unit 202 extracts local knowledge from the input data acquired by the input unit 201.

[0068] Here, "local knowledge" refers to information such as the individual characteristics and preferences of the person receiving care, situations requiring special attention, or subtle know-how based on the caregiver's experience. Local knowledge is a general term encompassing the individual characteristics, preferences, and know-how of each person receiving care.

[0069] In this specification, "local knowledge" can also be rephrased as specific knowledge, on-site insights, situation-dependent information, individual-specific knowledge, or rules of thumb in a particular environment. Furthermore, local knowledge can also be rephrased as support information, user information, care recipient information, patient information, or information on persons requiring support.

[0070] For example, local knowledge includes points that require individual attention, such as "It's easier for Person A to hear if you speak to them from the left side," "Person B is prone to aspiration, so use a shallower angle for the spoon," and "Person C, who has dementia, strongly resists being touched suddenly when transferring them to a wheelchair, so check how they are feeling beforehand."

[0071] The analysis unit 202 extracts local knowledge from the input data. The extracted local knowledge is stored in the local knowledge table. The analysis unit 202 performs specific processing for each type of input data with the aim of building a system for systematically managing the extracted local knowledge.

[0072] Specifically, if the input data is audio data, the analysis unit 202 uses natural language processing to extract keywords and detect important terms such as "verbal cues during bathing assistance" and "rejection reactions" from the input data. Furthermore, the analysis unit 202 matches the detected important terms with an existing local knowledge table and registers them as new points of caution, or updates the links in the local knowledge table to reinforce existing knowledge.

[0073] If the input data is text data, the analysis unit 202 extracts key points through document clustering and syntactic analysis. For example, the analysis unit 202 finds specific know-how such as "timing of swallowing during meal assistance" and "posture during excretion assistance." The analysis unit 202 assigns tags to the text data according to its content and organizes it on a per-care recipient basis (per care recipient ID).

[0074] If the input data is video data, the analysis unit 202 performs facial recognition and motion recognition. For example, the analysis unit 202 extracts information such as "Person A feels safe when their waist is supported" as local knowledge based on video footage showing the optimal positional relationship between the caregiver and the person being cared for when the person is bathing, or situations in which the person being cared for is likely to refuse.

[0075] If the input data is sensor data, the analysis unit 202 analyzes vital signs, acceleration sensor information, environmental data, etc., in a time series. For example, the analysis unit 202 detects patterns such as "a tendency to become irritable when the pulse rate exceeds a certain value" using a machine learning model. The analysis unit 202 adds the detected patterns to the individual profile. Through this operation of the analysis unit 202, the care support system can make specific suggestions in real time, such as "the timing of the caregiver's calls needs to be adjusted."

[0076] The analysis unit 202 associates the analysis results described above with the care recipient's information in the local knowledge table, for example, so that the following configurations are possible (so that the following functions can be realized).

[0077] (1) Tagging search function: For example, when a caregiver enters a phrase such as "Ms. A refused to eat" into the system using voice commands or text search, relevant know-how is listed and presented to the caregiver.

[0078] (2) Version control function: For example, based on feedback from caregivers or new findings, "updated versions of aspiration prevention measures" can be added as local knowledge.

[0079] (3) Automatic annotation function: For example, based on the actions detected by video analysis, a tag such as "assistance needed when stepping over a step" is automatically added.

[0080] The local knowledge accumulated in this way is immediately presented as an optimized care manual, referencing the profile of the person receiving care and their vital signs on that day. Therefore, local knowledge functions not merely as a static record, but as a dynamic and individualized information asset.

[0081] Furthermore, even when multiple facilities or home care settings collaborate to share data, the "core" of local knowledge corresponding to the care recipient's ID and the variations at each setting are managed separately, enabling both data generalization and individualization.

[0082] Furthermore, by introducing reinforcement learning and explainable AI (Artificial Intelligence), the system (server device 10) can learn on its own what assistance procedure is optimal. For example, it is expected that the system will extract new knowledge (local knowledge) such as, "In Mr. / Ms. B's case, increasing the number of verbal cues compared to before will allow them to sit down more smoothly."

[0083] Furthermore, the analysis unit 202 may also have functions for statistically analyzing data from multiple care recipients, in addition to analyzing data from individual care recipients, and for grouping and analyzing care recipients with similar characteristics. The analysis unit 202 may provide not only individualized information, but also group-based information, generalized recommendations using a statistical approach, and stepwise information based on the caregiver's skill level. For example, the analysis unit 202 may use machine learning to analyze the similarities of care recipients and extract common characteristics and trends to achieve more efficient information provision.

[0084] These advanced analytical flows allow local knowledge to move beyond mere notes of rules of thumb and be managed as an integrated and reusable data infrastructure.

[0085] <Data Structure of Analysis Results> Here, we will explain the data structure of the analysis results again, referring to the diagram.

[0086] Figure 6 is a schematic diagram illustrating an example of a data structure that shows how the personal profiles and analysis results of care recipients are managed as data.

[0087] In the embodiment disclosed herein, data managed by the system (server device 10) is divided and stored in a "personal profile DB" and a "local knowledge table". Furthermore, the system manages care scenes using a "care scene table".

[0088] The former, the personal profile database, stores information such as the care recipient's care recipient ID (user ID), name, age, level of care needed, medical history, food preferences, and communication preferences (communication characteristics).

[0089] The latter local knowledge table stores information such as a knowledge ID to identify the local knowledge, a care recipient ID, "notes to be aware of," "related scene ID," registration date and time, update date and time, and importance level. For example, a specific note such as "Person A is likely to refuse if the spoon is not offered at a 45-degree angle" is linked to the care recipient ID in the local knowledge table.

[0090] The care scene table stores information such as the scene ID (related scene ID), scene name, video path, timestamp, tag information, and privacy level.

