System

Abstract

A system is provided.SOLUTION: The system includes means for collecting behavior data and conversation data of a resident, means for analyzing the collected data to grasp a rhythm and a health condition of the resident, means for generating an optimum care plan based on the grasped rhythm and health condition, and means for notifying a nursing staff of the generated care plan.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Japan's nursing care market is expected to expand further as the aging society progresses, raising concerns that the burden on each nursing care staff member will increase. Furthermore, it is difficult for nursing care staff to respond to the health conditions and individual needs of all residents, resulting in high facility operating costs. In these circumstances, there is a need for technology that can provide efficient nursing care services and reduce the burden on nursing care staff. [Means for solving the problem]

[0005] To solve this problem, the present invention employs the following means. Specifically, the system includes a means for collecting behavioral data and conversational data of residents, a means for analyzing the collected data to understand the resident's rhythm and health status, a means for generating an optimal care plan based on the understood rhythm and health status, and a means for notifying care staff of the generated care plan. Furthermore, by including a means for cleansing the collected data and a means for assessing health risks, the system can improve the reliability of the data and generate more accurate care plans. This reduces the burden on care staff and improves the quality of care services.

[0006] "Resident" refers to a person who resides in a nursing facility and receives nursing care services.

[0007] "Behavioral data" refers to data that includes information about the physical movements and daily rhythms of residents.

[0008] "Conversation data" refers to data that includes information about the words spoken by residents and the content of their conversations.

[0009] "Means for collection" refers to the equipment and methods for acquiring behavioral and conversational data using sensors and voice recognition devices.

[0010] "Means of analysis" refers to the algorithms and technologies used to process the collected data and identify the rhythms and health conditions of residents.

[0011] "Rhythm" refers to a regular series of movements and patterns of activities in a resident's daily life.

[0012] "Health status" refers to information that indicates the physical and mental health status of a resident.

[0013] A "care plan" refers to a plan that includes care measures and suggestions optimized based on the resident's health condition and lifestyle.

[0014] "Means for notifying" refers to a device or method for communicating the generated care plan and risk information to nursing staff.

[0015] "Cleansing means" refers to technology for removing noise and misrecognized data from collected data to improve the accuracy of analysis.

[0016] "Means for assessing health risks" refers to technologies that identify potential health problems of residents as determined from analyzed data and assess their risk levels. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. Furthermore, it generates an optimal care plan based on this information and notifies the care staff, thereby providing appropriate care to residents and reducing the care staff's workload.

[0039] 1. Data Collection Phase

[0040] Device:

[0041] The terminals are equipped with IoT sensors and voice recognition devices to detect the residents' behavior and conversations. This allows them to accurately capture the actions and words that residents perform in their daily lives. For example, the terminals can detect and collect data on when residents wake up in the morning, when they eat meals, and even when they say they're hungry.

[0042] server:

[0043] The server receives and securely stores data sent from the device, which includes building a database for each resident and centrally managing behavioral and conversation data.

[0044] 2. Data analysis phase

[0045] server:

[0046] The server cleanses the collected data, removing incomplete data and noise, improving the accuracy of the analysis. Next, machine learning algorithms are used to analyze the data and identify the resident's behavioral patterns and daily rhythms. For example, it can detect patterns such as a resident waking up at roughly the same time every day and eating meals at the same time. The analysis results can also be used to assess the resident's health risks and detect irregular behavior or abnormal health signs.

[0047] 3. Care plan proposal phase

[0048] server:

[0049] Based on the analysis results, an optimal care plan is generated for each resident. The care plan includes suggestions for daily living support and medical intervention. For example, if a resident's frequent trips to the toilet at night are detected, the server will generate a care plan suggesting a review of fluid intake and a doctor's consultation.

[0050] Care staff (users):

[0051] The generated care plan is sent to the care staff's devices in real time, allowing them to provide appropriate care for the resident, such as suggesting specific meal plans and conducting regular health checks.

[0052] Specific examples

[0053] Device:

[0054] Resident A wakes up at 7:00 every morning, and the voice recognition device detects utterances such as "It's about time to get up" and sends them to the server.

[0055] server:

[0056] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm.

[0057] Care staff (users):

[0058] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" are notified.

[0059] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities and reduce the burden on nursing staff.

[0060] The processing flow will be explained below.

[0061] Program processing steps

[0062] 1. Data Collection Phase

[0063] Step 1:

[0064] The device detects the resident's actions and conversations. For example, it uses IoT sensors and voice recognition devices to capture the time the resident wakes up in the morning or when they say they're hungry.

[0065] Step 2:

[0066] The data detected by the device is sent in real time to a server, where it is classified as behavioral data or conversation data.

[0067] Step 3:

[0068] The server stores the received data in a database. The data is organized for each resident and stored together with past data.

[0069] 2. Data analysis phase

[0070] Step 4:

[0071] The server cleanses the collected data, removing incomplete data and noise to improve the accuracy of the analysis.

[0072] Step 5:

[0073] The server analyzes the data using machine learning algorithms to extract the resident's daily behavioral patterns and identify regularities, such as waking up at 7am and having breakfast at 8am every day.

[0074] Step 6:

[0075] The server uses the analysis results to understand the resident's rhythm and health status, for example, by analyzing sleep and eating patterns, and assessing the health risks if any abnormalities are found.

[0076] 3. Care plan proposal phase

[0077] Step 7:

[0078] The server generates an optimal care plan based on the analysis results. For example, if a resident is prone to a particular health problem (such as frequent urination at night), the plan will include measures to improve that problem.

[0079] Step 8:

[0080] The server generates a care plan and notifies the care staff, including specific improvement measures and daily care suggestions.

[0081] Step 9:

[0082] Care staff are notified and review the care plan, including making dietary changes or additional health checks to accommodate each resident's health condition.

[0083] Step 10:

[0084] Nursing staff carry out actual care based on the care plan. For example, for a resident with nocturia, they may provide guidance on fluid intake and arrange appointments with medical institutions.

[0085] Through these steps, the AI ​​nursing care support system efficiently grasps the rhythm and health status of residents and provides optimal care plans, thereby reducing the burden on nursing staff and improving the quality of nursing care services.

[0086] Example 1

[0087] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0088] Conventional nursing care support systems face challenges in efficiently collecting and analyzing resident behavior and conversation data and notifying nursing staff of optimal care plans in real time. Furthermore, collected data often contains noise and incomplete data, necessitating data cleansing. However, this manual process requires time and effort, so there is a demand for more accurate analysis and faster response.

[0089] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0090] In this invention, the server includes means for collecting resident behavioral data and conversation data, means for transmitting the collected data to the server, means for storing the transmitted data in a database, means for cleansing the stored data, means for analyzing the cleansed data to grasp the resident's rhythm and health condition, means for generating an optimal care plan based on the grasped rhythm and health condition, and means for notifying the care staff of the generated care plan. This makes it possible to efficiently and accurately analyze resident data and provide an optimal care plan in real time.

[0091] "Resident behavior data" is information related to the actions and behaviors that residents perform in their daily lives.

[0092] "Conversation data" is information about the voices uttered by the resident and their contents.

[0093] "Means for collection" refers to devices and technologies for acquiring resident behavioral data and conversation data.

[0094] "Transmitting means" refers to the communication device or protocol used to transfer the collected data to the server.

[0095] "Means of storage" refers to a database or storage system for safely storing and managing the transmitted data.

[0096] A "cleansing method" is a process or algorithm for detecting, removing, or correcting incomplete data or noise.

[0097] "Means of analysis" refers to technology that analyzes cleansed data using machine learning algorithms, etc., to understand residents' behavioral patterns and health conditions.

[0098] "Means of understanding" refers to methods for recognizing and understanding the rhythms and health status of residents based on information obtained from analyzed data.

[0099] The "means of generation" refers to systems and algorithms that create optimal care plans for each resident based on their rhythms and health status.

[0100] The "notification means" refers to a communication means or interface for informing the care staff of the generated care plan in real time.

[0101] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. This system is realized using the following hardware and software.

[0102] Data Collection Phase

[0103] Device:

[0104] The terminal uses IoT sensors and voice recognition devices to detect the resident's actions and conversations. The IoT sensors detect when the resident gets up from the bed in real time, and the voice recognition device accurately recognizes when the resident says, "It's about time to get up." This data is sent from the terminal to the server.

[0105] Data storage phase

[0106] server:

[0107] The server receives the data sent from the device and stores it in a database for each resident. This database stores data in a format that includes resident ID, timestamp, behavioral data, conversation data, etc.

[0108] Data Cleansing Phase

[0109] server:

[0110] The server reads new data from the database and starts the data cleansing process, specifically detecting incomplete data and noise, and removing and correcting them. This process improves the accuracy of the analysis.

[0111] Data analysis phase

[0112] server:

[0113] The server uses machine learning algorithms to analyze the cleansed data, identifying resident behavior patterns and daily rhythms. For example, it can identify resident A's pattern of waking up at 7 a.m. every morning and evaluate this rhythm as healthy.

[0114] Care plan generation phase

[0115] server:

[0116] Based on the analysis results, the server automatically generates an optimal care plan for each resident. This care plan includes support for daily activities and medical intervention suggestions. For example, for a resident who frequently uses the toilet at night, it might include suggestions to review fluid intake and instructions to seek medical advice.

[0117] Care plan notification phase

[0118] server:

[0119] The generated care plan is notified to the care staff's device in real time.

[0120] User (care staff):

[0121] Care staff can then review the care plan sent to them via the device and provide specific care to the resident based on the plan's contents, such as praising the resident for getting up at 7 a.m. every morning and helping them prepare breakfast.

[0122] Specific examples

[0123] Device:

[0124] Resident A wakes up at 7am every morning, and the device detects when he or she says, "It's about time to get up," and sends this information to the server.

[0125] server:

[0126] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm.

[0127] Care staff (users):

[0128] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" are notified.

[0129] Example prompts for generative AI models

[0130] Example prompt sentence:

[0131] "I am designing a system that can analyze a resident's behavioral pattern, such as waking up at 7am every day. How can I use this system to assess the resident's health risks and generate and communicate an appropriate care plan?"

[0132] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities and reduce the burden on nursing staff.

[0133] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0134] Step 1:

[0135] The device collects data

[0136] Input: Resident actions and conversations

[0137] Output: Behavioral and conversational data

[0138] The device uses IoT sensors and voice recognition devices to monitor the resident's behavior and conversations in real time. Specifically, the IoT sensors detect when the resident gets up from the bed, and the voice recognition device records what the resident says. For example, it detects the voice saying, "It's about time to get up," and collects this as data.

[0139] Step 2:

[0140] The device sends the data to the server

[0141] Input: Behavioral and conversational data

[0142] Output: Data sent to the server

[0143] The collected behavioral and conversational data is sent to the server in batch format. Specifically, the data is packaged within the device and transferred to the server using a secure communication protocol. For example, data with timestamps is compiled into batches and sent to the server at regular intervals.

[0144] Step 3:

[0145] The server saves the data to a database

[0146] Input: Data sent

[0147] Output: Data stored in the database

[0148] The server receives the data sent from the device and stores it in a database for each resident. Specifically, it uses a database structure to store the resident ID, timestamp, behavioral data, conversation data, etc. For example, an entry such as "User ID: A, Time: 7:00 AM, Action: Wake up, Conversation: 'It's about time to get up'" is created.

[0149] Step 4:

[0150] The server cleanses the data

[0151] Input: Data stored in the database

[0152] Output: Cleansed data

[0153] The server reads new data from the database and performs a data cleansing process to detect, remove, and correct incomplete data and noise. Specifically, it complements missing values ​​and corrects outliers. For example, it filters noise from voice data to produce clear data.

[0154] Step 5:

[0155] The server analyzes the data

[0156] Input: Cleansed data

[0157] Output: Resident behavior patterns and health status

[0158] The server uses machine learning algorithms to analyze the cleansed data. This data analysis identifies the resident's behavioral patterns and daily rhythms. For example, it detects that resident A wakes up at 7:00 every morning. This information is then used to evaluate the resident's health status.

[0159] Step 6:

[0160] The server generates a care plan

[0161] Input: Behavioral patterns and health status

[0162] Output: Optimal care plan

[0163] Based on the analysis results, the server generates an optimal care plan for each resident. The generated care plan includes support for daily activities and medical intervention suggestions. For example, for a resident who frequently uses the toilet at night, it might include suggestions to review fluid intake and seek medical advice.

[0164] Step 7:

[0165] The server notifies the care plan

[0166] Input: Optimal Care Plan

[0167] Output: Care plan sent to the care staff's device

[0168] The generated care plan is sent to the care staff's device in real time. Specifically, a pop-up message or an alert is sent to the care staff's device via the notification system, allowing the necessary care to be carried out promptly.

[0169] Step 8:

[0170] The user (caregiver) performs care

[0171] Input: Notified Care Plan

[0172] Output: Specific care for residents

[0173] Care staff can then review the care plan sent to them via the device and provide specific care based on the plan, such as praising the resident for getting up at 7am every morning or helping them prepare breakfast.

[0174] (Application example 1)

[0175] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0176] Conventional factory robot maintenance and operation management is often performed at regular intervals, making it difficult to detect actual usage status and abnormalities in real time. This has led to maintenance delays and excessive maintenance work, making efficient operation difficult. It has also been impossible to properly collect instructions and comments from operators and generate optimal maintenance plans based on them.

[0177] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0178] In this invention, the server includes means for collecting behavioral data and conversation data of residents and automatic equipment, means for analyzing the collected data to grasp the work rhythms and operating status of the residents and automatic equipment, means for generating an optimal maintenance plan based on the grasped work rhythms and operating status, and means for notifying a manager or operator of the generated maintenance plan, thereby enabling the generation and notification of an optimal maintenance plan based on the operating status of factory robots and instructions from operators.

[0179] A "resident" is a person who lives within a particular facility or environment and about whom data is collected.

[0180] "Automatic equipment" refers to a device that automatically performs a task based on a control signal, and examples thereof include factory robots.

[0181] "Behavioral data" refers to information about the behavior and operation history of a target person or automated equipment, and is collected through sensors and devices.

[0182] "Conversation data" refers to information collected through a voice recognition device that includes statements and instructions from the target person or operator.

[0183] A "server" is a computer system for storing and analyzing collected data and communicating generated plans.

[0184] "Work rhythm" refers to the daily behavioral patterns and schedules of automated equipment and residents.

[0185] "Operational status" refers to information about the actual operating state of automated equipment and its performance.

[0186] "Maintenance plan" refers to the schedule and content of maintenance work required to keep equipment operating properly.

[0187] "Manager or operator" means a person responsible for the operation and management of automated equipment or systems.

[0188] The system according to the present invention collects and analyzes behavioral and conversational data of automated equipment and residents to understand work rhythms and operating conditions, and then generates and notifies optimal maintenance plans. Specific embodiments of the system are described below.

[0189] 1. Data Collection Phase

[0190] Device:

[0191] The terminal collects real-time operational data and operator conversation data from IoT sensors and voice recognition devices attached to automated equipment. For example, it can collect information on whether robots in a factory are operating normally or what operators say when giving maintenance instructions.

[0192] 2. Data analysis phase

[0193] server:

[0194] The server receives the data sent from the device and first cleans it. This removes noise and missing data to improve the accuracy of the analysis. Next, it uses machine learning algorithms to analyze the data and detect patterns and anomalies in the operation of automated equipment. For example, it can detect the phenomenon of a robot slowing down at certain times each day.

[0195] 3. Maintenance plan generation phase

[0196] server:

[0197] The server then generates an optimal maintenance plan based on the analysis results. This maintenance plan includes periodic part replacement for automated equipment and necessary actions when specific abnormalities occur. For example, if a robot's performance deteriorates, the server can add the replacement of a specific part to the schedule.

[0198] 4. Notification Phase

[0199] Device:

[0200] The generated maintenance plan is sent to the terminal of the manager or operator in real time, allowing the manager or operator to carry out the necessary maintenance work at the appropriate time. For example, a specific instruction such as "replace a specific part every weekend" may be sent to the operator's terminal.

[0201] Hardware and software used

[0202] IoT sensors: Collect operational data from the robot.

[0203] Voice recognition device: collects instructions and comments from the operator.

[0204] Server: Cleanses and analyzes the data.

[0205] Machine learning algorithms: Use Python's TENSORFLOW (registered trademark) or PyTorch.

[0206] Database: Collected data is managed using SQLite or MySQL (registered trademark).

[0207] Specific examples

[0208] Terminal: When factory robot A needs specific maintenance every weekend, a voice recognition device detects an operator's instruction such as "Robot A is exhibiting a malfunction" and sends it to the server.

[0209] Server: The server stores the received data in a database and uses machine learning algorithms to analyze abnormal patterns of Robot A. For example, it can detect irregular behavior patterns and recommend maintenance before an abnormality occurs.

[0210] Manager or operator: The optimal maintenance plan is sent to the operator's terminal, with specific suggestions such as "replace specific parts every weekend."

[0211] Prompt Sentence Examples

[0212] "Develop a program that uses IoT sensors and voice recognition devices attached to factory robots to collect operational data and operator instructions regarding maintenance for Robot A, and generate an optimal maintenance plan. This program will include a function to detect anomalies and notify maintenance plans in real time. Use Python's TensorFlow."

[0213] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0214] Step 1:

[0215] The terminal collects operation data and operator conversation data from IoT sensors and voice recognition devices attached to the automated equipment. Specifically, it acquires the robot's operating status and voice data of instructions and comments given by the operator in real time and converts this data into digital format. Input: Operation data from the automated equipment's sensors, voice data from the voice recognition device. Output: Digital operation data and conversation data.

[0216] Step 2:

[0217] The terminal sends the collected digital data to the server. Specifically, it uses a data communication protocol to properly transfer the data and store it in a database on the server. Input: Digital operating data and conversation data. Output: Operating data and conversation data stored on the server.

[0218] Step 3:

[0219] The server cleanses the received data. Specifically, it removes incomplete data and noise and prepares the data to improve the accuracy of analysis. For example, it filters background noise from voice data and fills in missing values ​​in sensor data. Input: Operational data and conversation data stored on the server. Output: Cleansed data.

[0220] Step 4:

[0221] The server analyzes the cleansed data. Specifically, it uses machine learning algorithms to analyze the data and understand work rhythms and operating conditions. For example, it uses TensorFlow or PyTorch to learn the operation patterns of equipment and detect anomalies. Input: Cleansed data. Output: Analysis results (work rhythms, operating conditions, anomaly detection results).

[0222] Step 5:

[0223] The server generates an optimal maintenance plan based on the analysis results. Specifically, it automatically creates a schedule for maintenance timing and necessary part replacement based on the output of the machine learning model. Input: Analysis results (work rhythm, operating status, anomaly detection results). Output: Optimal maintenance plan.

[0224] Step 6:

[0225] The server notifies the administrator or operator of the generated maintenance plan. Specifically, it sends a real-time notification to the operator's terminal, presenting the necessary maintenance work. For example, a notification containing specific instructions such as "replace a specific part every weekend" is sent. Input: Optimized maintenance plan. Output: Maintenance instructions displayed on the operator's terminal.

[0226] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0227] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. Furthermore, it generates an optimal care plan based on this information and notifies the care staff, thereby providing appropriate care for the residents and reducing the care staff's workload. Furthermore, by combining it with an emotion engine, it can also recognize the user's emotions and provide emotionally appropriate care.

[0228] 1. Data Collection Phase

[0229] Device:

[0230] The terminals are equipped with IoT sensors and voice recognition devices to detect the residents' behavior and conversations. This allows them to accurately capture the actions and words that residents perform in their daily lives. For example, the terminals can detect and collect data on when residents wake up in the morning, when they eat meals, and even when they say they're hungry.

