System

Abstract

A system is provided.SOLUTION: A system comprising: a data collection means; a pre-processing means; a data analysis means; a result generation means; a notification means; and a feedback collection means.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] Caring for dementia patients is extremely complex, requiring the appropriate analysis of a wide range of data to understand the patient's cognitive state. Current care methods make it difficult to comprehensively consider a variety of factors, including medical data, lifestyle behavior, and weather, increasing the burden on caregivers. Caregivers themselves are also prone to stress, often with adverse effects on their mental health. Furthermore, dementia symptoms fluctuate daily, making it difficult to find appropriate ways to deal with each change. Therefore, a system capable of real-time data analysis and providing accurate treatment options is needed. [Means for solving the problem]

[0005] The present invention provides a system including a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means. This system collects and preprocesses a patient's vital data, weather data, dietary image data, voice data, etc. Furthermore, the preprocessed data is analyzed using an AI model to evaluate the patient's cognitive state. Appropriate measures are generated based on the evaluation results and notified to the caregiver's device. This allows the caregiver to understand the patient's condition in real time and provide appropriate care. Furthermore, the feedback collection means makes it possible to incorporate the results of the implemented measures back into the AI ​​model to improve analysis accuracy. In this way, the burden on the caregiver can be reduced and higher quality care can be achieved.

[0006] "Data collection means" refers to devices and systems that collect a variety of information, such as patient vital signs, weather data, dietary image data, and audio data.

[0007] "Preprocessing means" refers to devices or systems that convert and prepare collected data into a format suitable for analysis.

[0008] "Data analysis means" refers to devices or systems that use AI models to analyze preprocessed data and estimate the patient's cognitive state.

[0009] The "result generation means" refers to a device or system that evaluates the cognitive state of a patient and generates an appropriate treatment method based on the analysis results obtained from the data analysis means.

[0010] "Notification means" refers to a device or system that notifies the caregiver's device of the generated results and countermeasures in real time.

[0011] "Feedback collection means" refers to a device or system that re-collects the results of the interventions taken by the caregiver and retrains the system's AI model. [Brief explanation of the drawings]

[0012] [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

[0013] 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.

[0014] First, the terms used in the following description will be explained.

[0015] 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).

[0016] 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.

[0017] 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.

[0018] 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.

[0019] 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."

[0020] [First embodiment]

[0021] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0022] 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.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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.

[0028] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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."

[0033] The present invention provides a system for analyzing a wide range of data, allowing caregivers to grasp the cognitive state of dementia patients in real time and take appropriate measures. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[0034] First, the server uses a data collection means to collect vital data, weather data, dietary image data, and voice data of the patient, thereby enabling comprehensive collection of a wide range of data related to the patient's daily life.

[0035] The server then uses preprocessing means to prepare the collected data in a format suitable for analysis, including converting heart rate data into time series data, food images into a data format for image recognition, and voice data into text.

[0036] The server then uses a data analysis means to input the pre-processed data into an AI model to analyze the patient's cognitive status, which is pre-trained to infer the cognitive status from the data.

[0037] Based on the analysis results, the server uses a result generation means to evaluate the patient's cognitive state and generate appropriate countermeasures. For example, if the patient's heart rate is high and stress is suspected, the server will suggest a countermeasure such as "recommending a walk to relax."

[0038] Next, the terminal uses a notification method to display the analysis results and countermeasures received from the server on the caregiver's device. For example, a message such as "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. A 15-minute light walk is recommended" may be displayed on a smartphone or tablet.

[0039] The user (caregiver) then implements the notified measures. For example, the caregiver may take a short walk with Patient A to reduce the patient's stress.

[0040] The server also uses a feedback collection method to collect the results of the implemented measures. For example, by sending the patient's heart rate after a walk or changes in facial expression back to the system, data can be provided to improve the analysis accuracy of the AI ​​model.

[0041] In this way, the system analyzes the patient's cognitive status in real time and promptly provides appropriate treatment to caregivers, thereby reducing the burden on caregivers and ensuring high-quality care.In addition, a feedback system continuously improves the accuracy of the analysis of the entire system.

[0042] Specific examples are given below.

[0043] Example 1:

[0044] At 7:00 AM, the server collects data on patient A's heart rate, blood pressure, sleep duration (from the smartwatch), current temperature and humidity (from the weather API), and an image of the patient's breakfast (from the smartphone).

[0045] Example 2:

[0046] The server inputs this data into the AI ​​model, and the AI ​​suggests that Patient A's heart rate is higher than normal and that he is not sleeping well at night, which may indicate stress or anxiety.

[0047] Example 3:

[0048] The device displays a message on the caregiver's smartphone such as, "Patient A's cognitive state has deteriorated slightly. This is thought to be due to stress. We recommend a 15-minute light walk."

[0049] Example 4:

[0050] The user (caregiver) checks the notification and tries to reduce stress by taking a walk around the neighborhood with Patient A.

[0051] Example 5:

[0052] The server again collects data on heart rate and changes in the patient's facial expression after the walk, and uses this data to retrain the AI ​​model.

[0053] Through this series of processes, the system constantly assesses the patient's cognitive status based on the latest information and provides appropriate advice to caregivers, thereby achieving higher quality care and reducing the burden on caregivers.

[0054] The processing flow will be explained below.

[0055] Step 1:

[0056] The server collects the patient's vital signs, weather data, meal image data, and audio data via a smartwatch, weather API, smartphone camera, audio recording device, etc. Specifically, at 7:00 AM, the server obtains Patient A's heart rate, blood pressure, sleep data, current temperature and humidity, and an image of his breakfast.

[0057] Step 2:

[0058] The server preprocesses the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure data into time-series data, converting food images into a format suitable for image recognition, and converting voice data into text. Specifically, the server organizes the heart rate data and checks for any abnormal values.

[0059] Step 3:

[0060] The server then inputs the preprocessed data into an AI model for analysis. The AI ​​model then analyzes the current data based on the previously learned data and predicts the patient's cognitive state. For example, if the patient's heart rate is higher than normal and they are not getting enough sleep, the AI ​​model will generate a result suggesting stress or anxiety.

[0061] Step 4:

[0062] The server evaluates the patient's cognitive state based on the analysis results obtained from the AI ​​model and generates appropriate treatment methods. For example, if stress is determined to be the cause, it generates specific advice such as "recommending a walk to relax."

[0063] Step 5:

[0064] The server sends the generated results and recommended actions to the caregiver's device. The device (the caregiver's smartphone or tablet) displays the results in real time. For example, a notification message might read, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0065] Step 6:

[0066] The user (caregiver) checks the notification and takes the recommended action, for example, taking a walk around the neighborhood with Patient A.