[0091] Through this linking, the analysis means (analysis unit 202) can refer to tags linked to video data, which can then be used in subsequent personalization processing. The care recipient ID recorded in the personal profile DB identifies the content of precautions and other information recorded in the local knowledge table. In addition, the related scene ID recorded in the local knowledge table identifies the video path (location where the video is stored) and other information recorded in the care scene table.

[0092] <Operation of the Analysis Unit> The analysis unit 202 acquires input data from the input unit 201. The analysis unit 202 estimates a caregiving scene from the content of the speech acquired by the voice input unit of the input unit 201. The analysis unit 202 functions as a means for estimating a caregiving scene.

[0093] The analysis unit 202 controls the start or end of video recording according to the estimation results from the care scene estimation means. The analysis unit 202 also functions as a shooting control means that instructs the terminal 20 to start or stop video recording.

[0094] Furthermore, the analysis unit 202 associates the video data and text data acquired by the input unit 201 from the terminal 20. The analysis unit 202 extracts local knowledge from the text data, audio data, etc., and associates the extracted local knowledge with the video data. In other words, the analysis unit 202 functions as an information management means that tags and manages local knowledge with video data. At that time, the analysis unit 202 (information management means) stores the local knowledge in a local knowledge table, associating it with the care recipient ID. In addition, the analysis unit 202 (information management unit) may be able to update the local knowledge based on feedback from multiple caregivers.

[0095] Furthermore, the analysis unit 202 divides the video data by caregiving scene (scenario) and refers to the care recipient's personal profile to generate a video manual specifically for that care recipient. In other words, the analysis unit 202 has the function of analyzing the video data, classifying it by caregiving scenario, and automatically generating a caregiving manual in storyboard format.

[0096] Furthermore, the analysis unit 202 can extract local knowledge not only from text data but also from other data. For example, the analysis unit 202 has the function of an action recognition means that recognizes caregiving actions performed by a caregiver from video acquired by the video recording unit of the input unit 201.

[0097] Alternatively, the analysis unit 202 may determine the health status of the person receiving care from the biological information (vital data) acquired by the sensor input unit of the input unit 201. In other words, the analysis unit 202 may also function as a means for determining the state. Furthermore, the analysis unit 202 may adjust the care manual using the estimated health status. For example, the analysis unit 202 may generate a care manual specifically for the person receiving care by associating the health status (local knowledge) extracted based on the vital data with video data.

[0098] <Screen Example> Figure 7 shows an example of the display screen of a terminal 20 such as a smartphone, output by the output unit 203.

[0099] The terminal 20 disclosed in this application has a video playback area 301 on its user interface screen, overlaying text and icons. For example, pop-ups containing local knowledge, such as "Be careful, there is a 5cm step here" or "Offer the spoon at a 45-degree angle," are overlaid in accordance with the corresponding timeline of the video. Such pop-up displays allow caregivers to grasp individual precautions in real time.

[0100] Thus, the output unit 203 is equipped with a display means that superimposes points of attention specific to the person being cared for onto the video as text overlays or icons.

[0101] Furthermore, the care support system disclosed in this application (output unit 203 of the server device 10) is equipped with a multilingual translation function, and by switching the display language or generating automatic subtitles, it can be made easy for foreign staff and family members to use. Specifically, the output unit 203 may be equipped with translation means that perform automatic subtitle generation and multilingual translation. The translation means translates the contents of the care manual into multiple languages. Alternatively, the translation means outputs audio guidance in multiple languages. Alternatively, the translation means overlays multilingual subtitles on the video.

[0102] In addition, the care support system disclosed in this application (output unit 203 of the server device 10) may also be equipped with an interactive assistant function. By using this interactive assistant function, caregivers can perform operations (instructions to the system) such as "Please play the manual for assisting with B's excretion" using voice input.

[0103] Furthermore, the manner in which the output unit 203 presents local knowledge to the caregiver is not limited to the example shown in Figure 7.

[0104] For example, the output unit 203 may output local knowledge via a voice assistant interface. Specifically, the output unit 203 may provide the caregiver with real-time voice guidance via earphones or a smart speaker, such as, "Now is the time to transfer A to the wheelchair. Supporting A's left leg will help stabilize them." In other words, the output unit 203 may function as a voice guidance means that provides voice guidance on care procedures.

[0105] Furthermore, the output unit 203 may transmit information to the caregiver via an augmented reality (AR) device. Specifically, the output unit 203 may overlay icons and other elements onto the image displayed on smart glasses or an AR-compatible tablet. The output unit 203 may also highlight the care recipient's body parts and risk areas (such as steps or handrail locations) and display individual precautions as pop-ups.

[0106] Furthermore, the output unit 203 may also be equipped with an automatic notification message function. Specifically, the output unit 203 may work in conjunction with a schedule management system to send push notifications to a terminal 20 such as a smartphone or wearable device, such as "It's time to talk to Person A before breakfast" or "Person B often dislikes taking medicine, so please add an explanation."

[0107] Furthermore, the output unit 203 may provide information using an interactive chatbot. Specifically, if a caregiver inputs text such as "What should I be careful about when assisting Ms. B with toileting?", the output unit 203 may respond with relevant local knowledge in text form, such as "You need to place a waterproof sheet first to avoid accidentally wetting her clothes."

[0108] Furthermore, the output unit 203 may print a simplified checklist. Alternatively, the output unit 203 may provide a digital simplified checklist to caregivers, etc. Specifically, the output unit 203 may output local knowledge extracted from the local knowledge table in a list format. By posting the local knowledge compiled in this list format at the bedside or on the facility's work board, new staff and busy caregivers can quickly refer to the local knowledge.

[0109] Furthermore, the output unit 203 may also be equipped with a periodic reporting function. Specifically, when generating weekly and monthly review reports, the output unit 203 may highlight and include newly accumulated local knowledge and updated manuals. Such a periodic reporting function makes it easier for caregivers to refer to local knowledge in the next care setting.