[0231] server:

[0232] The server receives and securely stores data sent from the device, which includes building a database for each resident and centrally managing behavioral and conversation data.

[0233] 2. Data analysis phase

[0234] server:

[0235] The server cleanses the collected data, removing incomplete data and noise, improving the accuracy of the analysis. Next, machine learning algorithms are used to analyze the data and identify the resident's behavioral patterns and daily rhythms. For example, it can detect patterns such as a resident waking up at roughly the same time every day and eating meals at the same time. The analysis results can also be used to assess the resident's health risks and detect irregular behavior or abnormal health signs.

[0236] Emotion engine integration:

[0237] Furthermore, an emotion engine is used to recognize the resident's emotions from conversation data. The voice recognition device analyzes the resident's tone of voice and choice of words to determine the resident's emotional state. For example, if a resident says, "I'm very tired today," the emotion engine analyzes this statement and recognizes that the resident is feeling fatigued.

[0238] 3. Care plan proposal phase

[0239] server:

[0240] Based on the analysis results, an optimal care plan is generated for each resident. The care plan includes suggestions for daily living support and medical interventions. For example, if a resident is prone to a particular health problem (such as nocturia), measures to improve that problem will be included in the plan. Care suggestions appropriate to the resident's emotional state are also generated based on emotional data obtained from the emotion engine. For example, if a resident is feeling fatigued, suggestions may include ensuring a relaxing environment and taking time to rest.

[0241] Care staff (users):

[0242] The generated care plan is sent to the care staff's devices in real time, allowing them to provide appropriate care for the resident, for example, by suggesting specific meal plans or planning additional health checks based on the resident's emotional state.

[0243] Specific examples

[0244] Device:

[0245] Resident A wakes up at 7:00 every morning, and the voice recognition device detects utterances such as "It's about time to get up" and sends them to the server.

[0246] server:

[0247] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm. If the emotion engine detects fatigue from Resident A's tone of voice, that emotion data is also stored on the server for analysis.

[0248] Care staff (users):

[0249] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time and emotional state. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" and "providing a relaxing environment" are provided.

[0250] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities, reduce the burden on nursing staff, and provide care that takes into consideration the emotions of users.

[0251] The processing flow will be explained below.

[0252] Program processing steps

[0253] 1. Data Collection Phase

[0254] Step 1:

[0255] The device uses IoT sensors to detect the resident's behavior, specifically capturing the resident's movements in real time, such as getting up from the bed or moving around the room.

[0256] Step 2:

[0257] The device detects the resident's conversation using a voice recognition device. For example, if a resident says "I'm hungry," this voice data is collected.

[0258] Step 3:

[0259] The device transmits the acquired behavioral and conversation data to a server, where the data is categorized and stored.

[0260] Step 4:

[0261] The device analyzes the resident's voice data and requests the emotion engine to determine their emotional state, for example, by detecting their emotional state such as "happiness" or "sadness" from their tone of voice and vocabulary.

[0262] 2. Data analysis phase

[0263] Step 5:

[0264] The server stores the received data in a database, where it is stored together with past data in a centralized management system for each resident.

[0265] Step 6:

[0266] The server cleanses the collected data, removing incomplete data and noise, a process that improves the accuracy of the analysis.

[0267] Step 7:

[0268] The server uses machine learning algorithms to analyze the data and extract patterns and rhythms of the resident's daily behavior, such as identifying a regular pattern where the resident wakes up at 7 a.m. and eats breakfast at 8 a.m.

[0269] Step 8:

[0270] The server evaluates the health risks of residents based on the analysis results. For example, if a pattern of frequent toilet visits at night is detected, the server evaluates the risk.

[0271] Step 9:

[0272] The server also analyzes emotional data from the emotion engine. For example, if a resident is feeling fatigued, that information is added to the risk assessment.

[0273] 3. Care plan proposal phase

[0274] Step 10:

[0275] The server generates an optimal care plan based on the analysis results. For example, if a resident is at risk of nocturia, it generates a plan that includes adjusting fluid intake and suggesting a doctor's appointment.

[0276] Step 11:

[0277] The server generates a care plan based on the emotional data, taking into account the resident's emotional state. For example, if the resident is feeling fatigued, it will include suggestions for extending rest time and relaxation.

[0278] Step 12:

[0279] The server notifies care staff of the care plan and risk assessment results in real time, including specific care suggestions.

[0280] Step 13:

[0281] Care staff (users) receive notifications and review the care plan, for example, making changes to the resident's diet or planning additional health checks.

[0282] Step 14:

[0283] Care staff carry out actual care based on the care plan, such as providing guidance on fluid intake and creating an environment for relaxation.

[0284] Specific examples

[0285] Step 1:

[0286] The device uses an IoT sensor to detect that resident A gets out of bed at 7am.

[0287] Step 2:

[0288] The device detects the voice of resident A saying, "I'm a little tired today."

[0289] Step 3:

[0290] The terminal transmits the acquired behavioral data and conversation data to the server.

[0291] Step 4:

[0292] The device sends the detected voice data to the emotion engine, which determines resident A's emotional state as "fatigue."

[0293] Step 5:

[0294] The server stores the received data in a database.

[0295] Step 6:

[0296] The server cleanses the collected data.

[0297] Step 7:

[0298] The server analyzes the data and identifies the behavioral patterns of Resident A. For example, it detects that Resident A wakes up at 7:00 every morning and has breakfast at 8:00.

[0299] Step 8:

[0300] The server evaluates health risks and determines that resident A is at risk of nocturia.

[0301] Step 9:

[0302] The server adds "fatigue" data from the emotion engine to the evaluation.

[0303] Step 10:

[0304] The server generates an optimal care plan.

[0305] Step 11:

[0306] The server generates a care plan based on the emotional state, for example, including relaxation recommendations for Resident A who feels fatigued.

[0307] Step 12:

[0308] The server notifies the care staff of the generated care plan.

[0309] Step 13:

[0310] The care staff (user) receives a notification and checks the care plan.

[0311] Step 14:

[0312] Nursing staff carry out actual care based on the care plan.

[0313] Through these specific processing steps, the AI ​​nursing care support system can efficiently grasp the resident's rhythm and health status, provide an optimal care plan, and utilize an emotion engine to provide emotion-based care.

[0314] Example 2

[0315] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0316] As the aging population continues to grow, improving the quality of nursing care services and reducing the workload of nursing staff at nursing facilities have become important issues. There is also a need to provide appropriate care tailored to residents' emotions and health conditions. However, current systems lack sufficient analysis of residents' behavioral and conversational data, making it difficult to provide care plans that take emotions into account. Therefore, a system that can provide efficient and effective care is needed.

[0317] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting resident behavioral information and speech information, means for analyzing the collected information to understand the resident's rhythm and health condition, means for generating an optimal care plan based on the understood rhythm and health condition, means for notifying care staff of the generated care plan, means for analyzing the resident's emotions using a voice recognition device and an emotion analysis engine, and means for adjusting the care plan based on the results of the emotion analysis. This makes it possible to efficiently understand the resident's health condition based on data obtained from the resident's behavior and speech and to generate a care plan accordingly. Furthermore, by adjusting the care content based on the resident's emotions, it is possible to improve the quality of care services and reduce the workload of care staff.

[0318] "Residents" refers to elderly people who live in nursing homes or senior housing, and people who require nursing care support.

[0319] "Behavioral information" refers to data on specific actions taken by residents in their daily lives (for example, waking up, eating, moving around, etc.).

[0320] "Speech information" refers to the words and conversations that residents make in their daily lives.

[0321] "Means of collection" refers to devices and systems for collecting behavioral information and speech information using sensors, voice recognition devices, etc.

[0322] "Means of analysis" refers to methods and systems for analyzing collected data to understand residents' behavioral patterns, health status, emotions, etc.

[0323] "Rhythm" refers to the temporal patterns and habits of residents' daily lives.

[0324] "Health status" refers to the state of a resident's physical and mental health.

[0325] A "care plan" refers to a plan that specifically outlines the optimal content and methods of care to be provided to residents based on the results of the analysis.

[0326] "Care workers" refers to employees and staff who provide care services to residents.

[0327] "Means for notifying" refers to a method or system for notifying care staff of the generated care plan.

[0328] A "voice recognition device" refers to a device or system that detects what residents say and converts it into text data.

[0329] "Emotion analysis engine" refers to an algorithm or system that analyzes data obtained from a voice recognition device to estimate the emotional state of a resident.

[0330] "Results of emotion analysis" refers to the emotional state of the resident analyzed by the emotion analysis engine.

[0331] "Adjustment means" refers to a method or system for appropriately changing the content or method of the care plan based on the analyzed emotional data.

[0332] The AI ​​nursing care support system of this invention collects and analyzes information on residents' behavior and speech to understand their rhythm, health condition, and emotions. Furthermore, it generates an optimal care plan based on this information and notifies caregivers, thereby providing appropriate care for residents and reducing the caregiver's workload.

[0333] Data collection

[0334] Device:

[0335] The devices are equipped with IoT sensors and voice recognition devices to detect the resident's actions and comments. This allows them to accurately capture the actions and comments that residents make in their daily lives. For example, if a resident wakes up at 7:00 a.m. and says, "It's about time to get up," the voice recognition device will detect this. Motion sensors are also used to collect data on the resident's movements.

[0336] Data transmission and storage

[0337] Device:

[0338] The device sends the collected behavioral information and speech information to a server using a secure communication protocol (e.g., HTTPS).

[0339] server:

[0340] The server receives the data sent from the devices and stores it in a database for each resident, which records behavioral and conversation data along with timestamps.

[0341] Data Cleansing and Analysis

[0342] server:

[0343] The server cleanses the received data, removing incomplete data and noise. Specifically, it filters out extremely short audio data and erroneous sensor values. Once cleansed, the data is analyzed using machine learning algorithms. For example, it can detect a behavioral pattern, such as "Resident A wakes up at 7:00 every morning."

[0344] Emotional Data Analysis

[0345] server:

[0346] The server uses an emotion analysis engine to analyze the resident's emotions from the collected speech information. Data obtained from the voice recognition device is input into the emotion analysis engine to determine the resident's emotional state. For example, a statement such as "I'm very tired today" is analyzed and recognized as indicating that the resident is feeling fatigued.

[0347] Generate a care plan

[0348] server:

[0349] The server generates an optimal care plan for each resident based on the analyzed behavioral patterns and emotional data. The care plan includes support for daily life and medical intervention suggestions. For example, if a resident wakes up at 7 a.m. every morning and feels tired, the server will create a care plan that provides a relaxing environment after waking up.

[0350] Care plan notification

[0351] server:

[0352] The server then sends the generated care plan to the caregiver's device, where the caregiver can check the plan in real time on a tablet or smartphone.

[0353] Care delivery

[0354] Caregiver (user):

[0355] Care staff provide care to residents based on the care plan they receive, for example, praising residents for getting up at 7 a.m. every morning, helping them prepare breakfast, and providing a relaxing environment.

[0356] Prompt Sentence Examples

[0357] "Resident A wakes up at 7:00 every morning, and the voice recognition device detects that he says, 'It's about time to get up.' Please also analyze Resident A's emotional data and generate the optimal care plan."

[0358] In this way, the present invention improves the quality of nursing care services in nursing facilities and reduces the workload of nursing staff. By providing care based on the feelings of residents, more detailed care becomes possible.

[0359] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0360] Step 1: Sensor configuration and data collection

[0361] Terminal: The terminal collects information on the residents' behavior and speech. Specifically, it uses IoT sensors and voice recognition devices. The sensors detect the residents' movements and actions, and the voice recognition device records what the residents say. The input is raw data obtained from the sensors and voice device. The output is recorded as the residents' behavior data and speech data. For example, if a resident wakes up at 7am, the device detects when they say, "It's about time to get up."

[0362] Step 2: Data transmission and storage

[0363] Terminal: The terminal sends the collected behavioral and speech information to the server. The input is the data collected in step 1. The output is a message confirming the completion of the transmission. The data is sent to the server via a secure communication protocol (such as HTTPS).

[0364] Server: The server receives the data sent from the device and stores it in a database. The input is the data sent from the device. The output is the behavioral data and speech data stored in the database. This records the data with a timestamp and makes it available for subsequent analysis.

[0365] Step 3: Data cleansing and preprocessing

[0366] Server: The server cleanses the received data and removes incomplete data and noise. The input is the data stored in step 2. The output is the cleansed data. Specifically, it filters out extremely short audio data and erroneous sensor values.

[0367] Step 4: Data analysis

[0368] Server: The server uses a machine learning algorithm to analyze the cleansed data and understand the resident's behavioral patterns and health status. The input is the clean data obtained in step 3. The output is the analysis results of the resident's behavioral patterns and health status. For example, a pattern such as "Resident A wakes up at 7 o'clock every morning" is detected.

[0369] Step 5: Analyze the sentiment data

[0370] Server: The server uses an emotion analysis engine to analyze the resident's emotions from the utterance data. The input is the utterance data obtained in step 3. The output is the analysis result of the resident's emotional state. For example, the emotion analysis engine analyzes the utterance "I'm very tired today" and outputs the result as "Feeling tired."

[0371] Step 6: Generate a care plan

[0372] Server: The server generates an optimal care plan for each resident based on the analyzed behavioral patterns and emotional data. The input is the analysis results from steps 4 and 5. The output is a care plan for each resident. For example, if a resident wakes up at 7 a.m. every morning and feels tired, a care plan is generated that provides a relaxing environment after waking up.

[0373] Step 7: Inform the plan of care

[0374] Server: The server notifies the care staff device of the care plan it has generated. The input is the care plan generated in step 6. The output is a push notification to the care staff device. This allows the care staff to check the care plan in real time on devices such as tablets and smartphones.

[0375] Step 8: Implementing care

[0376] Care staff (user): Care staff provide care to residents based on the notified care plan. The input is the care plan notified from the server. The output is a record of the care provided. For example, they may praise a resident for waking up at 7am every morning, help them prepare breakfast, and provide a relaxing environment.

[0377] (Application example 2)

[0378] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0379] Current care support systems are effective in collecting residents' behavioral and conversational data, understanding their rhythms and health status based on that data, and generating optimal care plans. However, they have limitations in providing comprehensive health support based on residents' health status and eating patterns. They lack specific suggestions for residents to eat an appropriate diet and the integration of food delivery services to implement those suggestions. Therefore, an effective means of ensuring residents receive optimal nutrition and maintain their health is needed.

[0380] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting resident behavioral data and conversation data, means for analyzing the collected data to understand the resident's rhythm and health condition, means for generating an optimal care plan based on the understood rhythm and health condition, means for notifying care staff of the generated care plan, means for proposing an optimal meal menu based on the user's eating patterns and health condition, and means for notifying a delivery service of the proposed meal menu. This enables comprehensive health support for residents, and by linking appropriate nutritional intake with delivery services, it becomes possible to improve the quality of life of residents and reduce the workload of care staff.

[0381] "Resident" means an individual who resides in a particular facility or residence.

[0382] "Behavioral data" refers to data that indicates the resident's movements and activities in daily life.

[0383] "Conversation data" refers to data collected as text information of the sounds and words uttered by residents.

[0384] "Rhythm" refers to the regular patterns of behavior and lifestyle of residents.

[0385] "Health status" refers to information related to the resident's physical condition and health.

[0386] A "care plan" is a written plan that lists specific care methods and support for residents.

[0387] "Care staff" refers to people whose occupations support the health and lives of residents.

[0388] "Dietary patterns" refer to the habits and tendencies of residents regarding what meals they eat and at what times.

[0389] "Health Status" refers to the state of a resident's physical and mental health.

[0390] A "meal menu" refers to the specific meal contents and menu to be provided to residents.

[0391] "Delivery service" refers to a service for delivering meal menus to residents.

[0392] This invention is a system that collects and analyzes behavioral and conversational data of residents to understand their rhythms and health status. It also generates an optimal care plan based on this data and notifies care staff, providing appropriate care and reducing their workload. It also includes a function to suggest optimal meal menus based on eating patterns and health status and notify the delivery service.

[0393] The server first collects behavioral and conversational data from residents. Behavioral data includes residents' movements and daily activities, while conversational data includes what residents say. This data is collected using smartphones and IoT sensors and sent to the server. A voice recognition API, commonly known as a voice recognition engine, is used to convert the voice data into text data.

[0394] The server cleanses the collected data, removing incomplete data and noise, and then analyzes it using machine learning algorithms to understand the resident's rhythm and health status. This analysis is performed using a generic machine learning platform. For example, if the server detects that a resident has a habit of eating at a specific time, it analyzes their eating patterns. It also evaluates the resident's emotional state using an emotion analysis engine.

[0395] The server then generates an optimal care plan based on the analysis, including assisted living and medical recommendations. It also suggests optimal meal plans based on the resident's eating patterns and emotional state, and notifies a food delivery service. A food delivery API is used to arrange for the optimal menu to be delivered to the resident.

[0396] This system is used in the following specific example.

[0397] When a resident says "I'm hungry," the behavioral data collection module detects this and converts it into text using a speech recognition engine. The data cleansing module processes this text data, and the behavioral pattern analysis module analyzes eating patterns. The emotion analysis module evaluates the resident's emotional state, and the menu generation module generates an optimal meal menu. The delivery notification module notifies the resident of delivery suggestions using a food delivery API.

[0398] Example prompt sentence:

[0399] When a user says "I'm hungry," the behavioral data collection module detects this and converts it into text using a speech recognition engine. The data cleansing module processes this text data, and the behavioral pattern analysis module analyzes eating patterns. The emotion analysis module evaluates the user's emotional state, and the menu generation module generates an optimal meal menu. The delivery notification module notifies the user of delivery suggestions using a food delivery API.

[0400] This system will enable comprehensive health support for residents, improving their quality of life through proper nutritional intake and linking delivery services, and reducing the workload of care staff.

[0401] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0402] Step 1:

[0403] The terminal uses IoT sensors and voice recognition devices to collect resident behavioral and conversational data. For example, if a resident says "I'm hungry," the voice is detected and converted into text data by a voice recognition engine. The input for this step is the resident's behavior and speech, and the output is text data and behavioral data.

[0404] Step 2:

[0405] The behavioral and conversational data sent from the device is sent to a server and securely stored. This involves building and centrally managing a database for each specific resident. The input for this step is text data and behavioral data, and the output is the results stored in the database.

[0406] Step 3:

[0407] The server cleanses the collected data and removes incomplete data and noise. This is a data preprocessing stage and is an important step for accurate analysis. Specifically, it detects outliers and missing data and removes or imputes them. The input for this step is the stored behavioral data and text data, and the output is the cleansed data.

[0408] Step 4:

[0409] Based on the cleansed data, the server analyzes the data using machine learning algorithms. Common machine learning platforms such as TensorFlow and PyTorch are used here. The analysis identifies resident behavior patterns and daily rhythms and assesses health risks. The input for this step is the cleansed data, and the output is behavior patterns and health risk assessment results.

[0410] Step 5:

[0411] Using an emotion analysis engine, the server recognizes the resident's emotions from the conversation data. Specifically, it uses data from a voice recognition device and uses an emotion analysis API to determine the resident's emotional state from the tone of voice and vocabulary. The input for this step is the conversation data, and the output is the emotion analysis results.

[0412] Step 6:

[0413] Based on the analysis results, the server generates an optimal care plan for each resident. This care plan includes specific suggestions for health management and meal plans. The analysis results are integrated with the emotional state to suggest the optimal meal menu for the resident. This results in a care plan that corresponds to the resident's health condition. The inputs for this step are behavioral patterns, health risk assessment, and emotional analysis results, and the output is a care plan and meal menu suggestions.