[0067] Step 7:

[0068] The user (caregiver) records the results of the measures taken and provides feedback to the server. Specifically, the user records the patient's heart rate and changes in facial expression after the walk using a smartphone and sends the results to the server.

[0069] Step 8:

[0070] The server retrains the AI ​​model based on the collected feedback data to improve future analysis accuracy, for example, by using the feedback data to re-learn the relationship between stress levels and heart rate.

[0071] Through this series of processes, the system constantly assesses the patient's cognitive status based on the latest information and provides appropriate advice to caregivers, thereby achieving higher quality care and reducing the burden on caregivers.

[0072] Example 1

[0073] 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."

[0074] In conventional dementia care, it has been difficult for caregivers to grasp the patient's cognitive state in real time and provide appropriate treatment promptly. Furthermore, there has been a lack of mechanisms for appropriately collecting feedback and utilizing it to improve the system's analysis accuracy. This has increased the burden on caregivers, making it difficult to provide high-quality care.

[0075] 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.

[0076] In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means. This allows the server to collect and analyze the patient's vital data, environmental data, nutritional data, and voice data in real time. The server also notifies the user device of the generated evaluation results and countermeasures, and recollects data after the countermeasures are implemented to improve the analytical accuracy of the AI ​​model. This reduces the burden on caregivers and enables the provision of high-quality care.

[0077] "Data collection tools" are devices and techniques for collecting patient vital, environmental, nutritional, and audio data.

[0078] "Preprocessing means" are devices and techniques used to convert collected data into a form suitable for analysis.

[0079] "Data analysis means" refers to the equipment and technology used to input pre-processed data into an AI model for analysis.

[0080] The "result generation means" refers to devices and technologies for evaluating the cognitive state of a patient based on the results of data analysis and generating appropriate treatment methods.

[0081] The "notification means" refers to the device and technology for notifying the generated evaluation results and countermeasures to the user device.

[0082] The "feedback collection means" refers to devices and techniques for collecting data based on the results of measures taken by users and improving the accuracy of the system's analysis.

[0083] "Vital data" refers to a patient's biological information such as heart rate, blood pressure, and sleep time.

[0084] "Environmental data" refers to data related to the external environment, such as temperature and humidity.

[0085] "Nutrition data" refers to data on the types, amounts, and nutrients of food analyzed from dietary image data.

[0086] "Audio data" refers to data recorded from patients' everyday conversations and voice information.

[0087] An "AI model" is a pre-trained artificial intelligence system that includes algorithms for inferring a patient's cognitive state from input data.

[0088] "User device" refers to an electronic device, such as a smartphone or tablet, used by a caregiver.

[0089] The present invention provides a system for analyzing a wide range of data, allowing caregivers to grasp the cognitive state of dementia patients in real time and take appropriate measures. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[0090] Data collection

[0091] The server uses the data collection means to collect the following data:

[0092] Vital data: Collect the patient's heart rate, blood pressure, and sleep duration through the smartwatch.

[0093] Environmental data: Use the weather API to obtain temperature and humidity.

[0094] Nutritional data: Photos of meals are taken from the patient's smartphone and converted into data for nutritional analysis.

[0095] Voice data: The patient's everyday conversations and voices are recorded with a microphone and sent to the server.

[0096] Examples:

[0097] At 7:00 AM, the server accesses Patient A's smartwatch and captures the patient's heart rate and blood pressure data for the day.

[0098] Get the day's temperature and humidity data from the weather API.

[0099] Images of the day's breakfast are automatically uploaded from the patient's smartphone.

[0100] Data Preprocessing

[0101] The server uses pre-processing means to prepare the collected data in a suitable format for analysis.

[0102] Convert heart rate data: Format heart rate data into time series data.

[0103] Nutrition data conversion: Convert images into a format that is easy for AI to interpret.

[0104] Voice to Text Conversion: We use voice recognition software (e.g., Google Speech-to-Text) to convert voice data into text.

[0105] Examples:

[0106] The server takes the heart rate data and converts it into time series data.

[0107] Image recognition software is used to identify the types and quantities of food in the breakfast image and prepare it in a format suitable for analysis.

[0108] Everyday conversations received via voice are converted into text and prepared for sentiment analysis.

[0109] Data analysis and result generation

[0110] The server uses a data analysis means to input the pre-processed data into an AI model to analyze the patient's cognitive status, which is pre-trained to infer the cognitive status from the data.

[0111] Based on the analysis results, the server uses the result generation means to assess the patient's cognitive state and generate an appropriate course of action.

[0112] Examples:

[0113] The pre-processed heart rate data is fed into an AI model to detect values ​​outside the normal range.

[0114] Image data of breakfast is input into an AI model and the nutritional balance is analyzed.

[0115] The text data obtained by speech recognition is input into a sentiment analysis model to detect changes in emotion.

[0116] notification

[0117] The terminal uses a notification means to display the analysis results and countermeasures received from the server on the user's (caregiver's) device.

[0118] Notification method: A message will appear on your smartphone or tablet.

[0119] Notification content: Includes analysis results and details of how to deal with the issue.

[0120] Examples:

[0121] The caregiver's smartphone will receive a notification saying, "Patient A's cognitive status has deteriorated slightly. This is thought to be due to stress. We recommend a 15-minute light walk."

[0122] Implementing countermeasures

[0123] The user (caregiver) takes the notified action.

[0124] Action to be taken: Take the notified action.

[0125] Specific action: For example, a caregiver takes a walk around the neighborhood with patient A.

[0126] Examples:

[0127] The caregiver asks Patient A, "Let's take a 15-minute walk together," and they actually go for a walk.

[0128] Feedback collection

[0129] The server uses a feedback collection means to collect the results of the actions taken.

[0130] Feedback items: Heart rate after a walk, changes in the patient's facial expression, daily reactions, etc.

[0131] Data reuse: Collected data can be used to retrain AI models to improve analysis accuracy.

[0132] Examples:

[0133] After the walk, heart rate data is collected again from the smartwatch to assess the patient's stress level.

[0134] The patient's facial expressions and tone of voice are analyzed from camera and audio data and collected as feedback.

[0135] Example prompts to input to the generative AI model

[0136] "Patient A has a higher than normal heart rate and is not sleeping well at night. He is showing signs of stress and anxiety. Please suggest ways to address this."

[0137] "Based on Patient B's vital signs and dietary information today, please assess his cognitive status and suggest appropriate treatment options."

[0138] Through this series of processes, the system analyzes the patient's cognitive state in real time and promptly provides appropriate measures to caregivers, thereby reducing their burden and ensuring high-quality care. In addition, a feedback system continuously improves the analysis accuracy of the entire system.

[0139] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0140] Step 1:

[0141] Data collection

[0142] The server collects the following data using the data collection means:

[0143] Input: Patient's vital data (heart rate, blood pressure, sleep time), environmental data (temperature, humidity), nutritional data (images of meals), and audio data (daily conversation).