[0110] Furthermore, the output unit 203 may also be equipped with a function for coordinating with a care robot (coordination display). Specifically, the care robot may be equipped with a built-in touch panel and speech function. The output unit 203 may display instructions on the robot's screen or voice during transfer assistance or rehabilitation assistance by the caregiver, such as "Mr. / Ms. A has joint pain, so please support their waist when they stand up."

[0111] The output unit 203 may also have functions for linking with other caregiving equipment. Specifically, the output unit 203 may link with caregiving equipment such as caregiving lifts and electric beds, and manage the usage history and setting information of each device in association with local knowledge. For example, the output unit 203 automatically adjusts the settings of the caregiving equipment based on local knowledge. This automatic adjustment enables optimal equipment settings according to the physical condition and preferences of the person receiving care.

[0112] The output unit 203 may also have a smart home connectivity function. Specifically, the output unit 203 may connect with IoT devices such as lighting, air conditioning, hot water supply equipment, and automatic doors installed in nursing facilities or home care environments. The output unit 203 automatically optimizes the environmental settings of these devices based on the care recipient's local knowledge. For example, the output unit 203 adjusts the room temperature according to the care recipient's physical condition and time of day, controls lighting when they move, and automatically sets the hot water temperature when they bathe. Such smart home connectivity improves the comfort and safety of the care recipient and reduces the workload of the caregiver.

[0113] Furthermore, the output unit 203 may also be equipped with a vital alert linkage function. Specifically, if the sensor detects an abnormal value, the output unit 203 may dynamically suggest local knowledge such as, "Person B tends to experience sudden increases in blood pressure, so speak to them in a calm tone," or "Person C becomes unstable when the risk of falling increases, so increase the frequency of calling out to them."

[0114] Furthermore, the output unit 203 may also be equipped with a practical training support function. Specifically, the output unit 203 may output VR (Virtual Reality) / AR (Augmented Reality) teaching materials during training within the facility. Through the operation of such an output unit 203, local knowledge explained by veteran caregivers is newly embedded into videos and simulations, allowing trainees to repeatedly learn the know-how of veteran caregivers.

[0115] Furthermore, the output unit 203 may also be equipped with a face-to-face instruction support function. Specifically, when an experienced caregiver is instructing a newcomer, the output unit 203 outputs a "local knowledge map" to a terminal 20 such as a tablet. The experienced caregiver then refers to the output results and explains, "There have been many falls in this area in the past, so the timing of taking Ms. A's hand is important." In this way, information can be shared in a visualized form, rather than relying solely on verbal communication.

[0116] As described above, the care support system disclosed in this application can flexibly select and switch the presentation of local knowledge according to the user's skill level and situation. As a result, even with a large amount of information, the minimum essential information is transmitted in real time without omission, and continuous updates are reflected to maintain the latest care know-how at all times.

[0117] As described above, the care support system according to the first embodiment streamlines recording at the care site through multimodal input (voice, text, video, sensor data), and enables tagging of local knowledge and learning of behavioral patterns through analysis means. Furthermore, personalized care manuals and emergency response guides are presented to caregivers via output means, thereby realizing care support that is both individualized and real-time. In other words, the care support system according to the embodiment disclosed in this application significantly reduces the burden on caregivers with the above configuration, while simultaneously enabling meticulous care for those receiving care. That is, the care support system functions as an extremely useful system in home care and facility care settings in an aging society.

[0118] <Modified Version> The care support system disclosed in this application can also be used for anomaly detection and emergency response.

[0119] Figure 8 shows a flowchart for anomaly detection through analysis of vital data and behavioral information, as well as emergency response procedures.

[0120] The sensor transmits vital data of the person receiving care to the care support system (step S11).

[0121] The care support system (server device 10) uses an AI analysis engine to detect and determine abnormalities (step S12).

[0122] For example, the server device 10 makes an "anomaly detection request" to the AI ​​analysis engine and obtains anomaly detection results from the AI ​​analysis engine. More specifically, the analysis unit 202 detects anomalies in the person being cared for using sensor data and video data. The analysis unit 202 obtains anomaly detection results by inputting the sensor data and video data into the AI ​​analysis engine. In this way, the analysis unit 202 functions as an anomaly detection means.

[0123] If the care support system determines that an abnormality has occurred in the person being cared for, based on the abnormality detection result (degree of risk to the person being cared for), it sends an emergency notification to the terminal 20 held by the caregiver (step S13).

[0124] Furthermore, the care support system (analysis unit 202) displays the first aid manual on the terminal 20 (step S14). More specifically, the analysis unit 202 (anomaly detection means) calculates a risk level that quantifies the abnormal condition of the person receiving care, and if the risk level exceeds a preset threshold, it automatically retrieves the emergency response manual (first aid manual) and displays it on the terminal 20.

[0125] Furthermore, the care support system (analysis unit 202) makes an emergency contact with medical professionals (medical institutions) (step S15).

[0126] Furthermore, the care support system (analysis unit 202) notifies the family of the person receiving care of the situation if an abnormality has been detected (step S16).

[0127] Thus, the analysis unit 202 is equipped with an emergency notification function that notifies at least one of the caregiver, the care recipient's family, and medical professionals when an abnormality occurs in the person receiving care.

[0128] Once the caregiver has completed first aid, they operate terminal 20 to report this. Terminal 20 then sends a completion report to the care support system (step S17).

[0129] The care support system records the completion report (record the report; step S18).

[0130] In this way, the care support system calculates the risk level of the person receiving care based on sensor data and video analysis results. If the calculated risk level exceeds a preset threshold, the care support system automatically retrieves the emergency manual and displays first aid procedures on the display screen of terminal 20. Furthermore, along with displaying the first aid procedures, the care support system automatically contacts family members and medical professionals.