[0414] Step 7:

[0415] The generated care plan and meal menu are notified to the nursing staff and delivery service. Using the food delivery API, arrangements are made to quickly deliver the optimal meal menu for the resident. Nursing staff are notified of the resident's status in real time, helping them respond quickly. The input to this step is the care plan and meal menu suggestions, and the output is the notification results.

[0416] Through these steps, this system will provide comprehensive health support for residents, reduce the workload of care staff, and provide residents with appropriate nutritional intake.

[0417] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0418] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0419] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0420] [Second embodiment]

[0421] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0422] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0423] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0424] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0425] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0426] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0427] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0428] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0429] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0430] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0431] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0432] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0433] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. Furthermore, it generates an optimal care plan based on this information and notifies the care staff, thereby providing appropriate care to residents and reducing the care staff's workload.

[0434] 1. Data Collection Phase

[0435] Device:

[0436] The terminals are equipped with IoT sensors and voice recognition devices to detect the residents' behavior and conversations. This allows them to accurately capture the actions and words that residents perform in their daily lives. For example, the terminals can detect and collect data on when residents wake up in the morning, when they eat meals, and even when they say they're hungry.

[0437] server:

[0438] The server receives and securely stores data sent from the device, which includes building a database for each resident and centrally managing behavioral and conversation data.

[0439] 2. Data analysis phase

[0440] server:

[0441] The server cleanses the collected data, removing incomplete data and noise, improving the accuracy of the analysis. Next, machine learning algorithms are used to analyze the data and identify the resident's behavioral patterns and daily rhythms. For example, it can detect patterns such as a resident waking up at roughly the same time every day and eating meals at the same time. The analysis results can also be used to assess the resident's health risks and detect irregular behavior or abnormal health signs.

[0442] 3. Care plan proposal phase

[0443] server:

[0444] Based on the analysis results, an optimal care plan is generated for each resident. The care plan includes suggestions for daily living support and medical intervention. For example, if a resident's frequent trips to the toilet at night are detected, the server will generate a care plan suggesting a review of fluid intake and a doctor's consultation.

[0445] Care staff (users):

[0446] The generated care plan is sent to the care staff's devices in real time, allowing them to provide appropriate care for the resident, such as suggesting specific meal plans and conducting regular health checks.

[0447] Specific examples

[0448] Device:

[0449] Resident A wakes up at 7:00 every morning, and the voice recognition device detects utterances such as "It's about time to get up" and sends them to the server.

[0450] server:

[0451] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm.

[0452] Care staff (users):

[0453] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" are notified.

[0454] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities and reduce the burden on nursing staff.

[0455] The processing flow will be explained below.

[0456] Program processing steps

[0457] 1. Data Collection Phase

[0458] Step 1:

[0459] The device detects the resident's actions and conversations. For example, it uses IoT sensors and voice recognition devices to capture the time the resident wakes up in the morning or when they say they're hungry.

[0460] Step 2:

[0461] The data detected by the device is sent in real time to a server, where it is classified as behavioral data or conversation data.

[0462] Step 3:

[0463] The server stores the received data in a database. The data is organized for each resident and stored together with past data.

[0464] 2. Data analysis phase

[0465] Step 4:

[0466] The server cleanses the collected data, removing incomplete data and noise to improve the accuracy of the analysis.

[0467] Step 5:

[0468] The server analyzes the data using machine learning algorithms to extract the resident's daily behavioral patterns and identify regularities, such as waking up at 7am and having breakfast at 8am every day.

[0469] Step 6:

[0470] The server uses the analysis results to understand the resident's rhythm and health status, for example, by analyzing sleep and eating patterns, and assessing the health risks if any abnormalities are found.

[0471] 3. Care plan proposal phase

[0472] Step 7:

[0473] The server generates an optimal care plan based on the analysis results. For example, if a resident is prone to a particular health problem (such as frequent urination at night), the plan will include measures to improve that problem.

[0474] Step 8:

[0475] The server generates a care plan and notifies the care staff, including specific improvement measures and daily care suggestions.

[0476] Step 9:

[0477] Care staff are notified and review the care plan, including making dietary changes or additional health checks to accommodate each resident's health condition.

[0478] Step 10:

[0479] Nursing staff carry out actual care based on the care plan. For example, for a resident with nocturia, they may provide guidance on fluid intake and arrange appointments with medical institutions.

[0480] Through these steps, the AI ​​nursing care support system efficiently grasps the rhythm and health status of residents and provides optimal care plans, thereby reducing the burden on nursing staff and improving the quality of nursing care services.

[0481] Example 1

[0482] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0483] Conventional nursing care support systems face challenges in efficiently collecting and analyzing resident behavior and conversation data and notifying nursing staff of optimal care plans in real time. Furthermore, collected data often contains noise and incomplete data, necessitating data cleansing. However, this manual process requires time and effort, so there is a demand for more accurate analysis and faster response.

[0484] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0485] In this invention, the server includes means for collecting resident behavioral data and conversation data, means for transmitting the collected data to the server, means for storing the transmitted data in a database, means for cleansing the stored data, means for analyzing the cleansed data to grasp the resident's rhythm and health condition, means for generating an optimal care plan based on the grasped rhythm and health condition, and means for notifying the care staff of the generated care plan. This makes it possible to efficiently and accurately analyze resident data and provide an optimal care plan in real time.

[0486] "Resident behavior data" is information related to the actions and behaviors that residents perform in their daily lives.

[0487] "Conversation data" is information about the voices uttered by the resident and their contents.

[0488] "Means for collection" refers to devices and technologies for acquiring resident behavioral data and conversation data.

[0489] "Transmitting means" refers to the communication device or protocol used to transfer the collected data to the server.

[0490] "Means of storage" refers to a database or storage system for safely storing and managing the transmitted data.

[0491] A "cleansing method" is a process or algorithm for detecting, removing, or correcting incomplete data or noise.

[0492] "Means of analysis" refers to technology that analyzes cleansed data using machine learning algorithms, etc., to understand residents' behavioral patterns and health conditions.

[0493] "Means of understanding" refers to methods for recognizing and understanding the rhythms and health status of residents based on information obtained from analyzed data.

[0494] The "means of generation" refers to systems and algorithms that create optimal care plans for each resident based on their rhythms and health status.

[0495] The "notification means" refers to a communication means or interface for informing the care staff of the generated care plan in real time.

[0496] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. This system is realized using the following hardware and software.

[0497] Data Collection Phase

[0498] Device:

[0499] The terminal uses IoT sensors and voice recognition devices to detect the resident's actions and conversations. The IoT sensors detect when the resident gets up from the bed in real time, and the voice recognition device accurately recognizes when the resident says, "It's about time to get up." This data is sent from the terminal to the server.

[0500] Data storage phase

[0501] server:

[0502] The server receives the data sent from the device and stores it in a database for each resident. This database stores data in a format that includes resident ID, timestamp, behavioral data, conversation data, etc.

[0503] Data Cleansing Phase

[0504] server:

[0505] The server reads new data from the database and starts the data cleansing process, specifically detecting incomplete data and noise, and removing and correcting them. This process improves the accuracy of the analysis.

[0506] Data analysis phase

[0507] server:

[0508] The server uses machine learning algorithms to analyze the cleansed data, identifying resident behavior patterns and daily rhythms. For example, it can identify resident A's pattern of waking up at 7 a.m. every morning and evaluate this rhythm as healthy.

[0509] Care plan generation phase

[0510] server:

[0511] Based on the analysis results, the server automatically generates an optimal care plan for each resident. This care plan includes support for daily activities and medical intervention suggestions. For example, for a resident who frequently uses the toilet at night, it might include suggestions to review fluid intake and instructions to seek medical advice.

[0512] Care plan notification phase

[0513] server:

[0514] The generated care plan is notified to the care staff's device in real time.

[0515] User (care staff):

[0516] Care staff can then review the care plan sent to them via the device and provide specific care to the resident based on the plan's contents, such as praising the resident for getting up at 7 a.m. every morning and helping them prepare breakfast.

[0517] Specific examples

[0518] Device:

[0519] Resident A wakes up at 7am every morning, and the device detects when he or she says, "It's about time to get up," and sends this information to the server.

[0520] server:

[0521] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm.

[0522] Care staff (users):

[0523] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" are notified.

[0524] Example prompts for generative AI models

[0525] Example prompt sentence:

[0526] "I am designing a system that can analyze a resident's behavioral pattern, such as waking up at 7am every day. How can I use this system to assess the resident's health risks and generate and communicate an appropriate care plan?"

[0527] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities and reduce the burden on nursing staff.

[0528] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0529] Step 1:

[0530] The device collects data

[0531] Input: Resident actions and conversations

[0532] Output: Behavioral and conversational data

[0533] The device uses IoT sensors and voice recognition devices to monitor the resident's behavior and conversations in real time. Specifically, the IoT sensors detect when the resident gets up from the bed, and the voice recognition device records what the resident says. For example, it detects the voice saying, "It's about time to get up," and collects this as data.

[0534] Step 2:

[0535] The device sends the data to the server

[0536] Input: Behavioral and conversational data

[0537] Output: Data sent to the server

[0538] The collected behavioral and conversational data is sent to the server in batch format. Specifically, the data is packaged within the device and transferred to the server using a secure communication protocol. For example, data with timestamps is compiled into batches and sent to the server at regular intervals.

[0539] Step 3:

[0540] The server saves the data to a database

[0541] Input: Data sent

[0542] Output: Data stored in the database

[0543] The server receives the data sent from the device and stores it in a database for each resident. Specifically, it uses a database structure to store the resident ID, timestamp, behavioral data, conversation data, etc. For example, an entry such as "User ID: A, Time: 7:00 AM, Action: Wake up, Conversation: 'It's about time to get up'" is created.

[0544] Step 4:

[0545] The server cleanses the data

[0546] Input: Data stored in the database

[0547] Output: Cleansed data

[0548] The server reads new data from the database and performs a data cleansing process to detect, remove, and correct incomplete data and noise. Specifically, it complements missing values ​​and corrects outliers. For example, it filters noise from voice data to produce clear data.

[0549] Step 5:

[0550] The server analyzes the data

[0551] Input: Cleansed data

[0552] Output: Resident behavior patterns and health status

[0553] The server uses machine learning algorithms to analyze the cleansed data. This data analysis identifies the resident's behavioral patterns and daily rhythms. For example, it detects that resident A wakes up at 7:00 every morning. This information is then used to evaluate the resident's health status.

[0554] Step 6:

[0555] The server generates a care plan

[0556] Input: Behavioral patterns and health status

[0557] Output: Optimal care plan

[0558] Based on the analysis results, the server generates an optimal care plan for each resident. The generated care plan includes support for daily activities and medical intervention suggestions. For example, for a resident who frequently uses the toilet at night, it might include suggestions to review fluid intake and seek medical advice.

[0559] Step 7:

[0560] The server notifies the care plan

[0561] Input: Optimal Care Plan

[0562] Output: Care plan sent to the care staff's device

[0563] The generated care plan is sent to the care staff's device in real time. Specifically, a pop-up message or an alert is sent to the care staff's device via the notification system, allowing the necessary care to be carried out promptly.

[0564] Step 8:

[0565] The user (caregiver) performs care

[0566] Input: Notified Care Plan

[0567] Output: Specific care for residents

[0568] Care staff can then review the care plan sent to them via the device and provide specific care based on the plan, such as praising the resident for getting up at 7am every morning or helping them prepare breakfast.

[0569] (Application example 1)

[0570] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0571] Conventional factory robot maintenance and operation management is often performed at regular intervals, making it difficult to detect actual usage status and abnormalities in real time. This has led to maintenance delays and excessive maintenance work, making efficient operation difficult. It has also been impossible to properly collect instructions and comments from operators and generate optimal maintenance plans based on them.

[0572] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0573] In this invention, the server includes means for collecting behavioral data and conversation data of residents and automatic equipment, means for analyzing the collected data to grasp the work rhythms and operating status of the residents and automatic equipment, means for generating an optimal maintenance plan based on the grasped work rhythms and operating status, and means for notifying a manager or operator of the generated maintenance plan, thereby enabling the generation and notification of an optimal maintenance plan based on the operating status of factory robots and instructions from operators.

[0574] A "resident" is a person who lives within a particular facility or environment and about whom data is collected.

[0575] "Automatic equipment" refers to a device that automatically performs a task based on a control signal, and examples thereof include factory robots.

[0576] "Behavioral data" refers to information about the behavior and operation history of a target person or automated equipment, and is collected through sensors and devices.

[0577] "Conversation data" refers to information collected through a voice recognition device that includes statements and instructions from the target person or operator.

[0578] A "server" is a computer system for storing and analyzing collected data and communicating generated plans.

[0579] "Work rhythm" refers to the daily behavioral patterns and schedules of automated equipment and residents.

[0580] "Operational status" refers to information about the actual operating state of automated equipment and its performance.

[0581] "Maintenance plan" refers to the schedule and content of maintenance work required to keep equipment operating properly.

[0582] "Manager or operator" means a person responsible for the operation and management of automated equipment or systems.

[0583] The system according to the present invention collects and analyzes behavioral and conversational data of automated equipment and residents to understand work rhythms and operating conditions, and then generates and notifies optimal maintenance plans. Specific embodiments of the system are described below.

[0584] 1. Data Collection Phase

[0585] Device:

[0586] The terminal collects real-time operational data and operator conversation data from IoT sensors and voice recognition devices attached to automated equipment. For example, it can collect information on whether robots in a factory are operating normally or what operators say when giving maintenance instructions.

[0587] 2. Data analysis phase

[0588] server:

[0589] The server receives the data sent from the device and first cleans it. This removes noise and missing data to improve the accuracy of the analysis. Next, it uses machine learning algorithms to analyze the data and detect patterns and anomalies in the operation of automated equipment. For example, it can detect the phenomenon of a robot slowing down at certain times each day.

[0590] 3. Maintenance plan generation phase

[0591] server:

[0592] The server then generates an optimal maintenance plan based on the analysis results. This maintenance plan includes periodic part replacement for automated equipment and necessary actions when specific abnormalities occur. For example, if a robot's performance deteriorates, the server can add the replacement of a specific part to the schedule.

[0593] 4. Notification Phase

[0594] Device:

[0595] The generated maintenance plan is sent to the terminal of the manager or operator in real time, allowing the manager or operator to carry out the necessary maintenance work at the appropriate time. For example, a specific instruction such as "replace a specific part every weekend" may be sent to the operator's terminal.

[0596] Hardware and software used

[0597] IoT sensors: Collect operational data from the robot.

[0598] Voice recognition device: collects instructions and comments from the operator.

[0599] Server: Cleanses and analyzes the data.

[0600] Machine learning algorithms: Use Python's TensorFlow or PyTorch.

[0601] Database: Use SQLite or MySQL to manage collected data.

[0602] Specific examples

[0603] Terminal: When factory robot A needs specific maintenance every weekend, a voice recognition device detects an operator's instruction such as "Robot A is exhibiting a malfunction" and sends it to the server.

[0604] Server: The server stores the received data in a database and uses machine learning algorithms to analyze abnormal patterns of Robot A. For example, it can detect irregular behavior patterns and recommend maintenance before an abnormality occurs.

[0605] Manager or operator: The optimal maintenance plan is sent to the operator's terminal, with specific suggestions such as "replace specific parts every weekend."

[0606] Prompt Sentence Examples

[0607] "Develop a program that uses IoT sensors and voice recognition devices attached to factory robots to collect operational data and operator instructions regarding maintenance for Robot A, and generate an optimal maintenance plan. This program will include a function to detect anomalies and notify maintenance plans in real time. Use Python's TensorFlow."

[0608] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0609] Step 1:

[0610] The terminal collects operation data and operator conversation data from IoT sensors and voice recognition devices attached to the automated equipment. Specifically, it acquires the robot's operating status and voice data of instructions and comments given by the operator in real time and converts this data into digital format. Input: Operation data from the automated equipment's sensors, voice data from the voice recognition device. Output: Digital operation data and conversation data.

[0611] Step 2:

[0612] The terminal sends the collected digital data to the server. Specifically, it uses a data communication protocol to properly transfer the data and store it in a database on the server. Input: Digital operating data and conversation data. Output: Operating data and conversation data stored on the server.

[0613] Step 3:

[0614] The server cleanses the received data. Specifically, it removes incomplete data and noise and prepares the data to improve the accuracy of analysis. For example, it filters background noise from voice data and fills in missing values ​​in sensor data. Input: Operational data and conversation data stored on the server. Output: Cleansed data.

[0615] Step 4:

[0616] The server analyzes the cleansed data. Specifically, it uses machine learning algorithms to analyze the data and understand work rhythms and operating conditions. For example, it uses TensorFlow or PyTorch to learn the operation patterns of equipment and detect anomalies. Input: Cleansed data. Output: Analysis results (work rhythms, operating conditions, anomaly detection results).

[0617] Step 5:

[0618] The server generates an optimal maintenance plan based on the analysis results. Specifically, it automatically creates a schedule for maintenance timing and necessary part replacement based on the output of the machine learning model. Input: Analysis results (work rhythm, operating status, anomaly detection results). Output: Optimal maintenance plan.

[0619] Step 6:

[0620] The server notifies the administrator or operator of the generated maintenance plan. Specifically, it sends a real-time notification to the operator's terminal, presenting the necessary maintenance work. For example, a notification containing specific instructions such as "replace a specific part every weekend" is sent. Input: Optimized maintenance plan. Output: Maintenance instructions displayed on the operator's terminal.

[0621] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0622] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. Furthermore, it generates an optimal care plan based on this information and notifies the care staff, thereby providing appropriate care for the residents and reducing the care staff's workload. Furthermore, by combining it with an emotion engine, it can also recognize the user's emotions and provide emotionally appropriate care.

[0623] 1. Data Collection Phase

[0624] Device:

[0625] The terminals are equipped with IoT sensors and voice recognition devices to detect the residents' behavior and conversations. This allows them to accurately capture the actions and words that residents perform in their daily lives. For example, the terminals can detect and collect data on when residents wake up in the morning, when they eat meals, and even when they say they're hungry.

[0626] server:

[0627] The server receives and securely stores data sent from the device, which includes building a database for each resident and centrally managing behavioral and conversation data.

[0628] 2. Data analysis phase

[0629] server:

[0630] The server cleanses the collected data, removing incomplete data and noise, improving the accuracy of the analysis. Next, machine learning algorithms are used to analyze the data and identify the resident's behavioral patterns and daily rhythms. For example, it can detect patterns such as a resident waking up at roughly the same time every day and eating meals at the same time. The analysis results can also be used to assess the resident's health risks and detect irregular behavior or abnormal health signs.

[0631] Emotion engine integration:

[0632] Furthermore, an emotion engine is used to recognize the resident's emotions from conversation data. The voice recognition device analyzes the resident's tone of voice and choice of words to determine the resident's emotional state. For example, if a resident says, "I'm very tired today," the emotion engine analyzes this statement and recognizes that the resident is feeling fatigued.

[0633] 3. Care plan proposal phase

[0634] server:

[0635] Based on the analysis results, an optimal care plan is generated for each resident. The care plan includes suggestions for daily living support and medical interventions. For example, if a resident is prone to a particular health problem (such as nocturia), measures to improve that problem will be included in the plan. Care suggestions appropriate to the resident's emotional state are also generated based on emotional data obtained from the emotion engine. For example, if a resident is feeling fatigued, suggestions may include ensuring a relaxing environment and taking time to rest.

[0636] Care staff (users):

[0637] The generated care plan is sent to the care staff's devices in real time, allowing them to provide appropriate care for the resident, for example, by suggesting specific meal plans or planning additional health checks based on the resident's emotional state.

[0638] Specific examples

[0639] Device:

[0640] Resident A wakes up at 7:00 every morning, and the voice recognition device detects utterances such as "It's about time to get up" and sends them to the server.