[0144] Specific operation: The server retrieves heart rate, blood pressure, and sleep time data from Patient A's smartwatch at 7:00 AM. It also retrieves the temperature and humidity for that day using a weather API. It also uploads an image of breakfast from the patient's smartphone and records daily conversations using the microphone.

[0145] Output: Collected vital, environmental, nutritional and audio data.

[0146] Step 2:

[0147] Data Preprocessing

[0148] The server uses pre-processing means to prepare the collected data in a suitable format for analysis.

[0149] Input: Collected vital, environmental, nutritional, and audio data.

[0150] How it works: The server formats the heart rate data into time series data, converts the image of the breakfast into a format that is easy for AI to understand, and converts the audio data into text using speech recognition software.

[0151] Output: Preprocessed heart rate data, image data, and audio data converted to text.

[0152] Step 3:

[0153] Data analysis

[0154] The server uses data analysis means to input the preprocessed data into an AI model to analyze the patient's cognitive status.

[0155] Input: Preprocessed heart rate data, environmental data, nutrition data, and audio data.

[0156] Specific operations: Preprocessed heart rate data is input into an AI model to detect abnormal values. Image data of breakfast is input into an AI model to analyze nutritional balance. Text-converted voice data is input into an emotion analysis model to analyze changes in emotion.

[0157] Output: Analysis of the patient's cognitive status.

[0158] Step 4:

[0159] result generation

[0160] The server uses the result generation means to evaluate the cognitive state of the patient based on the analysis results and generates an appropriate course of action.

[0161] Input: Analysis results.

[0162] Specific behavior: Integrates the results of multiple analyses to assess the patient's cognitive state (e.g., stress level, lack of sleep, etc.), and uses a generative AI model to generate appropriate countermeasures (e.g., recommending a relaxing walk).

[0163] Output: Evaluation results and suggested solutions.

[0164] Step 5:

[0165] notification

[0166] The terminal uses the notification means to display the analysis results and countermeasures received from the server on the user device.

[0167] Input: Evaluation results and suggested actions.

[0168] What it does: Sends a detailed message to the caregiver's smartphone about the patient's cognitive status and how to address it. The notification might say, "Patient A's cognitive status has slightly deteriorated. This is thought to be due to stress. We recommend a 15-minute brisk walk."

[0169] Output: The notification message displayed on the user device.

[0170] Step 6:

[0171] Implementing countermeasures

[0172] The user (caregiver) takes the notified action.

[0173] Input: Notified suggested action.

[0174] Specific actions: The caregiver takes a walk around the neighborhood with Patient A. They say, "Let's take a 15-minute walk together," and then actually go for a walk.

[0175] Output: The outcome of the intervention (e.g., the patient's condition after the walk).

[0176] Step 7:

[0177] Feedback collection

[0178] The server collects the results of the actions taken using a feedback collection means.

[0179] Input: Results of the implemented measures (heart rate data, facial expression changes, etc.).

[0180] Specific operations: After the walk, the smartwatch collects heart rate data again to assess the patient's stress level, and analyzes the patient's facial expressions and tone of voice via camera and audio data to provide feedback.

[0181] Output: Collected feedback data, analysis results of improved AI model.

[0182] (Application example 1)

[0183] 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."

[0184] Until now, there has been no system that monitors a driver's fatigue and stress state in real time and provides appropriate countermeasures based on that information. This has made it difficult to improve driver safety and performance, potentially increasing the risk of traffic accidents. Therefore, the present invention aims to improve driver safety and comfort by analyzing the driver's vital signs, environmental data, and image data in real time and providing appropriate notifications and feedback according to the driver's condition.

[0185] 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.

[0186] In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, a means for collecting vital data, environmental data, and image data of the driver, a means for analyzing the data and generating recommendations based on the driver's condition, and a means for notifying the driver's device of the recommendations. This makes it possible to monitor the driver's fatigue and stress in real time and quickly provide appropriate countermeasures.

[0187] The "data collection means" is a means for acquiring various data including vital data of the driver, environmental data, and image data.

[0188] "Preprocessing means" refers to means for converting collected data into a format that is easy to analyze.

[0189] "Data analysis means" means a means for using pre-processed data to assess the driver's condition and analyzing it using specific algorithms or generative AI models.

[0190] The "result generation means" is a means for generating specific recommendations or notifications for the driver based on the results of the data analysis.

[0191] "Notification means" refers to a means for transmitting generated recommendations and notifications to the driver's device.

[0192] The "feedback collection means" is a means for recording the driver's response to the provided countermeasures and the progress of the improvement, and for use in reanalysis and model improvement.

[0193] "Vital data" refers to biometric information such as the driver's heart rate, blood pressure, and sleep data.

[0194] "Environmental data" refers to data relating to climatic conditions and the surrounding environment, including, for example, temperature, humidity, and weather information.

[0195] "Image data" refers to images of the driver's face and footage of the inside of the vehicle, and is data used to analyze the driver's facial expression and posture.

[0196] A "generative AI model" is a pre-trained artificial intelligence model that infers the driver's state based on collected data and generates appropriate recommendations.

[0197] The present invention provides a system for analyzing a driver's vital data, environmental data, and image data in real time and providing appropriate countermeasures based on the driver's condition. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[0198] The server uses data collection means to collect driver vital data, weather data, and driver image data. This data collection is done through wearable devices such as smartwatches and fitness trackers, weather APIs, and in-car cameras. For example, heart rate, blood pressure, and sleep data are acquired.

[0199] The collected data is preprocessed by the server. The preprocessing means standardizes vital data such as heart rate and blood pressure, converts image data to grayscale, and resizes it to an appropriate size. Weather data is obtained using an API and converted into a format suitable for analysis.

[0200] The server then inputs the preprocessed data into a generative AI model for data analysis. The generative AI model is pre-trained to infer the driver's state and detect conditions such as fatigue, stress, and distraction. Based on the results of this analysis, a specific recommendation is generated for the driver. For example, a message such as "You need a break. Please pull over and take a break" is generated.

[0201] The generated results are sent to the driver's device, such as a smartphone or an in-vehicle display, using a notification means, allowing the driver to immediately check their own condition and take any necessary measures.

[0202] After the driver actually takes the recommended action, the server uses a feedback collection method to collect the driver's reaction data again. For example, heart rate and facial expression changes after a break are collected and used to retrain the AI ​​model. This feedback function further improves the system's analysis accuracy.

[0203] As a concrete example, the following scenario can be considered.

[0204] Data collected:

[0205] Heart rate: 85 bpm

[0206] Blood pressure: 130 / 85 mmHg

[0207] Sleep data: 6 hours of sleep

[0208] Weather data: 22°C, 65% humidity

[0209] Driver's facial image

[0210] Generated recommendations:

[0211] The driver is likely fatigued, so a notification is generated saying, "You need a break. Please stop and take a break."