[0131] For example, if a fall is detected in the person being cared for, the server device 10 recognizes the person's fall motion through video analysis. The server device 10 then sends an alert to the attending physician or family via an emergency notification interface.

[0132] Furthermore, the emergency response manual includes videos illustrating procedures for responding to falls, checking the level of consciousness, and contacting the emergency services provider, enabling caregivers to take swift and appropriate action.

[0133] In this modified version, the server device 10 utilizes the functions of the analysis unit 202 for anomaly detection, thereby enabling 24-hour monitoring and rapid response in emergencies, as well as improving safety in caregiving settings.

[0134] <Typical Use Cases> Next, we will describe use cases illustrating how the care support system according to the embodiments of the present disclosure, including modified examples, operates.

[0135] For example, consider the case of person A, who has a high level of care needs and requires assistance with bathing. The caregiver carries a device 20 such as a smartphone and says to person A, "Person A, it's time to take a bath." In response to this call, the input means captures the utterance, and the analysis means estimates the care scene of bathing from the utterance. The analysis means automatically starts recording a video using the video recording unit installed in the device 20 (the analysis means instructs the device 20 to record a video).

[0136] The video (video information and audio information) captured by terminal 20 is transmitted to server device 10, and analysis unit 202 (analysis unit 202 as an information management means) generates a bathing manual incorporating local knowledge while referring to Mr. A's personal profile.

[0137] For example, if local knowledge records that Person A has weak knee joints and difficulty stepping over the bathtub, then individual precautions such as "Bend the knees slightly to provide support" and "Speak to them a little louder" will be superimposed as text overlays on the video screen of terminal 20. This superimposed display allows caregivers to obtain guidance for assistance in real time.

[0138] Furthermore, if a caregiver of foreign nationality is providing care (support) to the person being cared for, the output unit 203 (output unit 203 as a translation means) may perform automatic subtitling or multilingual display. As a result, problems caused by language barriers can be avoided.

[0139] Furthermore, if A's heart rate or blood pressure shows abnormal values ​​while bathing, the analysis unit 202 (analysis unit 202 as an emergency response measure) will be activated.

[0140] Alternatively, the analysis unit 202 (analysis unit 202 as an anomaly prediction means) uses a machine learning model to determine that Mr. A is at high risk of falling. In this case, it makes an emergency call to staff and family members and displays first aid procedures on the screen of terminal 20.

[0141] More specifically, the analysis unit 202 (anomaly prediction means) analyzes the care recipient's biometric information and behavioral patterns using a machine learning model to predict risks, including the care recipient's risk of falling and deterioration of their health. If a risk is predicted, the analysis unit 202 presents the caregiver with a care manual or warning that includes points to note (local knowledge) corresponding to the predicted risk.

[0142] The analysis unit 202 may also store the risk prediction results as local knowledge.

[0143] Caregivers (care staff, etc.) may input feedback regarding the abnormal situation into the system. For example, a caregiver may input a new point of caution into the system, such as "I should have spoken to them a little earlier." In response to this input, the learning algorithm (analysis unit 202) may update the local knowledge, thereby updating the care manual.

[0144] Thus, the analysis unit 202 may also function as a feedback input means for receiving feedback from caregivers and a learning means for updating local knowledge based on the feedback. The learning means may perform version control of the local knowledge and record the update history of the local knowledge.

[0145] This operation by the analysis unit 202 allows for improvements to the care manual, enabling more prompt verbal guidance and warnings in similar situations in the future.

[0146] Thus, the care support system disclosed in this application provides comprehensive care support by constantly monitoring Mr. A's individual condition, presenting care manuals to caregivers in real time and dynamically optimized, and covering emergency response in the event of an abnormality.

[0147] <Other Use Cases> Furthermore, the care support system disclosed in this application can also be used as a foundation for realizing the following business models. Specifically, by having an analysis means comprehensively analyze voice data, text data, video data, sensor data, etc. acquired by the input means disclosed in this application, and providing the results to local governments, companies, medical institutions, etc. via an output means, various forms of services and solutions can be constructed.

[0148] For example, one possible model is to build an AI-powered preventive medical care and nursing support platform as a cloud-based subscription service. In this case, an analysis unit (analysis unit 202, etc.) collects and learns individual biometric data, medical and nursing care history data, and lifestyle data, predicts abnormal signs in real time, and sends notifications encouraging early medical consultation to local governments and insurers via an output unit (output unit 203, etc.). In this way, a system that supports the prevention of serious illness and health promotion is provided.

[0149] Alternatively, a configuration is envisioned in which communication technologies such as 5G (5th Generation) / 6G (6th Generation) are combined with a cloud analysis platform on the server device 10 to provide a remote medical and nursing care network solution. Specifically, high-resolution images and vital data acquired by the analysis means can be immediately transmitted to specialists and nursing care staff, and remote medical consultations and remote monitoring can be performed through operation and calls from the terminal 20.

[0150] Furthermore, by utilizing the care support system disclosed in this application, a support service for foreign care workers combining multilingual support and biosensing can be provided. In this case, the input means acquires biometric information of staff and elderly people from wearable terminals 30, etc., which is then evaluated in real time by the analysis means. The output means, equipped with a multilingual translation function, supports smooth communication between caregivers and those receiving care, and also makes it possible to visualize emotions and stress levels to prevent accidents and troubles.

[0151] Alternatively, it is conceivable to provide a data utilization platform and Evidence-Based Policy Making (EBPM) services for local governments, utilizing the care support system disclosed in this application. The system disclosed in this application collects and analyzes regional medical and care data, which is stored in an integrated database on the server device 10. The analysis means calculates demand forecasts and facility placement plans, and the output means provides these to local governments and medical institutions, thereby enabling the provision of a model that supports evidence-based policy decisions.

[0152] Furthermore, as an integrated smart home and in-home care robot solution, IoT devices and care robots may be linked with the system disclosed herein, with the analysis means automating anomaly detection and transfer assistance based on sensor information, and the output means providing notifications and monitoring information to family members or specialists in remote locations.