[0641] server:

[0642] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm. If the emotion engine detects fatigue from Resident A's tone of voice, that emotion data is also stored on the server for analysis.

[0643] Care staff (users):

[0644] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time and emotional state. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" and "providing a relaxing environment" are provided.

[0645] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities, reduce the burden on nursing staff, and provide care that takes into consideration the emotions of users.

[0646] The processing flow will be explained below.

[0647] Program processing steps

[0648] 1. Data Collection Phase

[0649] Step 1:

[0650] The device uses IoT sensors to detect the resident's behavior, specifically capturing the resident's movements in real time, such as getting up from the bed or moving around the room.

[0651] Step 2:

[0652] The device detects the resident's conversation using a voice recognition device. For example, if a resident says "I'm hungry," this voice data is collected.

[0653] Step 3:

[0654] The device transmits the acquired behavioral and conversation data to a server, where the data is categorized and stored.

[0655] Step 4:

[0656] The device analyzes the resident's voice data and requests the emotion engine to determine their emotional state, for example, by detecting their emotional state such as "happiness" or "sadness" from their tone of voice and vocabulary.

[0657] 2. Data analysis phase

[0658] Step 5:

[0659] The server stores the received data in a database, where it is stored together with past data in a centralized management system for each resident.

[0660] Step 6:

[0661] The server cleanses the collected data, removing incomplete data and noise, a process that improves the accuracy of the analysis.

[0662] Step 7:

[0663] The server uses machine learning algorithms to analyze the data and extract patterns and rhythms of the resident's daily behavior, such as identifying a regular pattern where the resident wakes up at 7 a.m. and eats breakfast at 8 a.m.

[0664] Step 8:

[0665] The server evaluates the health risks of residents based on the analysis results. For example, if a pattern of frequent toilet visits at night is detected, the server evaluates the risk.

[0666] Step 9:

[0667] The server also analyzes emotional data from the emotion engine. For example, if a resident is feeling fatigued, that information is added to the risk assessment.

[0668] 3. Care plan proposal phase

[0669] Step 10:

[0670] The server generates an optimal care plan based on the analysis results. For example, if a resident is at risk of nocturia, it generates a plan that includes adjusting fluid intake and suggesting a doctor's appointment.

[0671] Step 11:

[0672] The server generates a care plan based on the emotional data, taking into account the resident's emotional state. For example, if the resident is feeling fatigued, it will include suggestions for extending rest time and relaxation.

[0673] Step 12:

[0674] The server notifies care staff of the care plan and risk assessment results in real time, including specific care suggestions.

[0675] Step 13:

[0676] Care staff (users) receive notifications and review the care plan, for example, making changes to the resident's diet or planning additional health checks.

[0677] Step 14:

[0678] Care staff carry out actual care based on the care plan, such as providing guidance on fluid intake and creating an environment for relaxation.

[0679] Specific examples

[0680] Step 1:

[0681] The device uses an IoT sensor to detect that resident A gets out of bed at 7am.

[0682] Step 2:

[0683] The device detects the voice of resident A saying, "I'm a little tired today."

[0684] Step 3:

[0685] The terminal transmits the acquired behavioral data and conversation data to the server.

[0686] Step 4:

[0687] The device sends the detected voice data to the emotion engine, which determines resident A's emotional state as "fatigue."

[0688] Step 5:

[0689] The server stores the received data in a database.

[0690] Step 6:

[0691] The server cleanses the collected data.

[0692] Step 7:

[0693] The server analyzes the data and identifies the behavioral patterns of Resident A. For example, it detects that Resident A wakes up at 7:00 every morning and has breakfast at 8:00.

[0694] Step 8:

[0695] The server evaluates health risks and determines that resident A is at risk of nocturia.

[0696] Step 9:

[0697] The server adds "fatigue" data from the emotion engine to the evaluation.

[0698] Step 10:

[0699] The server generates an optimal care plan.

[0700] Step 11:

[0701] The server generates a care plan based on the emotional state, for example, including relaxation recommendations for Resident A who feels fatigued.

[0702] Step 12:

[0703] The server notifies the care staff of the generated care plan.

[0704] Step 13:

[0705] The care staff (user) receives a notification and checks the care plan.

[0706] Step 14:

[0707] Nursing staff carry out actual care based on the care plan.

[0708] Through these specific processing steps, the AI ​​nursing care support system can efficiently grasp the resident's rhythm and health status, provide an optimal care plan, and utilize an emotion engine to provide emotion-based care.

[0709] Example 2

[0710] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0711] As the aging population continues to grow, improving the quality of nursing care services and reducing the workload of nursing staff at nursing facilities have become important issues. There is also a need to provide appropriate care tailored to residents' emotions and health conditions. However, current systems lack sufficient analysis of residents' behavioral and conversational data, making it difficult to provide care plans that take emotions into account. Therefore, a system that can provide efficient and effective care is needed.

[0712] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting resident behavioral information and speech information, means for analyzing the collected information to understand the resident's rhythm and health condition, means for generating an optimal care plan based on the understood rhythm and health condition, means for notifying care staff of the generated care plan, means for analyzing the resident's emotions using a voice recognition device and an emotion analysis engine, and means for adjusting the care plan based on the results of the emotion analysis. This makes it possible to efficiently understand the resident's health condition based on data obtained from the resident's behavior and speech and to generate a care plan accordingly. Furthermore, by adjusting the care content based on the resident's emotions, it is possible to improve the quality of care services and reduce the workload of care staff.

[0713] "Residents" refers to elderly people who live in nursing homes or senior housing, and people who require nursing care support.

[0714] "Behavioral information" refers to data on specific actions taken by residents in their daily lives (for example, waking up, eating, moving around, etc.).

[0715] "Speech information" refers to the words and conversations that residents make in their daily lives.

[0716] "Means of collection" refers to devices and systems for collecting behavioral information and speech information using sensors, voice recognition devices, etc.

[0717] "Means of analysis" refers to methods and systems for analyzing collected data to understand residents' behavioral patterns, health status, emotions, etc.

[0718] "Rhythm" refers to the temporal patterns and habits of residents' daily lives.

[0719] "Health status" refers to the state of a resident's physical and mental health.

[0720] A "care plan" refers to a plan that specifically outlines the optimal content and methods of care to be provided to residents based on the results of the analysis.

[0721] "Care workers" refers to employees and staff who provide care services to residents.

[0722] "Means for notifying" refers to a method or system for notifying care staff of the generated care plan.

[0723] A "voice recognition device" refers to a device or system that detects what residents say and converts it into text data.

[0724] "Emotion analysis engine" refers to an algorithm or system that analyzes data obtained from a voice recognition device to estimate the emotional state of a resident.

[0725] "Results of emotion analysis" refers to the emotional state of the resident analyzed by the emotion analysis engine.

[0726] "Adjustment means" refers to a method or system for appropriately changing the content or method of the care plan based on the analyzed emotional data.

[0727] The AI ​​nursing care support system of this invention collects and analyzes information on residents' behavior and speech to understand their rhythm, health condition, and emotions. Furthermore, it generates an optimal care plan based on this information and notifies caregivers, thereby providing appropriate care for residents and reducing the caregiver's workload.

[0728] Data collection

[0729] Device:

[0730] The devices are equipped with IoT sensors and voice recognition devices to detect the resident's actions and comments. This allows them to accurately capture the actions and comments that residents make in their daily lives. For example, if a resident wakes up at 7:00 a.m. and says, "It's about time to get up," the voice recognition device will detect this. Motion sensors are also used to collect data on the resident's movements.

[0731] Data transmission and storage

[0732] Device:

[0733] The device sends the collected behavioral information and speech information to a server using a secure communication protocol (e.g., HTTPS).

[0734] server:

[0735] The server receives the data sent from the devices and stores it in a database for each resident, which records behavioral and conversation data along with timestamps.

[0736] Data Cleansing and Analysis

[0737] server:

[0738] The server cleanses the received data, removing incomplete data and noise. Specifically, it filters out extremely short audio data and erroneous sensor values. Once cleansed, the data is analyzed using machine learning algorithms. For example, it can detect a behavioral pattern, such as "Resident A wakes up at 7:00 every morning."

[0739] Emotional Data Analysis

[0740] server:

[0741] The server uses an emotion analysis engine to analyze the resident's emotions from the collected speech information. Data obtained from the voice recognition device is input into the emotion analysis engine to determine the resident's emotional state. For example, a statement such as "I'm very tired today" is analyzed and recognized as indicating that the resident is feeling fatigued.

[0742] Generate a care plan

[0743] server:

[0744] The server generates an optimal care plan for each resident based on the analyzed behavioral patterns and emotional data. The care plan includes support for daily life and medical intervention suggestions. For example, if a resident wakes up at 7 a.m. every morning and feels tired, the server will create a care plan that provides a relaxing environment after waking up.

[0745] Care plan notification

[0746] server:

[0747] The server then sends the generated care plan to the caregiver's device, where the caregiver can check the plan in real time on a tablet or smartphone.

[0748] Care delivery

[0749] Caregiver (user):

[0750] Care staff provide care to residents based on the care plan they receive, for example, praising residents for getting up at 7 a.m. every morning, helping them prepare breakfast, and providing a relaxing environment.

[0751] Prompt Sentence Examples

[0752] "Resident A wakes up at 7:00 every morning, and the voice recognition device detects that he says, 'It's about time to get up.' Please also analyze Resident A's emotional data and generate the optimal care plan."

[0753] In this way, the present invention improves the quality of nursing care services in nursing facilities and reduces the workload of nursing staff. By providing care based on the feelings of residents, more detailed care becomes possible.

[0754] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0755] Step 1: Sensor configuration and data collection

[0756] Terminal: The terminal collects information on the residents' behavior and speech. Specifically, it uses IoT sensors and voice recognition devices. The sensors detect the residents' movements and actions, and the voice recognition device records what the residents say. The input is raw data obtained from the sensors and voice device. The output is recorded as the residents' behavior data and speech data. For example, if a resident wakes up at 7am, the device detects when they say, "It's about time to get up."

[0757] Step 2: Data transmission and storage

[0758] Terminal: The terminal sends the collected behavioral and speech information to the server. The input is the data collected in step 1. The output is a message confirming the completion of the transmission. The data is sent to the server via a secure communication protocol (such as HTTPS).

[0759] Server: The server receives the data sent from the device and stores it in a database. The input is the data sent from the device. The output is the behavioral data and speech data stored in the database. This records the data with a timestamp and makes it available for subsequent analysis.

[0760] Step 3: Data cleansing and preprocessing

[0761] Server: The server cleanses the received data and removes incomplete data and noise. The input is the data stored in step 2. The output is the cleansed data. Specifically, it filters out extremely short audio data and erroneous sensor values.

[0762] Step 4: Data analysis

[0763] Server: The server uses a machine learning algorithm to analyze the cleansed data and understand the resident's behavioral patterns and health status. The input is the clean data obtained in step 3. The output is the analysis results of the resident's behavioral patterns and health status. For example, a pattern such as "Resident A wakes up at 7 o'clock every morning" is detected.

[0764] Step 5: Analyze the sentiment data

[0765] Server: The server uses an emotion analysis engine to analyze the resident's emotions from the utterance data. The input is the utterance data obtained in step 3. The output is the analysis result of the resident's emotional state. For example, the emotion analysis engine analyzes the utterance "I'm very tired today" and outputs the result as "Feeling tired."

[0766] Step 6: Generate a care plan

[0767] Server: The server generates an optimal care plan for each resident based on the analyzed behavioral patterns and emotional data. The input is the analysis results from steps 4 and 5. The output is a care plan for each resident. For example, if a resident wakes up at 7 a.m. every morning and feels tired, a care plan is generated that provides a relaxing environment after waking up.

[0768] Step 7: Inform the plan of care

[0769] Server: The server notifies the care staff device of the care plan it has generated. The input is the care plan generated in step 6. The output is a push notification to the care staff device. This allows the care staff to check the care plan in real time on devices such as tablets and smartphones.

[0770] Step 8: Implementing care

[0771] Care staff (user): Care staff provide care to residents based on the notified care plan. The input is the care plan notified from the server. The output is a record of the care provided. For example, they may praise a resident for waking up at 7am every morning, help them prepare breakfast, and provide a relaxing environment.

[0772] (Application example 2)

[0773] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0774] Current care support systems are effective in collecting residents' behavioral and conversational data, understanding their rhythms and health status based on that data, and generating optimal care plans. However, they have limitations in providing comprehensive health support based on residents' health status and eating patterns. They lack specific suggestions for residents to eat an appropriate diet and the integration of food delivery services to implement those suggestions. Therefore, an effective means of ensuring residents receive optimal nutrition and maintain their health is needed.

[0775] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting resident behavioral data and conversation data, means for analyzing the collected data to understand the resident's rhythm and health condition, means for generating an optimal care plan based on the understood rhythm and health condition, means for notifying care staff of the generated care plan, means for proposing an optimal meal menu based on the user's eating patterns and health condition, and means for notifying a delivery service of the proposed meal menu. This enables comprehensive health support for residents, and by linking appropriate nutritional intake with delivery services, it becomes possible to improve the quality of life of residents and reduce the workload of care staff.

[0776] "Resident" means an individual who resides in a particular facility or residence.

[0777] "Behavioral data" refers to data that indicates the resident's movements and activities in daily life.

[0778] "Conversation data" refers to data collected as text information of the sounds and words uttered by residents.

[0779] "Rhythm" refers to the regular patterns of behavior and lifestyle of residents.

[0780] "Health status" refers to information related to the resident's physical condition and health.

[0781] A "care plan" is a written plan that lists specific care methods and support for residents.

[0782] "Care staff" refers to people whose occupations support the health and lives of residents.

[0783] "Dietary patterns" refer to the habits and tendencies of residents regarding what meals they eat and at what times.

[0784] "Health Status" refers to the state of a resident's physical and mental health.

[0785] A "meal menu" refers to the specific meal contents and menu to be provided to residents.

[0786] "Delivery service" refers to a service for delivering meal menus to residents.

[0787] This invention is a system that collects and analyzes behavioral and conversational data of residents to understand their rhythms and health status. It also generates an optimal care plan based on this data and notifies care staff, providing appropriate care and reducing their workload. It also includes a function to suggest optimal meal menus based on eating patterns and health status and notify the delivery service.

[0788] The server first collects behavioral and conversational data from residents. Behavioral data includes residents' movements and daily activities, while conversational data includes what residents say. This data is collected using smartphones and IoT sensors and sent to the server. A voice recognition API, commonly known as a voice recognition engine, is used to convert the voice data into text data.

[0789] The server cleanses the collected data, removing incomplete data and noise, and then analyzes it using machine learning algorithms to understand the resident's rhythm and health status. This analysis is performed using a generic machine learning platform. For example, if the server detects that a resident has a habit of eating at a specific time, it analyzes their eating patterns. It also evaluates the resident's emotional state using an emotion analysis engine.

[0790] The server then generates an optimal care plan based on the analysis, including assisted living and medical recommendations. It also suggests optimal meal plans based on the resident's eating patterns and emotional state, and notifies a food delivery service. A food delivery API is used to arrange for the optimal menu to be delivered to the resident.

[0791] This system is used in the following specific example.

[0792] When a resident says "I'm hungry," the behavioral data collection module detects this and converts it into text using a speech recognition engine. The data cleansing module processes this text data, and the behavioral pattern analysis module analyzes eating patterns. The emotion analysis module evaluates the resident's emotional state, and the menu generation module generates an optimal meal menu. The delivery notification module notifies the resident of delivery suggestions using a food delivery API.

[0793] Example prompt sentence:

[0794] When a user says "I'm hungry," the behavioral data collection module detects this and converts it into text using a speech recognition engine. The data cleansing module processes this text data, and the behavioral pattern analysis module analyzes eating patterns. The emotion analysis module evaluates the user's emotional state, and the menu generation module generates an optimal meal menu. The delivery notification module notifies the user of delivery suggestions using a food delivery API.

[0795] This system will enable comprehensive health support for residents, improving their quality of life through proper nutritional intake and linking delivery services, and reducing the workload of care staff.

[0796] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0797] Step 1:

[0798] The terminal uses IoT sensors and voice recognition devices to collect resident behavioral and conversational data. For example, if a resident says "I'm hungry," the voice is detected and converted into text data by a voice recognition engine. The input for this step is the resident's behavior and speech, and the output is text data and behavioral data.

[0799] Step 2:

[0800] The behavioral and conversational data sent from the device is sent to a server and securely stored. This involves building and centrally managing a database for each specific resident. The input for this step is text data and behavioral data, and the output is the results stored in the database.

[0801] Step 3:

[0802] The server cleanses the collected data and removes incomplete data and noise. This is a data preprocessing stage and is an important step for accurate analysis. Specifically, it detects outliers and missing data and removes or imputes them. The input for this step is the stored behavioral data and text data, and the output is the cleansed data.

[0803] Step 4:

[0804] Based on the cleansed data, the server analyzes the data using machine learning algorithms. Common machine learning platforms such as TensorFlow and PyTorch are used here. The analysis identifies resident behavior patterns and daily rhythms and assesses health risks. The input for this step is the cleansed data, and the output is behavior patterns and health risk assessment results.

[0805] Step 5:

[0806] Using an emotion analysis engine, the server recognizes the resident's emotions from the conversation data. Specifically, it uses data from a voice recognition device and uses an emotion analysis API to determine the resident's emotional state from the tone of voice and vocabulary. The input for this step is the conversation data, and the output is the emotion analysis results.

[0807] Step 6:

[0808] Based on the analysis results, the server generates an optimal care plan for each resident. This care plan includes specific suggestions for health management and meal plans. The analysis results are integrated with the emotional state to suggest the optimal meal menu for the resident. This results in a care plan that corresponds to the resident's health condition. The inputs for this step are behavioral patterns, health risk assessment, and emotional analysis results, and the output is a care plan and meal menu suggestions.

[0809] Step 7:

[0810] The generated care plan and meal menu are notified to the nursing staff and delivery service. Using the food delivery API, arrangements are made to quickly deliver the optimal meal menu for the resident. Nursing staff are notified of the resident's status in real time, helping them respond quickly. The input to this step is the care plan and meal menu suggestions, and the output is the notification results.

[0811] Through these steps, this system will provide comprehensive health support for residents, reduce the workload of care staff, and provide residents with appropriate nutritional intake.

[0812] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0813] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0814] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0815] [Third embodiment]

[0816] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0817] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0818] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0819] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0820] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0821] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0822] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0823] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0824] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0825] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0826] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0827] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0828] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. Furthermore, it generates an optimal care plan based on this information and notifies the care staff, thereby providing appropriate care to residents and reducing the care staff's workload.

[0829] 1. Data Collection Phase

[0830] Device:

[0831] The terminals are equipped with IoT sensors and voice recognition devices to detect the residents' behavior and conversations. This allows them to accurately capture the actions and words that residents perform in their daily lives. For example, the terminals can detect and collect data on when residents wake up in the morning, when they eat meals, and even when they say they're hungry.

[0832] server:

[0833] The server receives and securely stores data sent from the device, which includes building a database for each resident and centrally managing behavioral and conversation data.

[0834] 2. Data analysis phase

[0835] server:

[0836] The server cleanses the collected data, removing incomplete data and noise, improving the accuracy of the analysis. Next, machine learning algorithms are used to analyze the data and identify the resident's behavioral patterns and daily rhythms. For example, it can detect patterns such as a resident waking up at roughly the same time every day and eating meals at the same time. The analysis results can also be used to assess the resident's health risks and detect irregular behavior or abnormal health signs.