[0212] An example of a prompt is as follows:

[0213] "Based on the collected data, a generative AI model will be used to analyze the driver's condition and generate a notification that a break is needed, if necessary.

[0214] In this way, the present invention monitors the driver's health condition in real time and helps prevent traffic accidents and health risks.

[0215] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0216] Step 1:

[0217] The server uses data collection means to collect the driver's vital signs, weather data, and image data of the driver. Specifically, it obtains heart rate, blood pressure, and sleep data from the wearable device, and obtains current weather data (temperature, humidity, etc.) using a weather API. It also captures images of the driver's face using the in-car camera. This allows it to collect a wide range of data on the driver's condition and environment.

[0218] Input: Vital data from wearable devices, weather data from weather API, image data of the driver from an in-car camera

[0219] Output: Raw data collected

[0220] Step 2:

[0221] The server uses preprocessing tools to convert the collected data into a format suitable for analysis. Heart rate and blood pressure are normalized to a standard scale, sleep data is converted into an index to evaluate sleep quality, weather data is formatted from the API, and image data is converted to grayscale and resized to a size suitable for analysis. This prepares the data in a unified format.

[0222] Input: Raw data collected

[0223] Output: Preprocessed data

[0224] Step 3:

[0225] The server uses data analysis means to input preprocessed data into the generative AI model, which analyzes the driver's condition. The generative AI model combines and analyzes heart rate, blood pressure, sleep data, weather data, and image data to estimate the driver's fatigue level, stress level, distraction state, etc. The AI ​​model is pre-trained and uses this data to predict the driver's condition with high accuracy.

[0226] Input: Preprocessed data

[0227] Output: Analysis results (driver's current status evaluation)

[0228] Step 4:

[0229] The server generates specific recommendations for the driver based on the analysis results using the result generation means. For example, a message such as "The driver is highly fatigued, so a 15-minute break is recommended." This recommendation serves as a specific guideline for action to ensure the driver's safety.

[0230] Input: Analysis results (driver's current status evaluation)

[0231] Output: Recommendations

[0232] Step 5:

[0233] The terminal uses notification methods to notify the driver of the generated recommendations. A message is displayed on the driver's smartphone or in-vehicle display, allowing the driver to immediately check the contents. This allows the driver to recognize their own condition and take any necessary measures promptly.

[0234] Input: Recommendations

[0235] Output: Notification to the driver

[0236] Step 6:

[0237] The user (driver) takes the appropriate action. For example, after checking the notification on the device, the user stops the car in a safe place and takes a break for the specified amount of time. This is expected to reduce the driver's fatigue and stress.

[0238] Input: Notification content

[0239] Output: Coping behavior

[0240] Step 7:

[0241] The server collects the results of the implemented measures using a feedback collection method. Feedback data such as heart rate, blood pressure, and changes in the driver's facial expressions after the break are collected again and used as training data for the generative AI model. This feedback improves the accuracy of the system.

[0242] Input: New vital data and image data after response actions

[0243] Output: Feedback data

[0244] Through the above series of processing steps, the system of the present invention can monitor the driver's health condition in real time and provide appropriate countermeasures promptly, thereby reducing the risk of traffic accidents and improving the safety and comfort of the driver.

[0245] 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.

[0246] The present invention is a system for supporting care of dementia patients, and includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion engine. The system aims to analyze a wide range of data, grasp the cognitive state of the patient in real time, and provide appropriate measures to caregivers.

[0247] First, the server collects vital data (e.g., heart rate, blood pressure), weather data (e.g., temperature, humidity), meal image data, and voice data through a data collection means. It also uses an emotion engine to recognize emotions from the voice data and facial expression data of users (caregivers and patients).

[0248] The server then uses preprocessing means to convert the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure into time-series data, converting food images into a format suitable for image recognition, and converting voice data into text. Furthermore, an emotion engine extracts emotions from the user's voice and facial expressions and records them as data.

[0249] The server then uses data analysis tools to input the preprocessed data into the AI ​​model to analyze the patient's cognitive state. The AI ​​model uses past learning data and analyzes current data to infer cognitive state and stress level. By including emotional data obtained from the emotion engine in this analysis, more detailed insights into the mental state of the patient and caregiver can be gained.

[0250] Based on the analysis results, the server uses the result generation means to determine an appropriate course of action. Specifically, if it determines that the cognitive state is slightly deteriorating, it generates a course of action to reduce stress, such as "recommending a walk to relax." It can also suggest a course of action that takes emotions into account by referring to the data obtained through the emotion engine.

[0251] The server then sends the generated results and advice on how to respond to the situation to the caregiver's device. The device (the caregiver's smartphone or tablet) displays the received notification in real time. An example message might read, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0252] The user (caregiver) then checks the notification received from the device and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the emotion engine also monitors changes in the caregiver's and patient's emotions and collects further data.

[0253] Finally, the user (caregiver) sends the results of the measures they took as feedback to the server. The server analyzes the collected feedback data and uses it to retrain the AI ​​model to improve its accuracy. In particular, since this feedback includes emotional data, it can make the next recommendation more accurate.

[0254] As a specific example, at 7:00 AM, the server collects Patient A's heart rate, sleep data, current temperature and humidity, and an image of their breakfast. At the same time, the emotion engine analyzes the caregiver's morning voice tone and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that Patient A's heart rate is higher than normal and that they are not getting enough sleep, suggesting stress. The result generation method generates a solution such as, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. We recommend a 15-minute light walk." The device then sends a notification to the caregiver, who then follows the recommendation and takes the walk, providing feedback on the results. This feedback data is collected and analyzed by the server to help further improve the system.

[0255] In this way, the system analyzes data from multiple perspectives and provides appropriate measures that take emotional data into consideration, thereby providing effective support for both caregivers and patients.

[0256] The processing flow will be explained below.

[0257] Step 1:

[0258] The server uses data collection tools to collect the patient's vital signs, weather data, meal image data, voice data, and user emotion data. For example, it obtains heart rate and blood pressure from a smartwatch and current temperature and humidity from a weather API. Meanwhile, the caregiver's smartphone camera takes pictures of the patient's breakfast and records the conversation using a voice recording device.

[0259] Step 2:

[0260] The server uses an emotion engine to analyze emotions from the voice and facial expression data of the caregiver and patient. For example, it analyzes emotions from voice and recognizes emotions such as stress and anxiety from the tone of the voice. It also analyzes the patient's facial expression from captured images to extract emotions such as smile and tension.

[0261] Step 3:

[0262] The server uses preprocessing means to convert the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure data into time-series data, converts food image data into a format suitable for image recognition, converts voice data into text, and quantifies and stores emotion data.