[0153] Thus, the care support system disclosed in this application functions as a platform that integrates voice input, video analysis, and biosensor information management, and can easily realize service models tailored to various industries and markets (such as preventive medicine, telemedicine, multilingual care support, EBPM for local governments, and home robot integration).

[0154] [Second Embodiment] Next, a second embodiment will be described in detail with reference to the drawings.

[0155] In the second embodiment, we describe an example in which local knowledge is jointly utilized, rewritten, and modified by multiple caregivers and management staff.

[0156] In the second embodiment, as in the first embodiment, a personalized care manual is created by utilizing multimodal information obtained from voice input, video recording, sensor data, etc.

[0157] In the second embodiment, we describe a care support system equipped with enhanced management functions that ensures local knowledge is continuously passed on even when caregivers are replaced or new caregivers are added, and that all stakeholders can refer to the same content when local knowledge is modified.

[0158] Specifically, the server device 10 is equipped with a local knowledge table as well as a version control function.

[0159] The caregiver, as the user, is able to access the server device 10 from multiple devices within the facility (e.g., smartphones, tablets, PCs, etc.). For example, when a new caregiver assists person A with toileting for the first time, local knowledge tagged with "Person A always needs support from the left side when transferring to the toilet seat" is displayed on the caregiver's smartphone screen.

[0160] If a new staff member, after actually implementing the precautions, notices that "Person A has a weak core and should have more support around their waist," the new caregiver can immediately add information to their local knowledge or revise existing precautions through the system's feedback input screen.

[0161] When a modification is made, the version control function automatically updates the version of Person A's local knowledge, such as "Person A_Excretion Assistance_V2," and records who (which user) made the modification and when. Furthermore, when other caregivers access Person A's local knowledge, the latest modified version is displayed first.

[0162] Next, we will describe an example of an addition in the second embodiment.

[0163] <Caregiver Skill Evaluation Function> Furthermore, the care support system according to the second embodiment may include a function in which the AI ​​evaluates the care actions performed by each caregiver, including new caregivers (new caregivers), and suggests areas for improvement.

[0164] For example, after a caregiver performs tasks such as toileting assistance or bathing assistance on a person receiving care, while referring to local knowledge, the care support system (server device 10) comprehensively analyzes video, audio, sensor data, etc., and calculates a skill evaluation score from perspectives such as the time required for care, the person receiving care's reactions, and the accuracy of operations. For example, the server device 10 may assign scores for each type of care, such as "toileting assistance: 4", "bathing assistance: 3", and "meal assistance: 4", and output the results of comparison with the average value and points for improvement as comments.

[0165] The results of the skill evaluation may be presented as a list screen for each caregiver. For example, a message such as, "You are taking longer than average to assist with bathing. Adjusting the timing of verbal cues in the bathroom may help make the process smoother," may be automatically generated and displayed. This system operation allows caregivers to objectively reflect on their own performance, and also allows senior staff and managers to easily understand each staff member's areas of weakness.

[0166] <Skill Visualization and Guidance Support Function> In addition, the care support system according to the second embodiment may also include a guidance support function that allows senior staff and managers to provide appropriate advice while reviewing care videos of new staff members, in addition to the skill evaluation function.

[0167] Specifically, a comment writing function will be added to the feedback input screen and video playback screen, allowing senior staff to guide new staff by visualizing key points such as "It would be safer if you supported their waist a little more here" or "The interaction was delayed, which is likely to cause a negative reaction."

[0168] New staff members can refer to these suggestions and immediately check the relevant sections of the manual or register them as new local knowledge.

[0169] <Screen Example> Next, we will explain the feedback submission screen using Figure 9.

[0170] As shown in Figure 9, the feedback input screen of the care support system displays the target information of the manual (for example, "Ms. A_Meal Assistance Manual_V2") and its last update date at the top of the screen, and buttons at the bottom of the screen allow the user to select the type of feedback from "Improvement Suggestion," "Addition of Points to Note," and "Correction Request."

[0171] Feedback can be entered in a free-form text field in the central input area. For example, a user can write a specific suggestion for improvement in the input area in the center of the screen, such as, "Regarding the angle of the spoon, it is currently 45 degrees, but recent observations suggest that it is easier to swallow at around 30 degrees."

[0172] Furthermore, buttons for attaching images and videos, as well as a microphone button for voice input, are provided at the bottom of the screen. This allows users to attach not only text input, but also videos and audio explanations of actual caregiving situations.

[0173] The entered information is registered in the system when the "Send Feedback" button is pressed, and the manual's content is automatically updated through a learning algorithm by the analysis means (analysis unit 202).

[0174] Furthermore, if "Request for Correction" is selected as the type of feedback, a note at the bottom of the screen may indicate that it will be processed with priority due to its urgency.

[0175] Thus, the care support system disclosed in this application is configured to quickly share requested modifications with system administrators and other caregivers. Furthermore, the care support system disclosed in this application provides a mechanism that allows users to easily input and share observations and areas for improvement in the care setting, thereby supporting the continuous improvement of care manuals.

[0176] As described above, in the second embodiment, information is updated and shared among multiple staff members (updating and sharing takes place in real time). As a result, local knowledge, including tacit knowledge initially entered by veteran caregivers, is constantly updated and refined by new insights and the experiences of different staff members, enabling the same quality of care service to be provided regardless of who is in charge of caring for the care recipient.

[0177] Furthermore, when multiple facilities use the same platform, caregivers at each facility can access local knowledge across facilities using the user ID (care recipient ID) as a key.

[0178] Thus, the analysis unit 202 functions as a knowledge sharing means for sharing local knowledge among multiple care facilities. The analysis unit 202 (knowledge sharing means) makes it possible to access the local knowledge across facilities based on the care recipient ID.