[0837] 3. Care plan proposal phase

[0838] server:

[0839] Based on the analysis results, an optimal care plan is generated for each resident. The care plan includes suggestions for daily living support and medical intervention. For example, if a resident's frequent trips to the toilet at night are detected, the server will generate a care plan suggesting a review of fluid intake and a doctor's consultation.

[0840] Care staff (users):

[0841] The generated care plan is sent to the care staff's devices in real time, allowing them to provide appropriate care for the resident, such as suggesting specific meal plans and conducting regular health checks.

[0842] Specific examples

[0843] Device:

[0844] Resident A wakes up at 7:00 every morning, and the voice recognition device detects utterances such as "It's about time to get up" and sends them to the server.

[0845] server:

[0846] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm.

[0847] Care staff (users):

[0848] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" are notified.

[0849] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities and reduce the burden on nursing staff.

[0850] The processing flow will be explained below.

[0851] Program processing steps

[0852] 1. Data Collection Phase

[0853] Step 1:

[0854] The device detects the resident's actions and conversations. For example, it uses IoT sensors and voice recognition devices to capture the time the resident wakes up in the morning or when they say they're hungry.

[0855] Step 2:

[0856] The data detected by the device is sent in real time to a server, where it is classified as behavioral data or conversation data.

[0857] Step 3:

[0858] The server stores the received data in a database. The data is organized for each resident and stored together with past data.

[0859] 2. Data analysis phase

[0860] Step 4:

[0861] The server cleanses the collected data, removing incomplete data and noise to improve the accuracy of the analysis.

[0862] Step 5:

[0863] The server analyzes the data using machine learning algorithms to extract the resident's daily behavioral patterns and identify regularities, such as waking up at 7am and having breakfast at 8am every day.

[0864] Step 6:

[0865] The server uses the analysis results to understand the resident's rhythm and health status, for example, by analyzing sleep and eating patterns, and assessing the health risks if any abnormalities are found.

[0866] 3. Care plan proposal phase

[0867] Step 7:

[0868] The server generates an optimal care plan based on the analysis results. For example, if a resident is prone to a particular health problem (such as frequent urination at night), the plan will include measures to improve that problem.

[0869] Step 8:

[0870] The server generates a care plan and notifies the care staff, including specific improvement measures and daily care suggestions.

[0871] Step 9:

[0872] Care staff are notified and review the care plan, including making dietary changes or additional health checks to accommodate each resident's health condition.

[0873] Step 10:

[0874] Nursing staff carry out actual care based on the care plan. For example, for a resident with nocturia, they may provide guidance on fluid intake and arrange appointments with medical institutions.

[0875] Through these steps, the AI ​​nursing care support system efficiently grasps the rhythm and health status of residents and provides optimal care plans, thereby reducing the burden on nursing staff and improving the quality of nursing care services.

[0876] Example 1

[0877] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0878] Conventional nursing care support systems face challenges in efficiently collecting and analyzing resident behavior and conversation data and notifying nursing staff of optimal care plans in real time. Furthermore, collected data often contains noise and incomplete data, necessitating data cleansing. However, this manual process requires time and effort, so there is a demand for more accurate analysis and faster response.

[0879] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0880] In this invention, the server includes means for collecting resident behavioral data and conversation data, means for transmitting the collected data to the server, means for storing the transmitted data in a database, means for cleansing the stored data, means for analyzing the cleansed data to grasp the resident's rhythm and health condition, means for generating an optimal care plan based on the grasped rhythm and health condition, and means for notifying the care staff of the generated care plan. This makes it possible to efficiently and accurately analyze resident data and provide an optimal care plan in real time.

[0881] "Resident behavior data" is information related to the actions and behaviors that residents perform in their daily lives.

[0882] "Conversation data" is information about the voices uttered by the resident and their contents.

[0883] "Means for collection" refers to devices and technologies for acquiring resident behavioral data and conversation data.

[0884] "Transmitting means" refers to the communication device or protocol used to transfer the collected data to the server.

[0885] "Means of storage" refers to a database or storage system for safely storing and managing the transmitted data.

[0886] A "cleansing method" is a process or algorithm for detecting, removing, or correcting incomplete data or noise.

[0887] "Means of analysis" refers to technology that analyzes cleansed data using machine learning algorithms, etc., to understand residents' behavioral patterns and health conditions.

[0888] "Means of understanding" refers to methods for recognizing and understanding the rhythms and health status of residents based on information obtained from analyzed data.

[0889] The "means of generation" refers to systems and algorithms that create optimal care plans for each resident based on their rhythms and health status.

[0890] The "notification means" refers to a communication means or interface for informing the care staff of the generated care plan in real time.

[0891] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. This system is realized using the following hardware and software.

[0892] Data Collection Phase

[0893] Device:

[0894] The terminal uses IoT sensors and voice recognition devices to detect the resident's actions and conversations. The IoT sensors detect when the resident gets up from the bed in real time, and the voice recognition device accurately recognizes when the resident says, "It's about time to get up." This data is sent from the terminal to the server.

[0895] Data storage phase

[0896] server:

[0897] The server receives the data sent from the device and stores it in a database for each resident. This database stores data in a format that includes resident ID, timestamp, behavioral data, conversation data, etc.

[0898] Data Cleansing Phase

[0899] server:

[0900] The server reads new data from the database and starts the data cleansing process, specifically detecting incomplete data and noise, and removing and correcting them. This process improves the accuracy of the analysis.

[0901] Data analysis phase

[0902] server:

[0903] The server uses machine learning algorithms to analyze the cleansed data, identifying resident behavior patterns and daily rhythms. For example, it can identify resident A's pattern of waking up at 7 a.m. every morning and evaluate this rhythm as healthy.

[0904] Care plan generation phase

[0905] server:

[0906] Based on the analysis results, the server automatically generates an optimal care plan for each resident. This care plan includes support for daily activities and medical intervention suggestions. For example, for a resident who frequently uses the toilet at night, it might include suggestions to review fluid intake and instructions to seek medical advice.

[0907] Care plan notification phase

[0908] server:

[0909] The generated care plan is notified to the care staff's device in real time.

[0910] User (care staff):

[0911] Care staff can then review the care plan sent to them via the device and provide specific care to the resident based on the plan's contents, such as praising the resident for getting up at 7 a.m. every morning and helping them prepare breakfast.

[0912] Specific examples

[0913] Device:

[0914] Resident A wakes up at 7am every morning, and the device detects when he or she says, "It's about time to get up," and sends this information to the server.

[0915] server:

[0916] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm.

[0917] Care staff (users):

[0918] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" are notified.

[0919] Example prompts for generative AI models

[0920] Example prompt sentence:

[0921] "I am designing a system that can analyze a resident's behavioral pattern, such as waking up at 7am every day. How can I use this system to assess the resident's health risks and generate and communicate an appropriate care plan?"

[0922] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities and reduce the burden on nursing staff.

[0923] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0924] Step 1:

[0925] The device collects data

[0926] Input: Resident actions and conversations

[0927] Output: Behavioral and conversational data

[0928] The device uses IoT sensors and voice recognition devices to monitor the resident's behavior and conversations in real time. Specifically, the IoT sensors detect when the resident gets up from the bed, and the voice recognition device records what the resident says. For example, it detects the voice saying, "It's about time to get up," and collects this as data.

[0929] Step 2:

[0930] The device sends the data to the server

[0931] Input: Behavioral and conversational data

[0932] Output: Data sent to the server

[0933] The collected behavioral and conversational data is sent to the server in batch format. Specifically, the data is packaged within the device and transferred to the server using a secure communication protocol. For example, data with timestamps is compiled into batches and sent to the server at regular intervals.

[0934] Step 3:

[0935] The server saves the data to a database

[0936] Input: Data sent

[0937] Output: Data stored in the database

[0938] The server receives the data sent from the device and stores it in a database for each resident. Specifically, it uses a database structure to store the resident ID, timestamp, behavioral data, conversation data, etc. For example, an entry such as "User ID: A, Time: 7:00 AM, Action: Wake up, Conversation: 'It's about time to get up'" is created.

[0939] Step 4:

[0940] The server cleanses the data

[0941] Input: Data stored in the database

[0942] Output: Cleansed data

[0943] The server reads new data from the database and performs a data cleansing process to detect, remove, and correct incomplete data and noise. Specifically, it complements missing values ​​and corrects outliers. For example, it filters noise from voice data to produce clear data.

[0944] Step 5:

[0945] The server analyzes the data

[0946] Input: Cleansed data

[0947] Output: Resident behavior patterns and health status

[0948] The server uses machine learning algorithms to analyze the cleansed data. This data analysis identifies the resident's behavioral patterns and daily rhythms. For example, it detects that resident A wakes up at 7:00 every morning. This information is then used to evaluate the resident's health status.

[0949] Step 6:

[0950] The server generates a care plan

[0951] Input: Behavioral patterns and health status

[0952] Output: Optimal care plan

[0953] Based on the analysis results, the server generates an optimal care plan for each resident. The generated care plan includes support for daily activities and medical intervention suggestions. For example, for a resident who frequently uses the toilet at night, it might include suggestions to review fluid intake and seek medical advice.

[0954] Step 7:

[0955] The server notifies the care plan

[0956] Input: Optimal Care Plan

[0957] Output: Care plan sent to the care staff's device

[0958] The generated care plan is sent to the care staff's device in real time. Specifically, a pop-up message or an alert is sent to the care staff's device via the notification system, allowing the necessary care to be carried out promptly.

[0959] Step 8:

[0960] The user (caregiver) performs care

[0961] Input: Notified Care Plan

[0962] Output: Specific care for residents

[0963] Care staff can then review the care plan sent to them via the device and provide specific care based on the plan, such as praising the resident for getting up at 7am every morning or helping them prepare breakfast.

[0964] (Application example 1)

[0965] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0966] Conventional factory robot maintenance and operation management is often performed at regular intervals, making it difficult to detect actual usage status and abnormalities in real time. This has led to maintenance delays and excessive maintenance work, making efficient operation difficult. It has also been impossible to properly collect instructions and comments from operators and generate optimal maintenance plans based on them.

[0967] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0968] In this invention, the server includes means for collecting behavioral data and conversation data of residents and automatic equipment, means for analyzing the collected data to grasp the work rhythms and operating status of the residents and automatic equipment, means for generating an optimal maintenance plan based on the grasped work rhythms and operating status, and means for notifying a manager or operator of the generated maintenance plan, thereby enabling the generation and notification of an optimal maintenance plan based on the operating status of factory robots and instructions from operators.

[0969] A "resident" is a person who lives within a particular facility or environment and about whom data is collected.

[0970] "Automatic equipment" refers to a device that automatically performs a task based on a control signal, and examples thereof include factory robots.

[0971] "Behavioral data" refers to information about the behavior and operation history of a target person or automated equipment, and is collected through sensors and devices.

[0972] "Conversation data" refers to information collected through a voice recognition device that includes statements and instructions from the target person or operator.

[0973] A "server" is a computer system for storing and analyzing collected data and communicating generated plans.

[0974] "Work rhythm" refers to the daily behavioral patterns and schedules of automated equipment and residents.

[0975] "Operational status" refers to information about the actual operating state of automated equipment and its performance.

[0976] "Maintenance plan" refers to the schedule and content of maintenance work required to keep equipment operating properly.

[0977] "Manager or operator" means a person responsible for the operation and management of automated equipment or systems.

[0978] The system according to the present invention collects and analyzes behavioral and conversational data of automated equipment and residents to understand work rhythms and operating conditions, and then generates and notifies optimal maintenance plans. Specific embodiments of the system are described below.

[0979] 1. Data Collection Phase

[0980] Device:

[0981] The terminal collects real-time operational data and operator conversation data from IoT sensors and voice recognition devices attached to automated equipment. For example, it can collect information on whether robots in a factory are operating normally or what operators say when giving maintenance instructions.

[0982] 2. Data analysis phase

[0983] server:

[0984] The server receives the data sent from the device and first cleans it. This removes noise and missing data to improve the accuracy of the analysis. Next, it uses machine learning algorithms to analyze the data and detect patterns and anomalies in the operation of automated equipment. For example, it can detect the phenomenon of a robot slowing down at certain times each day.

[0985] 3. Maintenance plan generation phase

[0986] server:

[0987] The server then generates an optimal maintenance plan based on the analysis results. This maintenance plan includes periodic part replacement for automated equipment and necessary actions when specific abnormalities occur. For example, if a robot's performance deteriorates, the server can add the replacement of a specific part to the schedule.

[0988] 4. Notification Phase

[0989] Device:

[0990] The generated maintenance plan is sent to the terminal of the manager or operator in real time, allowing the manager or operator to carry out the necessary maintenance work at the appropriate time. For example, a specific instruction such as "replace a specific part every weekend" may be sent to the operator's terminal.

[0991] Hardware and software used

[0992] IoT sensors: Collect operational data from the robot.

[0993] Voice recognition device: collects instructions and comments from the operator.

[0994] Server: Cleanses and analyzes the data.

[0995] Machine learning algorithms: Use Python's TensorFlow or PyTorch.

[0996] Database: Use SQLite or MySQL to manage collected data.

[0997] Specific examples

[0998] Terminal: When factory robot A needs specific maintenance every weekend, a voice recognition device detects an operator's instruction such as "Robot A is exhibiting a malfunction" and sends it to the server.

[0999] Server: The server stores the received data in a database and uses machine learning algorithms to analyze abnormal patterns of Robot A. For example, it can detect irregular behavior patterns and recommend maintenance before an abnormality occurs.

[1000] Manager or operator: The optimal maintenance plan is sent to the operator's terminal, with specific suggestions such as "replace specific parts every weekend."

[1001] Prompt Sentence Examples

[1002] "Develop a program that uses IoT sensors and voice recognition devices attached to factory robots to collect operational data and operator instructions regarding maintenance for Robot A, and generate an optimal maintenance plan. This program will include a function to detect anomalies and notify maintenance plans in real time. Use Python's TensorFlow."

[1003] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1004] Step 1:

[1005] The terminal collects operation data and operator conversation data from IoT sensors and voice recognition devices attached to the automated equipment. Specifically, it acquires the robot's operating status and voice data of instructions and comments given by the operator in real time and converts this data into digital format. Input: Operation data from the automated equipment's sensors, voice data from the voice recognition device. Output: Digital operation data and conversation data.

[1006] Step 2:

[1007] The terminal sends the collected digital data to the server. Specifically, it uses a data communication protocol to properly transfer the data and store it in a database on the server. Input: Digital operating data and conversation data. Output: Operating data and conversation data stored on the server.

[1008] Step 3:

[1009] The server cleanses the received data. Specifically, it removes incomplete data and noise and prepares the data to improve the accuracy of analysis. For example, it filters background noise from voice data and fills in missing values ​​in sensor data. Input: Operational data and conversation data stored on the server. Output: Cleansed data.

[1010] Step 4:

[1011] The server analyzes the cleansed data. Specifically, it uses machine learning algorithms to analyze the data and understand work rhythms and operating conditions. For example, it uses TensorFlow or PyTorch to learn the operation patterns of equipment and detect anomalies. Input: Cleansed data. Output: Analysis results (work rhythms, operating conditions, anomaly detection results).

[1012] Step 5:

[1013] The server generates an optimal maintenance plan based on the analysis results. Specifically, it automatically creates a schedule for maintenance timing and necessary part replacement based on the output of the machine learning model. Input: Analysis results (work rhythm, operating status, anomaly detection results). Output: Optimal maintenance plan.

[1014] Step 6:

[1015] The server notifies the administrator or operator of the generated maintenance plan. Specifically, it sends a real-time notification to the operator's terminal, presenting the necessary maintenance work. For example, a notification containing specific instructions such as "replace a specific part every weekend" is sent. Input: Optimized maintenance plan. Output: Maintenance instructions displayed on the operator's terminal.

[1016] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1017] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. Furthermore, it generates an optimal care plan based on this information and notifies the care staff, thereby providing appropriate care for the residents and reducing the care staff's workload. Furthermore, by combining it with an emotion engine, it can also recognize the user's emotions and provide emotionally appropriate care.

[1018] 1. Data Collection Phase

[1019] Device:

[1020] The terminals are equipped with IoT sensors and voice recognition devices to detect the residents' behavior and conversations. This allows them to accurately capture the actions and words that residents perform in their daily lives. For example, the terminals can detect and collect data on when residents wake up in the morning, when they eat meals, and even when they say they're hungry.

[1021] server:

[1022] The server receives and securely stores data sent from the device, which includes building a database for each resident and centrally managing behavioral and conversation data.

[1023] 2. Data analysis phase

[1024] server:

[1025] The server cleanses the collected data, removing incomplete data and noise, improving the accuracy of the analysis. Next, machine learning algorithms are used to analyze the data and identify the resident's behavioral patterns and daily rhythms. For example, it can detect patterns such as a resident waking up at roughly the same time every day and eating meals at the same time. The analysis results can also be used to assess the resident's health risks and detect irregular behavior or abnormal health signs.

[1026] Emotion engine integration:

[1027] Furthermore, an emotion engine is used to recognize the resident's emotions from conversation data. The voice recognition device analyzes the resident's tone of voice and choice of words to determine the resident's emotional state. For example, if a resident says, "I'm very tired today," the emotion engine analyzes this statement and recognizes that the resident is feeling fatigued.

[1028] 3. Care plan proposal phase

[1029] server:

[1030] Based on the analysis results, an optimal care plan is generated for each resident. The care plan includes suggestions for daily living support and medical interventions. For example, if a resident is prone to a particular health problem (such as nocturia), measures to improve that problem will be included in the plan. Care suggestions appropriate to the resident's emotional state are also generated based on emotional data obtained from the emotion engine. For example, if a resident is feeling fatigued, suggestions may include ensuring a relaxing environment and taking time to rest.

[1031] Care staff (users):

[1032] The generated care plan is sent to the care staff's devices in real time, allowing them to provide appropriate care for the resident, for example, by suggesting specific meal plans or planning additional health checks based on the resident's emotional state.

[1033] Specific examples

[1034] Device:

[1035] Resident A wakes up at 7:00 every morning, and the voice recognition device detects utterances such as "It's about time to get up" and sends them to the server.

[1036] server:

[1037] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm. If the emotion engine detects fatigue from Resident A's tone of voice, that emotion data is also stored on the server for analysis.

[1038] Care staff (users):

[1039] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time and emotional state. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" and "providing a relaxing environment" are provided.

[1040] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities, reduce the burden on nursing staff, and provide care that takes into consideration the emotions of users.

[1041] The processing flow will be explained below.

[1042] Program processing steps

[1043] 1. Data Collection Phase

[1044] Step 1:

[1045] The device uses IoT sensors to detect the resident's behavior, specifically capturing the resident's movements in real time, such as getting up from the bed or moving around the room.

[1046] Step 2:

[1047] The device detects the resident's conversation using a voice recognition device. For example, if a resident says "I'm hungry," this voice data is collected.

[1048] Step 3:

[1049] The device transmits the acquired behavioral and conversation data to a server, where the data is categorized and stored.

[1050] Step 4:

[1051] The device analyzes the resident's voice data and requests the emotion engine to determine their emotional state, for example, by detecting their emotional state such as "happiness" or "sadness" from their tone of voice and vocabulary.

[1052] 2. Data analysis phase

[1053] Step 5:

[1054] The server stores the received data in a database, where it is stored together with past data in a centralized management system for each resident.

[1055] Step 6:

[1056] The server cleanses the collected data, removing incomplete data and noise, a process that improves the accuracy of the analysis.

[1057] Step 7:

[1058] The server uses machine learning algorithms to analyze the data and extract patterns and rhythms of the resident's daily behavior, such as identifying a regular pattern where the resident wakes up at 7 a.m. and eats breakfast at 8 a.m.