[0263] Step 4:

[0264] The server uses data analysis tools to input preprocessed data into the AI ​​model, which analyzes the patient's cognitive state. The AI ​​model suggests possible stress based on elevated heart rate and lack of sleep, and provides a complementary assessment of the degree of stress and anxiety based on the results of emotion analysis.

[0265] Step 5:

[0266] The server uses the result generation means to evaluate the patient's current cognitive state based on the analysis results obtained from the AI ​​model and generate appropriate measures to address the situation. For example, it generates specific advice such as, "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. We recommend a 15-minute light walk."

[0267] Step 6:

[0268] The server sends the generated results and recommended actions to the caregiver's device via a notification method. The device (caregiver's smartphone or tablet) displays this notification in real time. The caregiver receives a message saying, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0269] Step 7:

[0270] The user (caregiver) checks the notification and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the caregiver carefully observes the patient's reactions and records any changes in their emotions.

[0271] Step 8:

[0272] The user (caregiver) sends the results of the measures taken to the server through a feedback collection means. Specifically, the user records the patient's heart rate after the walk, changes in the patient's facial expression, and changes in the caregiver's own emotions using a smartphone and sends them to the server.

[0273] Step 9:

[0274] The server retrains the AI ​​model based on the data obtained through the feedback collection method to improve the accuracy of the next analysis. Based on the feedback data, the relationship between stress state and changes in heart rate and emotions is reevaluated and reflected in future countermeasures.

[0275] Through the above processing flow, the system can analyze the patient's cognitive state and emotions from multiple angles and provide caregivers with appropriate coping strategies in real time, thereby improving the quality of care and reducing the burden on caregivers.

[0276] Example 2

[0277] 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."

[0278] In providing care support for dementia patients, it is extremely important to understand the patient's cognitive state and emotional changes in real time and provide caregivers with appropriate measures. However, current systems have fragmented data collection and analysis, making it difficult to comprehensively understand the patient's overall situation. In addition, measures that take emotional data into consideration are insufficient, which hinders caregivers from providing optimal support.

[0279] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion recognition means. This makes it possible to collect biological data, environmental data, image data, and voice data of a patient from various sources, and through preprocessing and analysis, grasp the patient's cognitive state and emotions in real time and provide appropriate measures promptly.

[0280] The "data collection means" is a means for collecting various data such as patient biometric data, environmental data, image data, and voice data.

[0281] "Preprocessing means" refers to the means of converting collected data into a format suitable for analysis. Specifically, it generates time-series data, converts it into an image recognition format, converts voice data into text, etc.

[0282] The "data analysis means" is a means for inputting preprocessed data into the AI ​​model and analyzing it, which allows for the prediction of the patient's cognitive state and stress level.

[0283] The "result generation means" is a means for determining and generating appropriate measures based on the results of data analysis. Specifically, it suggests relaxation techniques, etc.

[0284] The "notification means" is a means for transmitting the generated countermeasure to the caregiver's device and displaying the content thereof.

[0285] The "feedback collection means" is a means for collecting the results of the measures taken by the caregiver and using them for subsequent analysis and to improve the accuracy of the AI ​​model.

[0286] "Emotion recognition means" is a means of extracting emotions from voice data and facial expression data and recording them as data. This data plays an important role in data analysis.

[0287] An "AI model" is an artificial intelligence model that analyzes collected data and generates recommendations, enabling highly accurate analysis using past data.

[0288] "Vital data" refers to biological information such as heart rate, blood pressure, and body temperature.

[0289] "Image data" refers to image information captured with a camera of food, facial expressions, etc.

[0290] "Audio data" refers to data that indicates acoustic information that records user speech and environmental sounds.

[0291] "Environmental data" refers to data relating to the external environment, such as temperature, humidity, and weather.

[0292] The present invention is a system for supporting care of dementia patients, which includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion recognition means. The system aims to analyze a wide range of data, grasp the cognitive state of the patient in real time, and provide appropriate measures to caregivers.

[0293] First, the server collects vital data (e.g., heart rate, blood pressure, body temperature, etc.), environmental data (e.g., temperature, humidity, etc.), image data (photos of meals and facial expressions), and voice data through a data collection means. Vital data is obtained from a wearable device (e.g., Fitbit or Apple Watch). Environmental data is acquired from a weather API (e.g., OpenWeatherMap API). Image data is collected through the caregiver's smartphone camera, and voice data is obtained through a dedicated app. In addition, an emotion recognition means is used to recognize emotions from the user's voice data and facial expression data.

[0294] Next, the server uses preprocessing means to convert and organize the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure data into time-series data, converts food images into a format suitable for image recognition (e.g., RGB pixel array), and converts audio data into text using speech recognition software (e.g., Google Cloud Speech-to-Text API). Furthermore, it uses emotion recognition means to extract emotions from voice and facial expressions and record them as data.

[0295] The server then uses data analysis tools to input the preprocessed data into an AI model (e.g., a model implemented in TensorFlow or PyTorch) to analyze the patient's cognitive state. The analysis involves analyzing past training data and using current data to estimate cognitive state and stress levels. Emotion data obtained from emotion recognition tools can also be incorporated into this analysis to provide detailed insights into the mental state of the patient and caregiver.

[0296] Based on the analysis results, the server uses a result generation method to determine an appropriate course of action. For example, if it determines that the cognitive state is slightly deteriorating, it will generate a suggestion such as "recommended walk to relax" as a measure to reduce stress. It is also possible to refer to emotional data and propose a course of action that takes emotions into consideration.

[0297] The server then sends the generated results and advice on how to respond to the situation to the caregiver's device, which then displays the received notification in real time. For example, a message might appear saying, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0298] The user (caregiver) then checks the notification received from the device and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the emotion recognition means monitors changes in the emotions of the caregiver and the patient and collects further data.

[0299] Finally, the user (caregiver) sends the results of the measures they took as feedback to the server. The server analyzes the collected feedback data and uses it to retrain the AI ​​model to improve its accuracy. In particular, since this feedback includes emotional data, it can make the next recommendation more accurate.

[0300] Examples:

[0301] At 7:00 AM, the server collects Patient A's heart rate, sleep data, current temperature and humidity, and an image of their breakfast. At the same time, the emotion recognition means analyzes the caregiver's morning tone of voice and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that Patient A's heart rate is higher than normal and that they are not getting enough sleep, suggesting stress. The result generation means generates a countermeasure such as, "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. We recommend a 15-minute light walk." The device then sends a notification to the caregiver, who then follows the recommendation and takes the walk, providing feedback on the results. This feedback data is collected and analyzed by the server to help further improve the system.