[0179] On the other hand, it may be possible to set permissions in accordance with the personal information protection regulations and internal rules of each facility, and caregivers who have moved to another facility may be able to extract and utilize only the necessary information. In other words, the analysis unit 202 may have a permission setting function that sets access permissions to local knowledge and care manuals based on personal information protection and facility-specific rules. In other words, the analysis unit 202 may perform privacy control.

[0180] As explained above, in the care support system according to the second embodiment, multiple users can access the same system and manage it while sharing and modifying local knowledge. Therefore, it is possible to maintain a care manual that always reflects the latest situation while preventing the discontinuation of know-how due to changes in caregivers.

[0181] Furthermore, the care support system according to the second embodiment can facilitate the smooth sharing of knowledge among multiple caregivers, which will be particularly important in future home care. As the aging society progresses and the shortage of caregivers becomes more serious, multiple caregivers will need to take turns providing 24-hour care in home care. In this regard, it is extremely important to reliably pass on the care procedures and precautions specific to the person being cared for in order to provide safe and high-quality care services.

[0182] The care support system disclosed in this application records and shares knowledge gained at care sites in real time via terminals such as smartphones and tablets 20. Furthermore, by accumulating systematic knowledge through AI analysis, the care support system makes it possible to always provide optimal care procedures regardless of the caregiver's experience or skill level. This system operation prevents information leaks and errors during caregiver shifts, enabling consistent care services 24 hours a day. Moreover, even in areas where securing caregivers is difficult, the care support system disclosed in this application is expected to allow even inexperienced caregivers to work with peace of mind, contributing to the resolution of the shortage of care personnel.

[0183] Next, we will describe the hardware of each device that makes up the information processing system. Figure 10 shows an example of the hardware configuration of the server device 10.

[0184] The server device 10 can be configured using a so-called computer and has the configuration illustrated in Figure 10. For example, the server device 10 includes a processor 311, memory 312, input / output interface 313, and communication interface 314, etc. The components of the processor 311, etc. are connected by an internal bus or the like and are configured to communicate with each other.

[0185] However, the configuration shown in Figure 10 is not intended to limit the hardware configuration of the server device 10. The server device 10 may include hardware not shown, and it may not have to have an input / output interface 313 if necessary. Also, the number of processors 311 etc. included in the server device 10 is not intended to be limited to the example in Figure 10; for example, multiple processors 311 may be included in the server device 10.

[0186] The processor 311 is a programmable device such as a CPU (Central Processing Unit), MPU (Micro Processing Unit), DSP (Digital Signal Processor), TPU (Tensor Processing Unit), or GPU (Graphics Processing Unit). Alternatively, the processor 311 may be a device such as an FPGA (Field Programmable Gate Array) or ASIC (Application Specific Integrated Circuit). The processor 311 executes various programs, including an operating system (OS).

[0187] Memory 312 can be RAM (Random Access Memory), ROM (Read Only Memory), HDD (Hard Disk Drive), SSD (Solid State Drive), etc. Memory 312 stores the OS program, application programs, and various data.

[0188] The input / output interface 313 is an interface for a display device or input device (not shown). The display device is, for example, a liquid crystal display. The input device is, for example, a device that accepts user input such as a keyboard or mouse.

[0189] The communication interface 314 is a circuit, module, etc., that communicates with other devices. For example, the communication interface 314 may include a NIC (Network Interface Card).

[0190] The functions of the server device 10 are realized by various processing modules. These processing modules are realized, for example, by the processor 311 executing a program stored in the memory 312. The program can also be recorded on a computer-readable storage medium. The storage medium can be a non-transitory material such as semiconductor memory, hard disk, magnetic recording medium, or optical recording medium. In other words, the present invention can also be embodied as a computer program product. Furthermore, the program can be downloaded via a network or updated using the storage medium on which the program is stored. Moreover, the processing module may be realized by a semiconductor chip.

[0191] In addition, terminal 20 can also be configured to include a computer, similar to server device 10, and its basic hardware configuration is no different from that of server device 10, so a detailed explanation will be omitted.

[0192] The server device 10 is equipped with a computer, and its functions can be realized by having the computer execute a program. Furthermore, the server device 10 executes control methods and information processing methods for the server device 10 using this program.

[0193] [Modification 1] As a modification of the second embodiment, the care support system may include an incentive management function based on the contribution of local knowledge. Specifically, the analysis unit 202 includes a contribution evaluation function that detects activities such as the recording of new local knowledge by caregivers, beneficial editing and updating of existing local knowledge, or referencing such local knowledge by other caregivers, and awards points to caregivers according to predetermined criteria in accordance with these activities.

[0194] Furthermore, the analysis unit 202 periodically analyzes the correlation between the frequency of use of local knowledge and the quality of care (for example, the satisfaction level of the person receiving care and the degree of improvement in their health), and if a positive correlation is confirmed, it has a correlation analysis function that awards additional points to the caregiver who recorded the local knowledge.

[0195] The output unit 203 provides a dashboard that visualizes each caregiver's contribution level and point acquisition history, and also notifies them of the status of the granting of rewards (monetary rewards, special leave, training opportunities, etc.) based on accumulated points. Furthermore, the output unit 203 publishes point rankings and implements motivation-enhancing measures such as awarding top performers. System administrators and facility managers can evaluate each caregiver's contribution level through this dashboard.

[0196] This incentive management function is expected to promote the voluntary sharing of knowledge among caregivers, leading to a continuous improvement in the quality of care.

[0197] Furthermore, the analysis unit 202 may also have a function to quantitatively evaluate the value of local knowledge in relation to the incentive management function described above. For example, the analysis unit 202 comprehensively analyzes indicators such as the number of times each piece of local knowledge has been referenced after recording, the frequency of its use in actual care settings, evaluations from other caregivers, and its contribution to improving the condition of the person receiving care, and calculates a usefulness score for the local knowledge. Caregivers who record local knowledge with a high score are awarded more points.