[1059] Step 8:

[1060] The server evaluates the health risks of residents based on the analysis results. For example, if a pattern of frequent toilet visits at night is detected, the server evaluates the risk.

[1061] Step 9:

[1062] The server also analyzes emotional data from the emotion engine. For example, if a resident is feeling fatigued, that information is added to the risk assessment.

[1063] 3. Care plan proposal phase

[1064] Step 10:

[1065] The server generates an optimal care plan based on the analysis results. For example, if a resident is at risk of nocturia, it generates a plan that includes adjusting fluid intake and suggesting a doctor's appointment.

[1066] Step 11:

[1067] The server generates a care plan based on the emotional data, taking into account the resident's emotional state. For example, if the resident is feeling fatigued, it will include suggestions for extending rest time and relaxation.

[1068] Step 12:

[1069] The server notifies care staff of the care plan and risk assessment results in real time, including specific care suggestions.

[1070] Step 13:

[1071] Care staff (users) receive notifications and review the care plan, for example, making changes to the resident's diet or planning additional health checks.

[1072] Step 14:

[1073] Care staff carry out actual care based on the care plan, such as providing guidance on fluid intake and creating an environment for relaxation.

[1074] Specific examples

[1075] Step 1:

[1076] The device uses an IoT sensor to detect that resident A gets out of bed at 7am.

[1077] Step 2:

[1078] The device detects the voice of resident A saying, "I'm a little tired today."

[1079] Step 3:

[1080] The terminal transmits the acquired behavioral data and conversation data to the server.

[1081] Step 4:

[1082] The device sends the detected voice data to the emotion engine, which determines resident A's emotional state as "fatigue."

[1083] Step 5:

[1084] The server stores the received data in a database.

[1085] Step 6:

[1086] The server cleanses the collected data.

[1087] Step 7:

[1088] The server analyzes the data and identifies the behavioral patterns of Resident A. For example, it detects that Resident A wakes up at 7:00 every morning and has breakfast at 8:00.

[1089] Step 8:

[1090] The server evaluates health risks and determines that resident A is at risk of nocturia.

[1091] Step 9:

[1092] The server adds "fatigue" data from the emotion engine to the evaluation.

[1093] Step 10:

[1094] The server generates an optimal care plan.

[1095] Step 11:

[1096] The server generates a care plan based on the emotional state, for example, including relaxation recommendations for Resident A who feels fatigued.

[1097] Step 12:

[1098] The server notifies the care staff of the generated care plan.

[1099] Step 13:

[1100] The care staff (user) receives a notification and checks the care plan.

[1101] Step 14:

[1102] Nursing staff carry out actual care based on the care plan.

[1103] Through these specific processing steps, the AI ​​nursing care support system can efficiently grasp the resident's rhythm and health status, provide an optimal care plan, and utilize an emotion engine to provide emotion-based care.

[1104] Example 2

[1105] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1106] As the aging population continues to grow, improving the quality of nursing care services and reducing the workload of nursing staff at nursing facilities have become important issues. There is also a need to provide appropriate care tailored to residents' emotions and health conditions. However, current systems lack sufficient analysis of residents' behavioral and conversational data, making it difficult to provide care plans that take emotions into account. Therefore, a system that can provide efficient and effective care is needed.

[1107] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting resident behavioral information and speech information, means for analyzing the collected information to understand the resident's rhythm and health condition, means for generating an optimal care plan based on the understood rhythm and health condition, means for notifying care staff of the generated care plan, means for analyzing the resident's emotions using a voice recognition device and an emotion analysis engine, and means for adjusting the care plan based on the results of the emotion analysis. This makes it possible to efficiently understand the resident's health condition based on data obtained from the resident's behavior and speech and to generate a care plan accordingly. Furthermore, by adjusting the care content based on the resident's emotions, it is possible to improve the quality of care services and reduce the workload of care staff.

[1108] "Residents" refers to elderly people who live in nursing homes or senior housing, and people who require nursing care support.

[1109] "Behavioral information" refers to data on specific actions taken by residents in their daily lives (for example, waking up, eating, moving around, etc.).

[1110] "Speech information" refers to the words and conversations that residents make in their daily lives.

[1111] "Means of collection" refers to devices and systems for collecting behavioral information and speech information using sensors, voice recognition devices, etc.

[1112] "Means of analysis" refers to methods and systems for analyzing collected data to understand residents' behavioral patterns, health status, emotions, etc.

[1113] "Rhythm" refers to the temporal patterns and habits of residents' daily lives.

[1114] "Health status" refers to the state of a resident's physical and mental health.

[1115] A "care plan" refers to a plan that specifically outlines the optimal content and methods of care to be provided to residents based on the results of the analysis.

[1116] "Care workers" refers to employees and staff who provide care services to residents.

[1117] "Means for notifying" refers to a method or system for notifying care staff of the generated care plan.

[1118] A "voice recognition device" refers to a device or system that detects what residents say and converts it into text data.

[1119] "Emotion analysis engine" refers to an algorithm or system that analyzes data obtained from a voice recognition device to estimate the emotional state of a resident.

[1120] "Results of emotion analysis" refers to the emotional state of the resident analyzed by the emotion analysis engine.

[1121] "Adjustment means" refers to a method or system for appropriately changing the content or method of the care plan based on the analyzed emotional data.

[1122] The AI ​​nursing care support system of this invention collects and analyzes information on residents' behavior and speech to understand their rhythm, health condition, and emotions. Furthermore, it generates an optimal care plan based on this information and notifies caregivers, thereby providing appropriate care for residents and reducing the caregiver's workload.

[1123] Data collection

[1124] Device:

[1125] The devices are equipped with IoT sensors and voice recognition devices to detect the resident's actions and comments. This allows them to accurately capture the actions and comments that residents make in their daily lives. For example, if a resident wakes up at 7:00 a.m. and says, "It's about time to get up," the voice recognition device will detect this. Motion sensors are also used to collect data on the resident's movements.

[1126] Data transmission and storage

[1127] Device:

[1128] The device sends the collected behavioral information and speech information to a server using a secure communication protocol (e.g., HTTPS).

[1129] server:

[1130] The server receives the data sent from the devices and stores it in a database for each resident, which records behavioral and conversation data along with timestamps.

[1131] Data Cleansing and Analysis

[1132] server:

[1133] The server cleanses the received data, removing incomplete data and noise. Specifically, it filters out extremely short audio data and erroneous sensor values. Once cleansed, the data is analyzed using machine learning algorithms. For example, it can detect a behavioral pattern, such as "Resident A wakes up at 7:00 every morning."

[1134] Emotional Data Analysis

[1135] server:

[1136] The server uses an emotion analysis engine to analyze the resident's emotions from the collected speech information. Data obtained from the voice recognition device is input into the emotion analysis engine to determine the resident's emotional state. For example, a statement such as "I'm very tired today" is analyzed and recognized as indicating that the resident is feeling fatigued.

[1137] Generate a care plan

[1138] server:

[1139] The server generates an optimal care plan for each resident based on the analyzed behavioral patterns and emotional data. The care plan includes support for daily life and medical intervention suggestions. For example, if a resident wakes up at 7 a.m. every morning and feels tired, the server will create a care plan that provides a relaxing environment after waking up.

[1140] Care plan notification

[1141] server:

[1142] The server then sends the generated care plan to the caregiver's device, where the caregiver can check the plan in real time on a tablet or smartphone.

[1143] Care delivery

[1144] Caregiver (user):

[1145] Care staff provide care to residents based on the care plan they receive, for example, praising residents for getting up at 7 a.m. every morning, helping them prepare breakfast, and providing a relaxing environment.

[1146] Prompt Sentence Examples

[1147] "Resident A wakes up at 7:00 every morning, and the voice recognition device detects that he says, 'It's about time to get up.' Please also analyze Resident A's emotional data and generate the optimal care plan."

[1148] In this way, the present invention improves the quality of nursing care services in nursing facilities and reduces the workload of nursing staff. By providing care based on the feelings of residents, more detailed care becomes possible.

[1149] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1150] Step 1: Sensor configuration and data collection

[1151] Terminal: The terminal collects information on the residents' behavior and speech. Specifically, it uses IoT sensors and voice recognition devices. The sensors detect the residents' movements and actions, and the voice recognition device records what the residents say. The input is raw data obtained from the sensors and voice device. The output is recorded as the residents' behavior data and speech data. For example, if a resident wakes up at 7am, the device detects when they say, "It's about time to get up."

[1152] Step 2: Data transmission and storage

[1153] Terminal: The terminal sends the collected behavioral and speech information to the server. The input is the data collected in step 1. The output is a message confirming the completion of the transmission. The data is sent to the server via a secure communication protocol (such as HTTPS).

[1154] Server: The server receives the data sent from the device and stores it in a database. The input is the data sent from the device. The output is the behavioral data and speech data stored in the database. This records the data with a timestamp and makes it available for subsequent analysis.

[1155] Step 3: Data cleansing and preprocessing

[1156] Server: The server cleanses the received data and removes incomplete data and noise. The input is the data stored in step 2. The output is the cleansed data. Specifically, it filters out extremely short audio data and erroneous sensor values.

[1157] Step 4: Data analysis

[1158] Server: The server uses a machine learning algorithm to analyze the cleansed data and understand the resident's behavioral patterns and health status. The input is the clean data obtained in step 3. The output is the analysis results of the resident's behavioral patterns and health status. For example, a pattern such as "Resident A wakes up at 7 o'clock every morning" is detected.

[1159] Step 5: Analyze the sentiment data

[1160] Server: The server uses an emotion analysis engine to analyze the resident's emotions from the utterance data. The input is the utterance data obtained in step 3. The output is the analysis result of the resident's emotional state. For example, the emotion analysis engine analyzes the utterance "I'm very tired today" and outputs the result as "Feeling tired."

[1161] Step 6: Generate a care plan

[1162] Server: The server generates an optimal care plan for each resident based on the analyzed behavioral patterns and emotional data. The input is the analysis results from steps 4 and 5. The output is a care plan for each resident. For example, if a resident wakes up at 7 a.m. every morning and feels tired, a care plan is generated that provides a relaxing environment after waking up.

[1163] Step 7: Inform the plan of care

[1164] Server: The server notifies the care staff device of the care plan it has generated. The input is the care plan generated in step 6. The output is a push notification to the care staff device. This allows the care staff to check the care plan in real time on devices such as tablets and smartphones.

[1165] Step 8: Implementing care

[1166] Care staff (user): Care staff provide care to residents based on the notified care plan. The input is the care plan notified from the server. The output is a record of the care provided. For example, they may praise a resident for waking up at 7am every morning, help them prepare breakfast, and provide a relaxing environment.

[1167] (Application example 2)

[1168] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1169] Current care support systems are effective in collecting residents' behavioral and conversational data, understanding their rhythms and health status based on that data, and generating optimal care plans. However, they have limitations in providing comprehensive health support based on residents' health status and eating patterns. They lack specific suggestions for residents to eat an appropriate diet and the integration of food delivery services to implement those suggestions. Therefore, an effective means of ensuring residents receive optimal nutrition and maintain their health is needed.

[1170] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting resident behavioral data and conversation data, means for analyzing the collected data to understand the resident's rhythm and health condition, means for generating an optimal care plan based on the understood rhythm and health condition, means for notifying care staff of the generated care plan, means for proposing an optimal meal menu based on the user's eating patterns and health condition, and means for notifying a delivery service of the proposed meal menu. This enables comprehensive health support for residents, and by linking appropriate nutritional intake with delivery services, it becomes possible to improve the quality of life of residents and reduce the workload of care staff.

[1171] "Resident" means an individual who resides in a particular facility or residence.

[1172] "Behavioral data" refers to data that indicates the resident's movements and activities in daily life.

[1173] "Conversation data" refers to data collected as text information of the sounds and words uttered by residents.

[1174] "Rhythm" refers to the regular patterns of behavior and lifestyle of residents.

[1175] "Health status" refers to information related to the resident's physical condition and health.

[1176] A "care plan" is a written plan that lists specific care methods and support for residents.

[1177] "Care staff" refers to people whose occupations support the health and lives of residents.

[1178] "Dietary patterns" refer to the habits and tendencies of residents regarding what meals they eat and at what times.

[1179] "Health Status" refers to the state of a resident's physical and mental health.

[1180] A "meal menu" refers to the specific meal contents and menu to be provided to residents.

[1181] "Delivery service" refers to a service for delivering meal menus to residents.

[1182] This invention is a system that collects and analyzes behavioral and conversational data of residents to understand their rhythms and health status. It also generates an optimal care plan based on this data and notifies care staff, providing appropriate care and reducing their workload. It also includes a function to suggest optimal meal menus based on eating patterns and health status and notify the delivery service.

[1183] The server first collects behavioral and conversational data from residents. Behavioral data includes residents' movements and daily activities, while conversational data includes what residents say. This data is collected using smartphones and IoT sensors and sent to the server. A voice recognition API, commonly known as a voice recognition engine, is used to convert the voice data into text data.

[1184] The server cleanses the collected data, removing incomplete data and noise, and then analyzes it using machine learning algorithms to understand the resident's rhythm and health status. This analysis is performed using a generic machine learning platform. For example, if the server detects that a resident has a habit of eating at a specific time, it analyzes their eating patterns. It also evaluates the resident's emotional state using an emotion analysis engine.

[1185] The server then generates an optimal care plan based on the analysis, including assisted living and medical recommendations. It also suggests optimal meal plans based on the resident's eating patterns and emotional state, and notifies a food delivery service. A food delivery API is used to arrange for the optimal menu to be delivered to the resident.

[1186] This system is used in the following specific example.

[1187] When a resident says "I'm hungry," the behavioral data collection module detects this and converts it into text using a speech recognition engine. The data cleansing module processes this text data, and the behavioral pattern analysis module analyzes eating patterns. The emotion analysis module evaluates the resident's emotional state, and the menu generation module generates an optimal meal menu. The delivery notification module notifies the resident of delivery suggestions using a food delivery API.

[1188] Example prompt sentence:

[1189] When a user says "I'm hungry," the behavioral data collection module detects this and converts it into text using a speech recognition engine. The data cleansing module processes this text data, and the behavioral pattern analysis module analyzes eating patterns. The emotion analysis module evaluates the user's emotional state, and the menu generation module generates an optimal meal menu. The delivery notification module notifies the user of delivery suggestions using a food delivery API.

[1190] This system will enable comprehensive health support for residents, improving their quality of life through proper nutritional intake and linking delivery services, and reducing the workload of care staff.

[1191] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1192] Step 1:

[1193] The terminal uses IoT sensors and voice recognition devices to collect resident behavioral and conversational data. For example, if a resident says "I'm hungry," the voice is detected and converted into text data by a voice recognition engine. The input for this step is the resident's behavior and speech, and the output is text data and behavioral data.

[1194] Step 2:

[1195] The behavioral and conversational data sent from the device is sent to a server and securely stored. This involves building and centrally managing a database for each specific resident. The input for this step is text data and behavioral data, and the output is the results stored in the database.

[1196] Step 3:

[1197] The server cleanses the collected data and removes incomplete data and noise. This is a data preprocessing stage and is an important step for accurate analysis. Specifically, it detects outliers and missing data and removes or imputes them. The input for this step is the stored behavioral data and text data, and the output is the cleansed data.

[1198] Step 4:

[1199] Based on the cleansed data, the server analyzes the data using machine learning algorithms. Common machine learning platforms such as TensorFlow and PyTorch are used here. The analysis identifies resident behavior patterns and daily rhythms and assesses health risks. The input for this step is the cleansed data, and the output is behavior patterns and health risk assessment results.

[1200] Step 5:

[1201] Using an emotion analysis engine, the server recognizes the resident's emotions from the conversation data. Specifically, it uses data from a voice recognition device and uses an emotion analysis API to determine the resident's emotional state from the tone of voice and vocabulary. The input for this step is the conversation data, and the output is the emotion analysis results.

[1202] Step 6:

[1203] Based on the analysis results, the server generates an optimal care plan for each resident. This care plan includes specific suggestions for health management and meal plans. The analysis results are integrated with the emotional state to suggest the optimal meal menu for the resident. This results in a care plan that corresponds to the resident's health condition. The inputs for this step are behavioral patterns, health risk assessment, and emotional analysis results, and the output is a care plan and meal menu suggestions.

[1204] Step 7:

[1205] The generated care plan and meal menu are notified to the nursing staff and delivery service. Using the food delivery API, arrangements are made to quickly deliver the optimal meal menu for the resident. Nursing staff are notified of the resident's status in real time, helping them respond quickly. The input to this step is the care plan and meal menu suggestions, and the output is the notification results.

[1206] Through these steps, this system will provide comprehensive health support for residents, reduce the workload of care staff, and provide residents with appropriate nutritional intake.

[1207] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1208] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1209] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1210] [Fourth embodiment]

[1211] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1212] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1213] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1214] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1215] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1216] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1217] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1218] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1219] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1220] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1221] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1222] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1223] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1224] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. Furthermore, it generates an optimal care plan based on this information and notifies the care staff, thereby providing appropriate care to residents and reducing the care staff's workload.

[1225] 1. Data Collection Phase

[1226] Device:

[1227] The terminals are equipped with IoT sensors and voice recognition devices to detect the residents' behavior and conversations. This allows them to accurately capture the actions and words that residents perform in their daily lives. For example, the terminals can detect and collect data on when residents wake up in the morning, when they eat meals, and even when they say they're hungry.

[1228] server:

[1229] The server receives and securely stores data sent from the device, which includes building a database for each resident and centrally managing behavioral and conversation data.

[1230] 2. Data analysis phase

[1231] server:

[1232] The server cleanses the collected data, removing incomplete data and noise, improving the accuracy of the analysis. Next, machine learning algorithms are used to analyze the data and identify the resident's behavioral patterns and daily rhythms. For example, it can detect patterns such as a resident waking up at roughly the same time every day and eating meals at the same time. The analysis results can also be used to assess the resident's health risks and detect irregular behavior or abnormal health signs.

[1233] 3. Care plan proposal phase

[1234] server:

[1235] Based on the analysis results, an optimal care plan is generated for each resident. The care plan includes suggestions for daily living support and medical intervention. For example, if a resident's frequent trips to the toilet at night are detected, the server will generate a care plan suggesting a review of fluid intake and a doctor's consultation.

[1236] Care staff (users):

[1237] The generated care plan is sent to the care staff's devices in real time, allowing them to provide appropriate care for the resident, such as suggesting specific meal plans and conducting regular health checks.

[1238] Specific examples

[1239] Device:

[1240] Resident A wakes up at 7:00 every morning, and the voice recognition device detects utterances such as "It's about time to get up" and sends them to the server.

[1241] server:

[1242] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm.

[1243] Care staff (users):

[1244] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" are notified.

[1245] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities and reduce the burden on nursing staff.

[1246] The processing flow will be explained below.

[1247] Program processing steps

[1248] 1. Data Collection Phase

[1249] Step 1:

[1250] The device detects the resident's actions and conversations. For example, it uses IoT sensors and voice recognition devices to capture the time the resident wakes up in the morning or when they say they're hungry.

[1251] Step 2:

[1252] The data detected by the device is sent in real time to a server, where it is classified as behavioral data or conversation data.

[1253] Step 3:

[1254] The server stores the received data in a database. The data is organized for each resident and stored together with past data.

[1255] 2. Data analysis phase

[1256] Step 4:

[1257] The server cleanses the collected data, removing incomplete data and noise to improve the accuracy of the analysis.

[1258] Step 5:

[1259] The server analyzes the data using machine learning algorithms to extract the resident's daily behavioral patterns and identify regularities, such as waking up at 7am and having breakfast at 8am every day.

[1260] Step 6:

[1261] The server uses the analysis results to understand the resident's rhythm and health status, for example, by analyzing sleep and eating patterns, and assessing the health risks if any abnormalities are found.