[0302] In this way, the system analyzes data from multiple perspectives and provides appropriate measures that take emotional data into consideration, thereby providing effective support for both caregivers and patients.

[0303] Example prompt sentence:

[0304] "Patient A's cognitive status has deteriorated slightly. This is likely due to stress. A 15-minute brisk walk is recommended."

[0305] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0306] Step 1: Data collection

[0307] The server collects various data through the data collection means.

[0308] Input: Patient vital data (heart rate, blood pressure, body temperature, etc.), environmental data (temperature, humidity, etc.), image data (photos of meals, images of facial expressions), voice data

[0309] Processing: Vital data is collected from wearable devices (e.g., Fitbit or Apple Watch). Environmental data is obtained from weather APIs (e.g., OpenWeatherMap API). Image data is collected through the caregiver's smartphone camera, and audio data is obtained through a dedicated app.

[0310] Output: A wide variety of raw data collected

[0311] Step 2: Data Preprocessing

[0312] The server uses pre-processing means to convert the collected data into a form suitable for analysis.

[0313] Input: Raw data collected

[0314] Processing: Heart rate and blood pressure data are converted into time-series data, and food images are converted into a format for image recognition (e.g., RGB pixel arrays). Audio data is converted into text using speech recognition software (e.g., Google Cloud Speech-to-Text API). Emotion recognition tools are used to extract emotions from voice and facial expressions and record them as data.

[0315] Output: Preprocessed datasets (time series data, formatted image data, text data, emotion data)

[0316] Step 3: Data analysis

[0317] The server uses data analysis means to input the preprocessed data into the AI ​​model and analyze the patient's cognitive status.

[0318] Input: Preprocessed datasets (time series data, formatted image data, text data, emotion data)

[0319] Processing: Data is fed into an AI model (e.g., a model implemented in TensorFlow or PyTorch) and current data is analyzed based on past training data. This allows for estimation of cognitive state and stress level. Emotion data obtained through emotion recognition methods is also included in the analysis.

[0320] Output: Analysis results (predictions of cognitive state, stress level, etc.)

[0321] Step 4: Result Generation

[0322] The server determines and generates an appropriate countermeasure based on the analysis result using the result generation means.

[0323] Input: Analysis results (predicted results such as cognitive state, stress level, etc.)

[0324] Processing: If the cognitive state is deteriorating, it will determine measures to reduce stress, such as suggesting a relaxing walk or a relaxation session. It will also take into account emotional data to generate coping strategies.

[0325] Output: A message with the recommendations

[0326] Step 5: Notification Delivery

[0327] The server transmits the generated results and actions to the caregiver's device.

[0328] Input: Message with recommendations

[0329] Processing: The generated message is sent to the caregiver's device (smartphone or tablet).

[0330] Output: Notification message displayed on the caregiver's device

[0331] Step 6: Implement the recommendations

[0332] The user (caregiver) checks the notification and takes the recommended action.

[0333] Input: Notification message

[0334] Action: Review the notification and take the recommended action with the patient, for example, take a walk with the patient.

[0335] Output: Action taken

[0336] Step 7: Gather feedback

[0337] The user (caregiver) sends the results of the execution to the server as feedback.

[0338] Input: Execution results (e.g., changes in the patient's condition or emotions after the walk)

[0339] Processing: The results of the countermeasures are recorded via the device and sent to the server, which collects and analyzes the feedback data and retrains the AI ​​model.

[0340] Output: Feedback data is stored and used on the server

[0341] Throughout this entire process, the system analyzes the data from multiple angles and provides appropriate responses that take emotional data into account, thereby providing effective support for both caregivers and patients.

[0342] (Application example 2)

[0343] 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."

[0344] In factories and other workplaces, it is necessary to grasp the physical and mental health status of employees in real time and quickly provide appropriate countermeasures. However, currently there is a lack of systems that can accurately analyze workload and stress and suggest appropriate countermeasures. This presents a problem in that it is not possible to provide appropriate support to ensure employee efficiency and safety.

[0345] 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 a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion analysis means. This makes it possible to analyze the physical and mental health conditions of workers in real time and quickly provide appropriate countermeasures.

[0346] The "data collection means" is a system element that collects vital data, work environment data, work image data, and voice data of workers.

[0347] A "preprocessing means" is an element of the system that converts and prepares collected data into a form suitable for analysis.

[0348] "Data analysis means" refers to the system element that inputs preprocessed data, including emotional data, into the AI ​​model for analysis.

[0349] A "result generation means" is an element of the system that determines the appropriate course of action based on the results of data analysis.

[0350] The "notification means" is a system element that sends the generated results and how to deal with them to the worker's device.

[0351] The "feedback collection means" is a system element that sends the results of the implemented countermeasures to the server and collects them.

[0352] "Emotion analysis means" is an element of the system that recognizes and analyzes emotions from the worker's facial expressions and voice data.

[0353] The present invention provides a system for monitoring and analyzing the health status and stress levels of employees in the workplace in real time, which includes a data collection means, a pre-processing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and a sentiment analysis means.

[0354] First, the server collects the worker's vital data (e.g., heart rate, blood pressure), work environment data (e.g., temperature, humidity), work image data, and voice data through a data collection means. It also uses an emotion analysis means to recognize the worker's emotions from their facial expressions and voice data.

[0355] Next, the server uses preprocessing means to convert the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure into time-series data, converting work images into a format suitable for image recognition, and converting voice data into text. Furthermore, emotion analysis means extracts emotions from the worker's voice and facial expressions and records them as data.

[0356] The server then uses data analysis means to input the preprocessed data into the AI ​​model, which analyzes the worker's health condition and stress level. The AI ​​model uses past learning data and analyzes current data to estimate stress levels and fatigue levels. By including emotional data obtained by the emotion analysis means in this analysis, detailed insights into the worker's mental state can be gained.

[0357] Based on the analysis results, the server uses the result generation means to determine an appropriate response. Specifically, if the stress level is determined to be high, the server generates a response such as recommending a break. Using data obtained through the emotion analysis means, the server can also suggest a response that takes emotions into consideration.

[0358] Next, the server sends the generated results and countermeasures to the worker's device (terminal). The terminal displays the received notification in real time. An example message might be, "Worker A's stress level is high, so we recommend that he take a 10-minute break."

[0359] The user (worker) then checks the notification received from the device and takes the recommended action. For example, the worker takes a 10-minute break in the break room. During this action, the emotion analysis means also monitors changes in the worker's emotions and collects further data.

[0360] Finally, the user (worker) sends the results of the implemented measures to the server as feedback. The server analyzes the collected feedback data and uses it for retraining to improve the accuracy of the AI ​​model. In particular, since this feedback includes emotional data, it will make the next recommendation more accurate.