[0198] Such quantitative evaluation mechanisms promote the accumulation of more practical and valuable local knowledge.

[0199] [Modification 2] Note that the configuration and operation of the information processing system described in the above embodiment are illustrative examples and are not intended to limit the system configuration.

[0200] The above embodiment describes a case where a care manual is provided to the caregiver. However, the care manual may also be provided to the person receiving care. The person receiving care may use their own care manual to improve their life, etc. In this case, the analysis unit 202 of the server device 10 may adjust the length of the video (care manual) and the difficulty level of the explanation according to the degree of cognitive decline and the visual and auditory condition of the person receiving care. The analysis unit 202 may detect cognitive decline and the visual and auditory condition using video data, sensor data, etc. The analysis unit 202 may also have a function as an optimization means (individualization means) that optimizes the care manual according to the condition of the person receiving care, etc.

[0201] In the flowcharts (sequence diagrams) used in the above description, multiple processes are shown in order, but the execution order of the processes performed in the embodiment is not limited to the order in which they are shown. In the embodiment, the order of the illustrated processes can be changed to the extent that it does not impede the content, for example, by executing each process in parallel.

[0202] The embodiments described above are explained in detail to facilitate understanding of the disclosure, and it is not intended that all the configurations described above are necessary. Furthermore, when multiple embodiments are described, each embodiment may be used individually or in combination. For example, it is possible to replace parts of the configuration of one embodiment with those of another embodiment, or to add configurations from other embodiments to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of one embodiment with those of another.

[0203] As described above, the industrial applicability of the present invention is clear, and it is particularly suitable for application to care support systems that provide users with care manuals and the like.

[0204] Some or all of the above embodiments may also be described as follows, but are not limited to the following:

[0205] [Note 1] A care support system comprising: an input means for acquiring data in multiple formats at a care setting; an analysis means for analyzing the data in multiple formats and extracting support information; and an output means for outputting information to the caregiver or the person being cared for based on the support information, wherein the support information includes care procedures and points to note specific to the person being cared for.

[0206] [Note 2] The care support system according to Note 1, wherein the input means includes a voice input unit for acquiring the caregiver's speech, a text input unit for acquiring care records, a video recording unit for filming care scenes, and a sensor input unit for acquiring the care recipient's biometric information, and the analysis means includes a care scene estimation means for estimating a care scene from the content of speech acquired by the voice input unit, an action recognition means for recognizing care actions performed by the caregiver from a video acquired by the video recording unit, and a state determination means for determining the care recipient's health status from biometric information acquired by the sensor input unit.

[0207] [Note 3] The care support system according to Note 2, wherein the analysis means further comprises a shooting control means that controls the start or end of video recording according to the estimation result by the care scene estimation means, and an information management means that tags and manages the support information with respect to the video data, and the information management means maintains the support information in association with the care recipient ID and can update the support information based on feedback from multiple caregivers.

[0208] [Note 4] The analysis means has a function to analyze the video data and classify it according to care scenarios and automatically generate a care manual, and a function to adjust the care manual based on the care recipient's personal profile and current health status, wherein the personal profile includes name, age, level of care required, medical history, food preferences, and communication preferences, as described in Note 3 of the care support system.

[0209] [Note 5] The care support system described in Note 4, wherein the output means comprises: an individualization means that adjusts the length of the video and the difficulty of the explanation according to the degree of cognitive decline and the visual and auditory condition of the person to be cared for; a display means that overlays points of attention specific to the person to be cared for onto the video; and an audio guide means that provides audio guidance on care procedures.

[0210] [Note 6] The care support system described in Note 5, wherein the output means includes a translation means that performs automatic subtitle generation and multilingual translation, and the translation means has the function of translating the contents of the care manual into multiple languages, outputting audio guidance in the multiple languages, and overlaying multilingual subtitles on the video.

[0211] [Note 7] The care support system according to any one of Notes 1 to 6, wherein the analysis means includes an abnormality detection means for detecting abnormalities of the person being cared for using sensor data and video data, the abnormality detection means calculates a risk level that quantifies the abnormal state of the person being cared for, and has an emergency notification function that automatically calls up an emergency response manual and notifies at least one of the caregiver, the family of the person being cared for, and medical personnel when the risk level exceeds a preset threshold.

[0212] [Note 8] The care support system as described in Note 7, wherein the analysis means includes an anomaly prediction means that analyzes the biometric information and behavioral patterns of the person to be cared for using a machine learning model and predicts risks including the risk of falling and deterioration of the person to be cared for, and when the anomaly prediction means predicts the risk, it presents the caregiver with a care manual or warning that includes points to note corresponding to the predicted risk, and stores the risk prediction results as support information.

[0213] [Note 9] The care support system according to any one of Notes 1 to 8, wherein the analysis means comprises a feedback input means for receiving feedback from the caregiver, and a learning means for updating the support information based on the feedback, and the learning means performs version control of the support information and records the update history of the support information.

[0214] [Note 10] The care support system described in Note 9, wherein the analysis means includes a knowledge sharing means for sharing the support information among multiple care facilities, the knowledge sharing means enables cross-facility access to the support information based on the care recipient ID, and has a function for protecting personal information and setting permissions based on facility-specific rules.

[0215] [Note 11] A server device comprising: an input means for acquiring data in a care setting, including data in multiple formats such as voice data, text data, video data, and sensor data; an analysis means for analyzing the data in multiple formats to extract support information including care procedures and points to note specific to the person being cared for; and an output means for providing personalized information to the caregiver or the person being cared for based on the support information.

[0216] [Note 12] A control method for a server device, wherein the server device acquires data in multiple formats, including voice data, text data, video data, and sensor data, which are data from a caregiving site; analyzes the data in multiple formats to extract support information including care procedures and points to note specific to the person being cared for; and provides personalized information to the caregiver or the person being cared for based on the support information.