[1262] 3. Care plan proposal phase

[1263] Step 7:

[1264] The server generates an optimal care plan based on the analysis results. For example, if a resident is prone to a particular health problem (such as frequent urination at night), the plan will include measures to improve that problem.

[1265] Step 8:

[1266] The server generates a care plan and notifies the care staff, including specific improvement measures and daily care suggestions.

[1267] Step 9:

[1268] Care staff are notified and review the care plan, including making dietary changes or additional health checks to accommodate each resident's health condition.

[1269] Step 10:

[1270] Nursing staff carry out actual care based on the care plan. For example, for a resident with nocturia, they may provide guidance on fluid intake and arrange appointments with medical institutions.

[1271] Through these steps, the AI ​​nursing care support system efficiently grasps the rhythm and health status of residents and provides optimal care plans, thereby reducing the burden on nursing staff and improving the quality of nursing care services.

[1272] Example 1

[1273] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1274] Conventional nursing care support systems face challenges in efficiently collecting and analyzing resident behavior and conversation data and notifying nursing staff of optimal care plans in real time. Furthermore, collected data often contains noise and incomplete data, necessitating data cleansing. However, this manual process requires time and effort, so there is a demand for more accurate analysis and faster response.

[1275] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1276] In this invention, the server includes means for collecting resident behavioral data and conversation data, means for transmitting the collected data to the server, means for storing the transmitted data in a database, means for cleansing the stored data, means for analyzing the cleansed data to grasp the resident's rhythm and health condition, means for generating an optimal care plan based on the grasped rhythm and health condition, and means for notifying the care staff of the generated care plan. This makes it possible to efficiently and accurately analyze resident data and provide an optimal care plan in real time.

[1277] "Resident behavior data" is information related to the actions and behaviors that residents perform in their daily lives.

[1278] "Conversation data" is information about the voices uttered by the resident and their contents.

[1279] "Means for collection" refers to devices and technologies for acquiring resident behavioral data and conversation data.

[1280] "Transmitting means" refers to the communication device or protocol used to transfer the collected data to the server.

[1281] "Means of storage" refers to a database or storage system for safely storing and managing the transmitted data.

[1282] A "cleansing method" is a process or algorithm for detecting, removing, or correcting incomplete data or noise.

[1283] "Means of analysis" refers to technology that analyzes cleansed data using machine learning algorithms, etc., to understand residents' behavioral patterns and health conditions.

[1284] "Means of understanding" refers to methods for recognizing and understanding the rhythms and health status of residents based on information obtained from analyzed data.

[1285] The "means of generation" refers to systems and algorithms that create optimal care plans for each resident based on their rhythms and health status.

[1286] The "notification means" refers to a communication means or interface for informing the care staff of the generated care plan in real time.

[1287] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. This system is realized using the following hardware and software.

[1288] Data Collection Phase

[1289] Device:

[1290] The terminal uses IoT sensors and voice recognition devices to detect the resident's actions and conversations. The IoT sensors detect when the resident gets up from the bed in real time, and the voice recognition device accurately recognizes when the resident says, "It's about time to get up." This data is sent from the terminal to the server.

[1291] Data storage phase

[1292] server:

[1293] The server receives the data sent from the device and stores it in a database for each resident. This database stores data in a format that includes resident ID, timestamp, behavioral data, conversation data, etc.

[1294] Data Cleansing Phase

[1295] server:

[1296] The server reads new data from the database and starts the data cleansing process, specifically detecting incomplete data and noise, and removing and correcting them. This process improves the accuracy of the analysis.

[1297] Data analysis phase

[1298] server:

[1299] The server uses machine learning algorithms to analyze the cleansed data, identifying resident behavior patterns and daily rhythms. For example, it can identify resident A's pattern of waking up at 7 a.m. every morning and evaluate this rhythm as healthy.

[1300] Care plan generation phase

[1301] server:

[1302] Based on the analysis results, the server automatically generates an optimal care plan for each resident. This care plan includes support for daily activities and medical intervention suggestions. For example, for a resident who frequently uses the toilet at night, it might include suggestions to review fluid intake and instructions to seek medical advice.

[1303] Care plan notification phase

[1304] server:

[1305] The generated care plan is notified to the care staff's device in real time.

[1306] User (care staff):

[1307] Care staff can then review the care plan sent to them via the device and provide specific care to the resident based on the plan's contents, such as praising the resident for getting up at 7 a.m. every morning and helping them prepare breakfast.

[1308] Specific examples

[1309] Device:

[1310] Resident A wakes up at 7am every morning, and the device detects when he or she says, "It's about time to get up," and sends this information to the server.

[1311] server:

[1312] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm.

[1313] Care staff (users):

[1314] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" are notified.

[1315] Example prompts for generative AI models

[1316] Example prompt sentence:

[1317] "I am designing a system that can analyze a resident's behavioral pattern, such as waking up at 7am every day. How can I use this system to assess the resident's health risks and generate and communicate an appropriate care plan?"

[1318] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities and reduce the burden on nursing staff.

[1319] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1320] Step 1:

[1321] The device collects data

[1322] Input: Resident actions and conversations

[1323] Output: Behavioral and conversational data

[1324] The device uses IoT sensors and voice recognition devices to monitor the resident's behavior and conversations in real time. Specifically, the IoT sensors detect when the resident gets up from the bed, and the voice recognition device records what the resident says. For example, it detects the voice saying, "It's about time to get up," and collects this as data.

[1325] Step 2:

[1326] The device sends the data to the server

[1327] Input: Behavioral and conversational data

[1328] Output: Data sent to the server

[1329] The collected behavioral and conversational data is sent to the server in batch format. Specifically, the data is packaged within the device and transferred to the server using a secure communication protocol. For example, data with timestamps is compiled into batches and sent to the server at regular intervals.

[1330] Step 3:

[1331] The server saves the data to a database

[1332] Input: Data sent

[1333] Output: Data stored in the database

[1334] The server receives the data sent from the device and stores it in a database for each resident. Specifically, it uses a database structure to store the resident ID, timestamp, behavioral data, conversation data, etc. For example, an entry such as "User ID: A, Time: 7:00 AM, Action: Wake up, Conversation: 'It's about time to get up'" is created.

[1335] Step 4:

[1336] The server cleanses the data

[1337] Input: Data stored in the database

[1338] Output: Cleansed data

[1339] The server reads new data from the database and performs a data cleansing process to detect, remove, and correct incomplete data and noise. Specifically, it complements missing values ​​and corrects outliers. For example, it filters noise from voice data to produce clear data.

[1340] Step 5:

[1341] The server analyzes the data

[1342] Input: Cleansed data

[1343] Output: Resident behavior patterns and health status

[1344] The server uses machine learning algorithms to analyze the cleansed data. This data analysis identifies the resident's behavioral patterns and daily rhythms. For example, it detects that resident A wakes up at 7:00 every morning. This information is then used to evaluate the resident's health status.

[1345] Step 6:

[1346] The server generates a care plan

[1347] Input: Behavioral patterns and health status

[1348] Output: Optimal care plan

[1349] Based on the analysis results, the server generates an optimal care plan for each resident. The generated care plan includes support for daily activities and medical intervention suggestions. For example, for a resident who frequently uses the toilet at night, it might include suggestions to review fluid intake and seek medical advice.

[1350] Step 7:

[1351] The server notifies the care plan

[1352] Input: Optimal Care Plan

[1353] Output: Care plan sent to the care staff's device

[1354] The generated care plan is sent to the care staff's device in real time. Specifically, a pop-up message or an alert is sent to the care staff's device via the notification system, allowing the necessary care to be carried out promptly.

[1355] Step 8:

[1356] The user (caregiver) performs care

[1357] Input: Notified Care Plan

[1358] Output: Specific care for residents

[1359] Care staff can then review the care plan sent to them via the device and provide specific care based on the plan, such as praising the resident for getting up at 7am every morning or helping them prepare breakfast.

[1360] (Application example 1)

[1361] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1362] Conventional factory robot maintenance and operation management is often performed at regular intervals, making it difficult to detect actual usage status and abnormalities in real time. This has led to maintenance delays and excessive maintenance work, making efficient operation difficult. It has also been impossible to properly collect instructions and comments from operators and generate optimal maintenance plans based on them.

[1363] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1364] In this invention, the server includes means for collecting behavioral data and conversation data of residents and automatic equipment, means for analyzing the collected data to grasp the work rhythms and operating status of the residents and automatic equipment, means for generating an optimal maintenance plan based on the grasped work rhythms and operating status, and means for notifying a manager or operator of the generated maintenance plan, thereby enabling the generation and notification of an optimal maintenance plan based on the operating status of factory robots and instructions from operators.

[1365] A "resident" is a person who lives within a particular facility or environment and about whom data is collected.

[1366] "Automatic equipment" refers to a device that automatically performs a task based on a control signal, and examples thereof include factory robots.

[1367] "Behavioral data" refers to information about the behavior and operation history of a target person or automated equipment, and is collected through sensors and devices.

[1368] "Conversation data" refers to information collected through a voice recognition device that includes statements and instructions from the target person or operator.

[1369] A "server" is a computer system for storing and analyzing collected data and communicating generated plans.

[1370] "Work rhythm" refers to the daily behavioral patterns and schedules of automated equipment and residents.

[1371] "Operational status" refers to information about the actual operating state of automated equipment and its performance.

[1372] "Maintenance plan" refers to the schedule and content of maintenance work required to keep equipment operating properly.

[1373] "Manager or operator" means a person responsible for the operation and management of automated equipment or systems.

[1374] The system according to the present invention collects and analyzes behavioral and conversational data of automated equipment and residents to understand work rhythms and operating conditions, and then generates and notifies optimal maintenance plans. Specific embodiments of the system are described below.

[1375] 1. Data Collection Phase

[1376] Device:

[1377] The terminal collects real-time operational data and operator conversation data from IoT sensors and voice recognition devices attached to automated equipment. For example, it can collect information on whether robots in a factory are operating normally or what operators say when giving maintenance instructions.

[1378] 2. Data analysis phase

[1379] server:

[1380] The server receives the data sent from the device and first cleans it. This removes noise and missing data to improve the accuracy of the analysis. Next, it uses machine learning algorithms to analyze the data and detect patterns and anomalies in the operation of automated equipment. For example, it can detect the phenomenon of a robot slowing down at certain times each day.

[1381] 3. Maintenance plan generation phase

[1382] server:

[1383] The server then generates an optimal maintenance plan based on the analysis results. This maintenance plan includes periodic part replacement for automated equipment and necessary actions when specific abnormalities occur. For example, if a robot's performance deteriorates, the server can add the replacement of a specific part to the schedule.

[1384] 4. Notification Phase

[1385] Device:

[1386] The generated maintenance plan is sent to the terminal of the manager or operator in real time, allowing the manager or operator to carry out the necessary maintenance work at the appropriate time. For example, a specific instruction such as "replace a specific part every weekend" may be sent to the operator's terminal.

[1387] Hardware and software used

[1388] IoT sensors: Collect operational data from the robot.

[1389] Voice recognition device: collects instructions and comments from the operator.

[1390] Server: Cleanses and analyzes the data.

[1391] Machine learning algorithms: Use Python's TensorFlow or PyTorch.

[1392] Database: Use SQLite or MySQL to manage collected data.

[1393] Specific examples

[1394] Terminal: When factory robot A needs specific maintenance every weekend, a voice recognition device detects an operator's instruction such as "Robot A is exhibiting a malfunction" and sends it to the server.

[1395] Server: The server stores the received data in a database and uses machine learning algorithms to analyze abnormal patterns of Robot A. For example, it can detect irregular behavior patterns and recommend maintenance before an abnormality occurs.

[1396] Manager or operator: The optimal maintenance plan is sent to the operator's terminal, with specific suggestions such as "replace specific parts every weekend."

[1397] Prompt Sentence Examples

[1398] "Develop a program that uses IoT sensors and voice recognition devices attached to factory robots to collect operational data and operator instructions regarding maintenance for Robot A, and generate an optimal maintenance plan. This program will include a function to detect anomalies and notify maintenance plans in real time. Use Python's TensorFlow."

[1399] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1400] Step 1:

[1401] The terminal collects operation data and operator conversation data from IoT sensors and voice recognition devices attached to the automated equipment. Specifically, it acquires the robot's operating status and voice data of instructions and comments given by the operator in real time and converts this data into digital format. Input: Operation data from the automated equipment's sensors, voice data from the voice recognition device. Output: Digital operation data and conversation data.

[1402] Step 2:

[1403] The terminal sends the collected digital data to the server. Specifically, it uses a data communication protocol to properly transfer the data and store it in a database on the server. Input: Digital operating data and conversation data. Output: Operating data and conversation data stored on the server.

[1404] Step 3:

[1405] The server cleanses the received data. Specifically, it removes incomplete data and noise and prepares the data to improve the accuracy of analysis. For example, it filters background noise from voice data and fills in missing values ​​in sensor data. Input: Operational data and conversation data stored on the server. Output: Cleansed data.

[1406] Step 4:

[1407] The server analyzes the cleansed data. Specifically, it uses machine learning algorithms to analyze the data and understand work rhythms and operating conditions. For example, it uses TensorFlow or PyTorch to learn the operation patterns of equipment and detect anomalies. Input: Cleansed data. Output: Analysis results (work rhythms, operating conditions, anomaly detection results).

[1408] Step 5:

[1409] The server generates an optimal maintenance plan based on the analysis results. Specifically, it automatically creates a schedule for maintenance timing and necessary part replacement based on the output of the machine learning model. Input: Analysis results (work rhythm, operating status, anomaly detection results). Output: Optimal maintenance plan.

[1410] Step 6:

[1411] The server notifies the administrator or operator of the generated maintenance plan. Specifically, it sends a real-time notification to the operator's terminal, presenting the necessary maintenance work. For example, a notification containing specific instructions such as "replace a specific part every weekend" is sent. Input: Optimized maintenance plan. Output: Maintenance instructions displayed on the operator's terminal.

[1412] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1413] The AI ​​nursing care support system of the present invention collects and analyzes behavioral and conversational data of residents to understand their rhythms and health conditions. Furthermore, it generates an optimal care plan based on this information and notifies the care staff, thereby providing appropriate care for the residents and reducing the care staff's workload. Furthermore, by combining it with an emotion engine, it can also recognize the user's emotions and provide emotionally appropriate care.

[1414] 1. Data Collection Phase

[1415] Device:

[1416] The terminals are equipped with IoT sensors and voice recognition devices to detect the residents' behavior and conversations. This allows them to accurately capture the actions and words that residents perform in their daily lives. For example, the terminals can detect and collect data on when residents wake up in the morning, when they eat meals, and even when they say they're hungry.

[1417] server:

[1418] The server receives and securely stores data sent from the device, which includes building a database for each resident and centrally managing behavioral and conversation data.

[1419] 2. Data analysis phase

[1420] server:

[1421] The server cleanses the collected data, removing incomplete data and noise, improving the accuracy of the analysis. Next, machine learning algorithms are used to analyze the data and identify the resident's behavioral patterns and daily rhythms. For example, it can detect patterns such as a resident waking up at roughly the same time every day and eating meals at the same time. The analysis results can also be used to assess the resident's health risks and detect irregular behavior or abnormal health signs.

[1422] Emotion engine integration:

[1423] Furthermore, an emotion engine is used to recognize the resident's emotions from conversation data. The voice recognition device analyzes the resident's tone of voice and choice of words to determine the resident's emotional state. For example, if a resident says, "I'm very tired today," the emotion engine analyzes this statement and recognizes that the resident is feeling fatigued.

[1424] 3. Care plan proposal phase

[1425] server:

[1426] Based on the analysis results, an optimal care plan is generated for each resident. The care plan includes suggestions for daily living support and medical interventions. For example, if a resident is prone to a particular health problem (such as nocturia), measures to improve that problem will be included in the plan. Care suggestions appropriate to the resident's emotional state are also generated based on emotional data obtained from the emotion engine. For example, if a resident is feeling fatigued, suggestions may include ensuring a relaxing environment and taking time to rest.

[1427] Care staff (users):

[1428] The generated care plan is sent to the care staff's devices in real time, allowing them to provide appropriate care for the resident, for example, by suggesting specific meal plans or planning additional health checks based on the resident's emotional state.

[1429] Specific examples

[1430] Device:

[1431] Resident A wakes up at 7:00 every morning, and the voice recognition device detects utterances such as "It's about time to get up" and sends them to the server.

[1432] server:

[1433] The server stores the received data in a database and uses a machine learning algorithm to analyze Resident A's behavioral patterns. For example, it detects a pattern of waking up at 7:00 and determines that this is a healthy behavioral rhythm. If the emotion engine detects fatigue from Resident A's tone of voice, that emotion data is also stored on the server for analysis.

[1434] Care staff (users):

[1435] The system notifies the care staff of an appropriate care plan based on Resident A's wake-up time and emotional state. For example, specific care suggestions such as "praising Resident A for getting up at 7 a.m. every day and helping him prepare breakfast" and "providing a relaxing environment" are provided.

[1436] The above is an embodiment of the present invention. The present invention can improve the quality of nursing care services in nursing facilities, reduce the burden on nursing staff, and provide care that takes into consideration the emotions of users.

[1437] The processing flow will be explained below.

[1438] Program processing steps

[1439] 1. Data Collection Phase

[1440] Step 1:

[1441] The device uses IoT sensors to detect the resident's behavior, specifically capturing the resident's movements in real time, such as getting up from the bed or moving around the room.

[1442] Step 2:

[1443] The device detects the resident's conversation using a voice recognition device. For example, if a resident says "I'm hungry," this voice data is collected.

[1444] Step 3:

[1445] The device transmits the acquired behavioral and conversation data to a server, where the data is categorized and stored.

[1446] Step 4:

[1447] The device analyzes the resident's voice data and requests the emotion engine to determine their emotional state, for example, by detecting their emotional state such as "happiness" or "sadness" from their tone of voice and vocabulary.

[1448] 2. Data analysis phase

[1449] Step 5:

[1450] The server stores the received data in a database, where it is stored together with past data in a centralized management system for each resident.

[1451] Step 6:

[1452] The server cleanses the collected data, removing incomplete data and noise, a process that improves the accuracy of the analysis.

[1453] Step 7:

[1454] The server uses machine learning algorithms to analyze the data and extract patterns and rhythms of the resident's daily behavior, such as identifying a regular pattern where the resident wakes up at 7 a.m. and eats breakfast at 8 a.m.

[1455] Step 8:

[1456] The server evaluates the health risks of residents based on the analysis results. For example, if a pattern of frequent toilet visits at night is detected, the server evaluates the risk.

[1457] Step 9:

[1458] The server also analyzes emotional data from the emotion engine. For example, if a resident is feeling fatigued, that information is added to the risk assessment.

[1459] 3. Care plan proposal phase

[1460] Step 10:

[1461] The server generates an optimal care plan based on the analysis results. For example, if a resident is at risk of nocturia, it generates a plan that includes adjusting fluid intake and suggesting a doctor's appointment.

[1462] Step 11:

[1463] The server generates a care plan based on the emotional data, taking into account the resident's emotional state. For example, if the resident is feeling fatigued, it will include suggestions for extending rest time and relaxation.

[1464] Step 12:

[1465] The server notifies care staff of the care plan and risk assessment results in real time, including specific care suggestions.

[1466] Step 13:

[1467] Care staff (users) receive notifications and review the care plan, for example, making changes to the resident's diet or planning additional health checks.

[1468] Step 14:

[1469] Care staff carry out actual care based on the care plan, such as providing guidance on fluid intake and creating an environment for relaxation.

[1470] Specific examples

[1471] Step 1:

[1472] The device uses an IoT sensor to detect that resident A gets out of bed at 7am.