[0361] As a specific example, at 7:00 AM, the server collects worker A's heart rate, sleep data, and current work environment data, while the emotion analysis means simultaneously analyzes the worker's morning tone of voice and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that worker A's heart rate is higher than normal and that he or she is highly fatigued. The result generation means generates a countermeasure such as, "Worker A's stress level is high, so we recommend that he or she take a 10-minute break." The device then sends a notification to the worker, and the user (worker) follows the recommendation and takes a break, providing feedback on the results. This feedback data is collected and analyzed by the server and used to further improve the system.

[0362] This system analyzes data from multiple angles and provides appropriate countermeasures that take emotional data into consideration, thereby effectively supporting both worker health and work efficiency.

[0363] Example prompt sentence:

[0364] Please explain the system that collects and analyzes the heart rate, blood pressure, facial expressions and voice data of factory workers, and recommends appropriate breaks and ways to deal with stress when the worker's stress level is high.

[0365] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0366] Step 1:

[0367] The server uses data collection means to collect worker vital data (heart rate, blood pressure), work environment data (temperature, humidity), work image data, and voice data. In this step, the input is real-time data obtained from various sensors and cameras, and the output is raw data compiled by the data collection means.

[0368] Step 2:

[0369] The server uses preprocessing means to convert and organize the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure into time-series data, converts work images into a format for image recognition, and converts voice data into text. The input to this step is the raw data collected in step 1, and the output is preprocessed data.

[0370] Step 3:

[0371] The server uses emotion analysis means to recognize and extract emotions from the worker's facial and voice data. The emotion data is used for subsequent analysis. The input for this step is facial and voice data obtained from a camera and microphone, and the output is emotion data recognized by the emotion analysis means.

[0372] Step 4:

[0373] The server uses data analysis means to input the preprocessed data and emotional data into the AI ​​model to analyze the worker's health condition and stress level. The AI ​​model performs analysis based on past learning data and makes an evaluation based on current data. The inputs for this step are the preprocessed data and emotional data, and the output is the analysis results.

[0374] Step 5:

[0375] The server uses the result generation means to determine an appropriate course of action based on the analysis results. For example, if the stress level is determined to be high, it generates a course of action such as recommending a break. The input of this step is the analysis result of step 4, and the output is a course of action.

[0376] Step 6:

[0377] The server uses the notification means to send the generated results and the countermeasures to the worker's terminal. The terminal displays the received notification to the worker in real time. The input of this step is information about the countermeasures, and the output is the notification sent to the worker's terminal.

[0378] Step 7:

[0379] The user (worker) checks the notification received from the terminal and takes the recommended action. For example, the user takes a specific action such as taking a 10-minute rest in the break room. The input of this step is the notification from the terminal, and the output is the action taken.

[0380] Step 8:

[0381] The user (operator) sends the results of the countermeasures they have taken as feedback to the server. The server analyzes the collected feedback data and uses it for retraining to improve the accuracy of the AI ​​model. The input of this step is the feedback information from the user, and the output is the data used for retraining.

[0382] 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.

[0383] 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.

[0384] 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.

[0385] [Second embodiment]

[0386] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0387] 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.

[0388] 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).

[0389] 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.

[0390] 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.

[0391] 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).

[0392] 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.

[0393] 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.

[0394] 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.

[0395] 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.

[0396] 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.

[0397] 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."

[0398] The present invention provides a system for analyzing a wide range of data, allowing caregivers to grasp the cognitive state of dementia patients in real time and take appropriate measures. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[0399] First, the server uses a data collection means to collect vital data, weather data, dietary image data, and voice data of the patient, thereby enabling comprehensive collection of a wide range of data related to the patient's daily life.

[0400] The server then uses preprocessing means to prepare the collected data in a format suitable for analysis, including converting heart rate data into time series data, food images into a data format for image recognition, and voice data into text.

[0401] The server then uses a data analysis means to input the pre-processed data into an AI model to analyze the patient's cognitive status, which is pre-trained to infer the cognitive status from the data.

[0402] Based on the analysis results, the server uses a result generation means to evaluate the patient's cognitive state and generate appropriate countermeasures. For example, if the patient's heart rate is high and stress is suspected, the server will suggest a countermeasure such as "recommending a walk to relax."

[0403] Next, the terminal uses a notification method to display the analysis results and countermeasures received from the server on the caregiver's device. For example, a message such as "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. A 15-minute light walk is recommended" may be displayed on a smartphone or tablet.

[0404] The user (caregiver) then implements the notified measures. For example, the caregiver may take a short walk with Patient A to reduce the patient's stress.

[0405] The server also uses a feedback collection method to collect the results of the implemented measures. For example, by sending the patient's heart rate after a walk or changes in facial expression back to the system, data can be provided to improve the analysis accuracy of the AI ​​model.

[0406] In this way, the system analyzes the patient's cognitive status in real time and promptly provides appropriate treatment to caregivers, thereby reducing the burden on caregivers and ensuring high-quality care.In addition, a feedback system continuously improves the accuracy of the analysis of the entire system.

[0407] Specific examples are given below.

[0408] Example 1:

[0409] At 7:00 AM, the server collects data on patient A's heart rate, blood pressure, sleep duration (from the smartwatch), current temperature and humidity (from the weather API), and an image of the patient's breakfast (from the smartphone).

[0410] Example 2:

[0411] The server inputs this data into the AI ​​model, and the AI ​​suggests that Patient A's heart rate is higher than normal and that he is not sleeping well at night, which may indicate stress or anxiety.

[0412] Example 3:

[0413] The device displays a message on the caregiver's smartphone such as, "Patient A's cognitive state has deteriorated slightly. This is thought to be due to stress. We recommend a 15-minute light walk."

[0414] Example 4:

[0415] The user (caregiver) checks the notification and tries to reduce stress by taking a walk around the neighborhood with Patient A.

[0416] Example 5:

[0417] The server again collects data on heart rate and changes in the patient's facial expression after the walk, and uses this data to retrain the AI ​​model.

[0418] Through this series of processes, the system constantly assesses the patient's cognitive status based on the latest information and provides appropriate advice to caregivers, thereby achieving higher quality care and reducing the burden on caregivers.

[0419] The processing flow will be explained below.

[0420] Step 1:

[0421] The server collects the patient's vital signs, weather data, meal image data, and audio data via a smartwatch, weather API, smartphone camera, audio recording device, etc. Specifically, at 7:00 AM, the server obtains Patient A's heart rate, blood pressure, sleep data, current temperature and humidity, and an image of his breakfast.

[0422] Step 2:

[0423] The server preprocesses the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure data into time-series data, converting food images into a format suitable for image recognition, and converting voice data into text. Specifically, the server organizes the heart rate data and checks for any abnormal values.