[0217] [Note 13] A computer-readable storage medium that stores a program for causing a computer mounted on a server device to execute: a process for acquiring data in multiple formats, including audio data, text data, video data, and sensor data, which are data from a caregiving site; a process for analyzing the data in multiple formats to extract support information, including care procedures and points to note specific to the person being cared for; and a process for providing personalized information to the caregiver or the person being cared for based on the support information.

[0218] Furthermore, some or all of the configurations described in Appendices 2 to 10, which are dependent on Appendice 1 above, may also be dependent on Appendices 11 to 13 in the same way as Appendices 2 to 10. Moreover, not limited to Appendices 1, 11, 12, and 13, some or all of the configurations described as appendices may also be dependent on various hardware, software, various recording means for recording software, or systems, without departing from the embodiments described above.

[0219] Furthermore, each disclosure of the above-mentioned prior art documents cited herein is incorporated herein by reference. Although embodiments of the present invention have been described above, the present invention is not limited to these embodiments. It will be understood by those skilled in the art that these embodiments are merely illustrative and that various modifications are possible without departing from the scope and spirit of the present invention. That is, the present invention naturally includes the entire disclosure, including the claims, and various modifications and alterations that can be made by those skilled in the art in accordance with the technical idea.

[0220] 10 Server device 20 Terminal 30 Wearable device 40 Environmental sensor 50 Medical device 101 Input means 102 Analysis means 103 Output means 201 Input unit 202 Analysis unit 203 Output unit 204 Database 301 Video playback area 311 Processor 312 Memory 313 Input / output interface 314 Communication interface

Claims

1. A care support system comprising: an input means for acquiring data in multiple formats at a care setting; an analysis means for analyzing the data in multiple formats and extracting support information; and an output means for outputting information to the caregiver or the person receiving care based on the support information, wherein the support information includes care procedures and points to note specific to the person receiving care.

2. The care support system according to claim 1, wherein the input means includes a voice input unit for acquiring the caregiver's speech, a text input unit for acquiring care records, a video recording unit for recording care scenes, and a sensor input unit for acquiring the care recipient's biometric information, and the analysis means includes a care scene estimation means for estimating a care scene from the content of speech acquired by the voice input unit, an action recognition means for recognizing care actions performed by the caregiver from a video acquired by the video recording unit, and a state determination means for determining the care recipient's health status from biometric information acquired by the sensor input unit.

3. The care support system according to claim 2, wherein the analysis means further comprises a shooting control means that controls the start or end of video recording according to the estimation result by the care scene estimation means, and an information management means that tags and manages the support information with respect to the video data, the information management means holds the support information in association with the care recipient ID, and is capable of updating the support information based on feedback from multiple caregivers.

4. The care support system according to claim 3, wherein the analysis means has a function to analyze the video data and classify it according to care scenarios and automatically generate a care manual, and a function to adjust the care manual based on the care recipient's personal profile and current health status, the personal profile including name, age, level of care required, medical history, food preferences, and communication preferences.

5. The care support system according to claim 4, wherein the output means comprises: an individualization means for adjusting the length of the video and the difficulty of the explanation according to the degree of cognitive decline and the visual and auditory condition of the person being cared for; a display means for overlaying points of attention specific to the person being cared for onto the video; and an audio guide means for providing audio guidance on care procedures.

6. The care support system according to claim 5, wherein the output means comprises a translation means for automatic subtitle generation and multilingual translation, and the translation means has the function of translating the contents of a care manual into multiple languages, outputting audio guidance in the multiple languages, and overlaying multilingual subtitles on a video.

7. The care support system according to any one of claims 1 to 6, wherein the analysis means includes an abnormality detection means for detecting abnormalities of the person being cared for using sensor data and video data, the abnormality detection means calculates a risk level that quantifies the abnormal state of the person being cared for, and has an emergency notification function that automatically calls up an emergency response manual and notifies at least one of the caregiver, the family of the person being cared for, and a medical professional when the risk level exceeds a preset threshold.

8. The care support system according to claim 7, wherein the analysis means includes an anomaly prediction means that analyzes the biological information and behavioral patterns of the person to be cared for using a machine learning model and predicts risks including the risk of falling and deterioration of the health condition of the person to be cared for, and when the anomaly prediction means predicts the risk, it presents the caregiver with a care manual or warning that includes points to note corresponding to the predicted risk, and stores the risk prediction results as support information.

9. The care support system according to any one of claims 1 to 8, wherein the analysis means comprises a feedback input means for receiving feedback from the caregiver, and a learning means for updating the support information based on the feedback, the learning means for version control of the support information and for recording the update history of the support information.

10. The care support system according to claim 9, wherein the analysis means comprises a knowledge sharing means for sharing the support information among multiple care facilities, the knowledge sharing means enables cross-facility access to the support information based on the care recipient ID, and has a function for protecting personal information and setting permissions based on facility-specific rules.

11. A server device comprising: an input means for acquiring data in a care setting, including data in multiple formats such as voice data, text data, video data, and sensor data; an analysis means for analyzing the data in multiple formats to extract support information including care procedures and points to note specific to the person being cared for; and an output means for providing personalized information to the caregiver or the person being cared for based on the support information.

12. A control method for a server device, comprising: acquiring data in a care setting, including data in multiple formats such as voice data, text data, video data, and sensor data; analyzing the data in multiple formats to extract support information including care procedures and points to note specific to the person being cared for; and providing personalized information to the caregiver or the person being cared for based on the support information.

13. A computer-readable storage medium that stores a program for causing a computer mounted on a server device to execute: a process for acquiring data in multiple formats, including audio data, text data, video data, and sensor data, which are data from a caregiving site; a process for analyzing the data in multiple formats to extract support information, including care procedures and points to note specific to the person being cared for; and a process for providing personalized information to the caregiver or the person being cared for based on the support information.

Images