[1473] Step 2:

[1474] The device detects the voice of resident A saying, "I'm a little tired today."

[1475] Step 3:

[1476] The terminal transmits the acquired behavioral data and conversation data to the server.

[1477] Step 4:

[1478] The device sends the detected voice data to the emotion engine, which determines resident A's emotional state as "fatigue."

[1479] Step 5:

[1480] The server stores the received data in a database.

[1481] Step 6:

[1482] The server cleanses the collected data.

[1483] Step 7:

[1484] The server analyzes the data and identifies the behavioral patterns of Resident A. For example, it detects that Resident A wakes up at 7:00 every morning and has breakfast at 8:00.

[1485] Step 8:

[1486] The server evaluates health risks and determines that resident A is at risk of nocturia.

[1487] Step 9:

[1488] The server adds "fatigue" data from the emotion engine to the evaluation.

[1489] Step 10:

[1490] The server generates an optimal care plan.

[1491] Step 11:

[1492] The server generates a care plan based on the emotional state, for example, including relaxation recommendations for Resident A who feels fatigued.

[1493] Step 12:

[1494] The server notifies the care staff of the generated care plan.

[1495] Step 13:

[1496] The care staff (user) receives a notification and checks the care plan.

[1497] Step 14:

[1498] Nursing staff carry out actual care based on the care plan.

[1499] Through these specific processing steps, the AI ​​nursing care support system can efficiently grasp the resident's rhythm and health status, provide an optimal care plan, and utilize an emotion engine to provide emotion-based care.

[1500] Example 2

[1501] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1502] As the aging population continues to grow, improving the quality of nursing care services and reducing the workload of nursing staff at nursing facilities have become important issues. There is also a need to provide appropriate care tailored to residents' emotions and health conditions. However, current systems lack sufficient analysis of residents' behavioral and conversational data, making it difficult to provide care plans that take emotions into account. Therefore, a system that can provide efficient and effective care is needed.

[1503] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting resident behavioral information and speech information, means for analyzing the collected information to understand the resident's rhythm and health condition, means for generating an optimal care plan based on the understood rhythm and health condition, means for notifying care staff of the generated care plan, means for analyzing the resident's emotions using a voice recognition device and an emotion analysis engine, and means for adjusting the care plan based on the results of the emotion analysis. This makes it possible to efficiently understand the resident's health condition based on data obtained from the resident's behavior and speech and to generate a care plan accordingly. Furthermore, by adjusting the care content based on the resident's emotions, it is possible to improve the quality of care services and reduce the workload of care staff.

[1504] "Residents" refers to elderly people who live in nursing homes or senior housing, and people who require nursing care support.

[1505] "Behavioral information" refers to data on specific actions taken by residents in their daily lives (for example, waking up, eating, moving around, etc.).

[1506] "Speech information" refers to the words and conversations that residents make in their daily lives.

[1507] "Means of collection" refers to devices and systems for collecting behavioral information and speech information using sensors, voice recognition devices, etc.

[1508] "Means of analysis" refers to methods and systems for analyzing collected data to understand residents' behavioral patterns, health status, emotions, etc.

[1509] "Rhythm" refers to the temporal patterns and habits of residents' daily lives.

[1510] "Health status" refers to the state of a resident's physical and mental health.

[1511] A "care plan" refers to a plan that specifically outlines the optimal content and methods of care to be provided to residents based on the results of the analysis.

[1512] "Care workers" refers to employees and staff who provide care services to residents.

[1513] "Means for notifying" refers to a method or system for notifying care staff of the generated care plan.

[1514] A "voice recognition device" refers to a device or system that detects what residents say and converts it into text data.

[1515] "Emotion analysis engine" refers to an algorithm or system that analyzes data obtained from a voice recognition device to estimate the emotional state of a resident.

[1516] "Results of emotion analysis" refers to the emotional state of the resident analyzed by the emotion analysis engine.

[1517] "Adjustment means" refers to a method or system for appropriately changing the content or method of the care plan based on the analyzed emotional data.

[1518] The AI ​​nursing care support system of this invention collects and analyzes information on residents' behavior and speech to understand their rhythm, health condition, and emotions. Furthermore, it generates an optimal care plan based on this information and notifies caregivers, thereby providing appropriate care for residents and reducing the caregiver's workload.

[1519] Data collection

[1520] Device:

[1521] The devices are equipped with IoT sensors and voice recognition devices to detect the resident's actions and comments. This allows them to accurately capture the actions and comments that residents make in their daily lives. For example, if a resident wakes up at 7:00 a.m. and says, "It's about time to get up," the voice recognition device will detect this. Motion sensors are also used to collect data on the resident's movements.

[1522] Data transmission and storage

[1523] Device:

[1524] The device sends the collected behavioral information and speech information to a server using a secure communication protocol (e.g., HTTPS).

[1525] server:

[1526] The server receives the data sent from the devices and stores it in a database for each resident, which records behavioral and conversation data along with timestamps.

[1527] Data Cleansing and Analysis

[1528] server:

[1529] The server cleanses the received data, removing incomplete data and noise. Specifically, it filters out extremely short audio data and erroneous sensor values. Once cleansed, the data is analyzed using machine learning algorithms. For example, it can detect a behavioral pattern, such as "Resident A wakes up at 7:00 every morning."

[1530] Emotional Data Analysis

[1531] server:

[1532] The server uses an emotion analysis engine to analyze the resident's emotions from the collected speech information. Data obtained from the voice recognition device is input into the emotion analysis engine to determine the resident's emotional state. For example, a statement such as "I'm very tired today" is analyzed and recognized as indicating that the resident is feeling fatigued.

[1533] Generate a care plan

[1534] server:

[1535] The server generates an optimal care plan for each resident based on the analyzed behavioral patterns and emotional data. The care plan includes support for daily life and medical intervention suggestions. For example, if a resident wakes up at 7 a.m. every morning and feels tired, the server will create a care plan that provides a relaxing environment after waking up.

[1536] Care plan notification

[1537] server:

[1538] The server then sends the generated care plan to the caregiver's device, where the caregiver can check the plan in real time on a tablet or smartphone.

[1539] Care delivery

[1540] Caregiver (user):

[1541] Care staff provide care to residents based on the care plan they receive, for example, praising residents for getting up at 7 a.m. every morning, helping them prepare breakfast, and providing a relaxing environment.

[1542] Prompt Sentence Examples

[1543] "Resident A wakes up at 7:00 every morning, and the voice recognition device detects that he says, 'It's about time to get up.' Please also analyze Resident A's emotional data and generate the optimal care plan."

[1544] In this way, the present invention improves the quality of nursing care services in nursing facilities and reduces the workload of nursing staff. By providing care based on the feelings of residents, more detailed care becomes possible.

[1545] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1546] Step 1: Sensor configuration and data collection

[1547] Terminal: The terminal collects information on the residents' behavior and speech. Specifically, it uses IoT sensors and voice recognition devices. The sensors detect the residents' movements and actions, and the voice recognition device records what the residents say. The input is raw data obtained from the sensors and voice device. The output is recorded as the residents' behavior data and speech data. For example, if a resident wakes up at 7am, the device detects when they say, "It's about time to get up."

[1548] Step 2: Data transmission and storage

[1549] Terminal: The terminal sends the collected behavioral and speech information to the server. The input is the data collected in step 1. The output is a message confirming the completion of the transmission. The data is sent to the server via a secure communication protocol (such as HTTPS).

[1550] Server: The server receives the data sent from the device and stores it in a database. The input is the data sent from the device. The output is the behavioral data and speech data stored in the database. This records the data with a timestamp and makes it available for subsequent analysis.

[1551] Step 3: Data cleansing and preprocessing

[1552] Server: The server cleanses the received data and removes incomplete data and noise. The input is the data stored in step 2. The output is the cleansed data. Specifically, it filters out extremely short audio data and erroneous sensor values.

[1553] Step 4: Data analysis

[1554] Server: The server uses a machine learning algorithm to analyze the cleansed data and understand the resident's behavioral patterns and health status. The input is the clean data obtained in step 3. The output is the analysis results of the resident's behavioral patterns and health status. For example, a pattern such as "Resident A wakes up at 7 o'clock every morning" is detected.

[1555] Step 5: Analyze the sentiment data

[1556] Server: The server uses an emotion analysis engine to analyze the resident's emotions from the utterance data. The input is the utterance data obtained in step 3. The output is the analysis result of the resident's emotional state. For example, the emotion analysis engine analyzes the utterance "I'm very tired today" and outputs the result as "Feeling tired."

[1557] Step 6: Generate a care plan

[1558] Server: The server generates an optimal care plan for each resident based on the analyzed behavioral patterns and emotional data. The input is the analysis results from steps 4 and 5. The output is a care plan for each resident. For example, if a resident wakes up at 7 a.m. every morning and feels tired, a care plan is generated that provides a relaxing environment after waking up.

[1559] Step 7: Inform the plan of care

[1560] Server: The server notifies the care staff device of the care plan it has generated. The input is the care plan generated in step 6. The output is a push notification to the care staff device. This allows the care staff to check the care plan in real time on devices such as tablets and smartphones.

[1561] Step 8: Implementing care

[1562] Care staff (user): Care staff provide care to residents based on the notified care plan. The input is the care plan notified from the server. The output is a record of the care provided. For example, they may praise a resident for waking up at 7am every morning, help them prepare breakfast, and provide a relaxing environment.

[1563] (Application example 2)

[1564] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1565] Current care support systems are effective in collecting residents' behavioral and conversational data, understanding their rhythms and health status based on that data, and generating optimal care plans. However, they have limitations in providing comprehensive health support based on residents' health status and eating patterns. They lack specific suggestions for residents to eat an appropriate diet and the integration of food delivery services to implement those suggestions. Therefore, an effective means of ensuring residents receive optimal nutrition and maintain their health is needed.

[1566] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting resident behavioral data and conversation data, means for analyzing the collected data to understand the resident's rhythm and health condition, means for generating an optimal care plan based on the understood rhythm and health condition, means for notifying care staff of the generated care plan, means for proposing an optimal meal menu based on the user's eating patterns and health condition, and means for notifying a delivery service of the proposed meal menu. This enables comprehensive health support for residents, and by linking appropriate nutritional intake with delivery services, it becomes possible to improve the quality of life of residents and reduce the workload of care staff.

[1567] "Resident" means an individual who resides in a particular facility or residence.

[1568] "Behavioral data" refers to data that indicates the resident's movements and activities in daily life.

[1569] "Conversation data" refers to data collected as text information of the sounds and words uttered by residents.

[1570] "Rhythm" refers to the regular patterns of behavior and lifestyle of residents.

[1571] "Health status" refers to information related to the resident's physical condition and health.

[1572] A "care plan" is a written plan that lists specific care methods and support for residents.

[1573] "Care staff" refers to people whose occupations support the health and lives of residents.

[1574] "Dietary patterns" refer to the habits and tendencies of residents regarding what meals they eat and at what times.

[1575] "Health Status" refers to the state of a resident's physical and mental health.

[1576] A "meal menu" refers to the specific meal contents and menu to be provided to residents.

[1577] "Delivery service" refers to a service for delivering meal menus to residents.

[1578] This invention is a system that collects and analyzes behavioral and conversational data of residents to understand their rhythms and health status. It also generates an optimal care plan based on this data and notifies care staff, providing appropriate care and reducing their workload. It also includes a function to suggest optimal meal menus based on eating patterns and health status and notify the delivery service.

[1579] The server first collects behavioral and conversational data from residents. Behavioral data includes residents' movements and daily activities, while conversational data includes what residents say. This data is collected using smartphones and IoT sensors and sent to the server. A voice recognition API, commonly known as a voice recognition engine, is used to convert the voice data into text data.

[1580] The server cleanses the collected data, removing incomplete data and noise, and then analyzes it using machine learning algorithms to understand the resident's rhythm and health status. This analysis is performed using a generic machine learning platform. For example, if the server detects that a resident has a habit of eating at a specific time, it analyzes their eating patterns. It also evaluates the resident's emotional state using an emotion analysis engine.

[1581] The server then generates an optimal care plan based on the analysis, including assisted living and medical recommendations. It also suggests optimal meal plans based on the resident's eating patterns and emotional state, and notifies a food delivery service. A food delivery API is used to arrange for the optimal menu to be delivered to the resident.

[1582] This system is used in the following specific example.

[1583] When a resident says "I'm hungry," the behavioral data collection module detects this and converts it into text using a speech recognition engine. The data cleansing module processes this text data, and the behavioral pattern analysis module analyzes eating patterns. The emotion analysis module evaluates the resident's emotional state, and the menu generation module generates an optimal meal menu. The delivery notification module notifies the resident of delivery suggestions using a food delivery API.

[1584] Example prompt sentence:

[1585] When a user says "I'm hungry," the behavioral data collection module detects this and converts it into text using a speech recognition engine. The data cleansing module processes this text data, and the behavioral pattern analysis module analyzes eating patterns. The emotion analysis module evaluates the user's emotional state, and the menu generation module generates an optimal meal menu. The delivery notification module notifies the user of delivery suggestions using a food delivery API.

[1586] This system will enable comprehensive health support for residents, improving their quality of life through proper nutritional intake and linking delivery services, and reducing the workload of care staff.

[1587] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1588] Step 1:

[1589] The terminal uses IoT sensors and voice recognition devices to collect resident behavioral and conversational data. For example, if a resident says "I'm hungry," the voice is detected and converted into text data by a voice recognition engine. The input for this step is the resident's behavior and speech, and the output is text data and behavioral data.

[1590] Step 2:

[1591] The behavioral and conversational data sent from the device is sent to a server and securely stored. This involves building and centrally managing a database for each specific resident. The input for this step is text data and behavioral data, and the output is the results stored in the database.

[1592] Step 3:

[1593] The server cleanses the collected data and removes incomplete data and noise. This is a data preprocessing stage and is an important step for accurate analysis. Specifically, it detects outliers and missing data and removes or imputes them. The input for this step is the stored behavioral data and text data, and the output is the cleansed data.

[1594] Step 4:

[1595] Based on the cleansed data, the server analyzes the data using machine learning algorithms. Common machine learning platforms such as TensorFlow and PyTorch are used here. The analysis identifies resident behavior patterns and daily rhythms and assesses health risks. The input for this step is the cleansed data, and the output is behavior patterns and health risk assessment results.

[1596] Step 5:

[1597] Using an emotion analysis engine, the server recognizes the resident's emotions from the conversation data. Specifically, it uses data from a voice recognition device and uses an emotion analysis API to determine the resident's emotional state from the tone of voice and vocabulary. The input for this step is the conversation data, and the output is the emotion analysis results.

[1598] Step 6:

[1599] Based on the analysis results, the server generates an optimal care plan for each resident. This care plan includes specific suggestions for health management and meal plans. The analysis results are integrated with the emotional state to suggest the optimal meal menu for the resident. This results in a care plan that corresponds to the resident's health condition. The inputs for this step are behavioral patterns, health risk assessment, and emotional analysis results, and the output is a care plan and meal menu suggestions.

[1600] Step 7:

[1601] The generated care plan and meal menu are notified to the nursing staff and delivery service. Using the food delivery API, arrangements are made to quickly deliver the optimal meal menu for the resident. Nursing staff are notified of the resident's status in real time, helping them respond quickly. The input to this step is the care plan and meal menu suggestions, and the output is the notification results.

[1602] Through these steps, this system will provide comprehensive health support for residents, reduce the workload of care staff, and provide residents with appropriate nutritional intake.

[1603] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1604] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1605] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1606] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1607] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1608] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1609] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1610] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1611] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1612] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1613] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1614] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1615] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1616] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1617] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1618] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1619] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1620] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1621] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1622] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1623] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1624] The following is further disclosed regarding the above embodiment.

[1625] (Claim 1)

[1626] a means for collecting resident behavioral and conversational data;

[1627] A means for analyzing the collected data to grasp the rhythm and health condition of the resident;

[1628] A means for generating an optimal care plan based on the rhythm and health condition that have been grasped;

[1629] a means for notifying a care staff member of the generated care plan;

[1630] A system including:

[1631] (Claim 2)

[1632] 10. The system of claim 1, further comprising means for cleansing the collected data.

[1633] (Claim 3)

[1634] The system according to claim 1, further comprising a means for assessing health risks based on the determined rhythms and health conditions.

[1635] "Example 1"

[1636] (Claim 1)

[1637] a means for collecting resident behavioral and conversational data;

[1638] means for transmitting the collected data to a server;

[1639] means for storing the transmitted data in a database;

[1640] means for cleansing the stored data;

[1641] a means for analyzing the cleansed data to grasp the rhythm and health condition of the resident;

[1642] A means for generating an optimal care plan based on the rhythm and health condition that have been grasped;

[1643] a means for notifying a care staff member of the generated care plan;

[1644] A system including:

[1645] (Claim 2)

[1646] 10. The system of claim 1, further comprising means for cleansing the collected data.

[1647] (Claim 3)

[1648] The system according to claim 1, further comprising a means for assessing health risks based on the determined rhythms and health conditions.

[1649] "Application Example 1"

[1650] (Claim 1)

[1651] a means for collecting occupant and automated device behavioral and conversational data;

[1652] A means for analyzing the collected data to grasp the work rhythm and operating status of residents and automated equipment;

[1653] A means for generating an optimal maintenance plan based on the work rhythm and operating status that have been grasped;

[1654] a means for notifying a manager or an operator of the generated maintenance plan;

[1655] A system including:

[1656] (Claim 2)

[1657] 10. The system of claim 1, further comprising means for cleansing the collected data.

[1658] (Claim 3)

[1659] The system according to claim 1, further comprising means for performing risk assessment based on the identified work rhythm and operating status.

[1660] "Example 2: Combining Emotion Engines"

[1661] (Claim 1)

[1662] A means for collecting information on resident behavior and speech;

[1663] A means for analyzing the collected information to grasp the rhythm and health condition of the resident;

[1664] A means for generating an optimal care plan based on the rhythm and health condition that has been determined;

[1665] means for notifying care staff of the generated care plan;

[1666] A means for analyzing the emotions of residents using a voice recognition device and an emotion analysis engine;

[1667] means for adjusting a care plan based on results of said sentiment analysis;

[1668] A system including:

[1669] (Claim 2)

[1670] 10. The system of claim 1, further comprising means for cleansing the collected information.

[1671] (Claim 3)

[1672] The system of claim 1, further comprising a means for assessing health risks based on the determined rhythms and health conditions.

[1673] "Application example 2 when combining emotion engines"

[1674] (Claim 1)

[1675] a means for collecting resident behavioral and conversational data;

[1676] A means for analyzing the collected data to grasp the rhythm and health condition of the resident;

[1677] A means for generating an optimal care plan based on the rhythm and health condition that have been grasped;

[1678] a means for notifying a care staff member of the generated care plan;

[1679] A means for suggesting an optimal meal menu based on the user's eating patterns and health condition;

[1680] means for notifying a delivery service of the proposed meal menu;

[1681] A system including:

[1682] (Claim 2)

[1683] 10. The system of claim 1, further comprising means for cleansing the collected data.

[1684] (Claim 3)

[1685] The system according to claim 1, further comprising a means for assessing health risks based on the determined rhythms and health conditions. [Explanation of symbols]

[1686] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for collecting resident behavioral and conversational data; A means for analyzing the collected data to grasp the rhythm and health condition of the resident; A means for generating an optimal care plan based on the rhythm and health condition that have been grasped; a means for notifying a care staff member of the generated care plan; A system including:

2. 10. The system of claim 1, further comprising means for cleansing the collected data.

3. The system according to claim 1 , further comprising a means for assessing health risks based on the determined rhythms and health conditions.

Images