[0424] Step 3:

[0425] The server then inputs the preprocessed data into an AI model for analysis. The AI ​​model then analyzes the current data based on the previously learned data and predicts the patient's cognitive state. For example, if the patient's heart rate is higher than normal and they are not getting enough sleep, the AI ​​model will generate a result suggesting stress or anxiety.

[0426] Step 4:

[0427] The server evaluates the patient's cognitive state based on the analysis results obtained from the AI ​​model and generates appropriate treatment methods. For example, if stress is determined to be the cause, it generates specific advice such as "recommending a walk to relax."

[0428] Step 5:

[0429] The server sends the generated results and recommended actions to the caregiver's device. The device (the caregiver's smartphone or tablet) displays the results in real time. For example, a notification message might read, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0430] Step 6:

[0431] The user (caregiver) checks the notification and takes the recommended action, for example, taking a walk around the neighborhood with Patient A.

[0432] Step 7:

[0433] The user (caregiver) records the results of the measures taken and provides feedback to the server. Specifically, the user records the patient's heart rate and changes in facial expression after the walk using a smartphone and sends the results to the server.

[0434] Step 8:

[0435] The server retrains the AI ​​model based on the collected feedback data to improve future analysis accuracy, for example, by using the feedback data to re-learn the relationship between stress levels and heart rate.

[0436] Through this series of processes, the system constantly assesses the patient's cognitive status based on the latest information and provides appropriate advice to caregivers, thereby achieving higher quality care and reducing the burden on caregivers.

[0437] Example 1

[0438] 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."

[0439] In conventional dementia care, it has been difficult for caregivers to grasp the patient's cognitive state in real time and provide appropriate treatment promptly. Furthermore, there has been a lack of mechanisms for appropriately collecting feedback and utilizing it to improve the system's analysis accuracy. This has increased the burden on caregivers, making it difficult to provide high-quality care.

[0440] 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.

[0441] In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means. This allows the server to collect and analyze the patient's vital data, environmental data, nutritional data, and voice data in real time. The server also notifies the user device of the generated evaluation results and countermeasures, and recollects data after the countermeasures are implemented to improve the analytical accuracy of the AI ​​model. This reduces the burden on caregivers and enables the provision of high-quality care.

[0442] "Data collection tools" are devices and techniques for collecting patient vital, environmental, nutritional, and audio data.

[0443] "Preprocessing means" are devices and techniques used to convert collected data into a form suitable for analysis.

[0444] "Data analysis means" refers to the equipment and technology used to input pre-processed data into an AI model for analysis.

[0445] The "result generation means" refers to devices and technologies for evaluating the cognitive state of a patient based on the results of data analysis and generating appropriate treatment methods.

[0446] The "notification means" refers to the device and technology for notifying the generated evaluation results and countermeasures to the user device.

[0447] The "feedback collection means" refers to devices and techniques for collecting data based on the results of measures taken by users and improving the accuracy of the system's analysis.

[0448] "Vital data" refers to a patient's biological information such as heart rate, blood pressure, and sleep time.

[0449] "Environmental data" refers to data related to the external environment, such as temperature and humidity.

[0450] "Nutrition data" refers to data on the types, amounts, and nutrients of food analyzed from dietary image data.

[0451] "Audio data" refers to data recorded from patients' everyday conversations and voice information.

[0452] An "AI model" is a pre-trained artificial intelligence system that includes algorithms for inferring a patient's cognitive state from input data.

[0453] "User device" refers to an electronic device, such as a smartphone or tablet, used by a caregiver.

[0454] The present invention provides a system for analyzing a wide range of data, allowing caregivers to grasp the cognitive state of dementia patients in real time and take appropriate measures. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[0455] Data collection

[0456] The server uses the data collection means to collect the following data:

[0457] Vital data: Collect the patient's heart rate, blood pressure, and sleep duration through the smartwatch.

[0458] Environmental data: Use the weather API to obtain temperature and humidity.

[0459] Nutritional data: Photos of meals are taken from the patient's smartphone and converted into data for nutritional analysis.

[0460] Voice data: The patient's everyday conversations and voices are recorded using a microphone and sent to the server.

[0461] Examples:

[0462] At 7:00 AM, the server accesses Patient A's smartwatch and captures the patient's heart rate and blood pressure data for the day.

[0463] Get the day's temperature and humidity data from the weather API.

[0464] Images of the day's breakfast are automatically uploaded from the patient's smartphone.

[0465] Data Preprocessing

[0466] The server uses pre-processing means to prepare the collected data in a suitable format for analysis.

[0467] Convert heart rate data: Format heart rate data into time series data.

[0468] Nutrition data conversion: Convert images into a format that is easy for AI to interpret.

[0469] Voice to Text Conversion: We use voice recognition software (e.g., Google Speech-to-Text) to convert voice data into text.

[0470] Examples:

[0471] The server takes the heart rate data and converts it into time series data.

[0472] Image recognition software is used to identify the types and quantities of food in the breakfast image and prepare it in a format suitable for analysis.

[0473] Everyday conversations received via voice are converted into text and prepared for sentiment analysis.

[0474] Data analysis and result generation

[0475] The server uses a data analysis means to input the pre-processed data into an AI model to analyze the patient's cognitive status, which is pre-trained to infer the cognitive status from the data.

[0476] Based on the analysis results, the server uses the result generation means to assess the patient's cognitive state and generate an appropriate course of action.

[0477] Examples:

[0478] The pre-processed heart rate data is fed into an AI model to detect values ​​outside the normal range.

[0479] Image data of breakfast is input into an AI model and the nutritional balance is analyzed.

[0480] The text data obtained by speech recognition is input into a sentiment analysis model to detect changes in emotion.

[0481] notification

[0482] The terminal uses a notification means to display the analysis results and countermeasures received from the server on the user's (caregiver's) device.

[0483] Notification method: A message will appear on your smartphone or tablet.

[0484] Notification content: Includes analysis results and details of how to deal with the issue.

[0485] Examples:

[0486] The caregiver's smartphone will receive a notification saying, "Patient A's cognitive state has deteriorated slightly. This is thought to be due to stress. We recommend a 15-minute light walk."

[0487] Implementing countermeasures

[0488] The user (caregiver) takes the notified action.

[0489] Action to be taken: Take the notified action.

[0490] Specific action: For example, a caregiver takes a walk around the neighborhood with patient A.

[0491] Examples:

[0492] The caregiver asks Patient A, "Let's take a 15-minute walk together," and they actually go for a walk.

[0493] Feedback collection

[0494] The server uses a feedback collection means to collect the results of the actions taken.

[0495] Feedback items: Heart rate after a walk, changes in the patient's facial expression, daily reactions, etc.

[0496] Data reuse: Collected data can be used to retrain AI models to improve analysis accuracy.

[0497] Examples:

[0498] After the walk, heart rate data is collected again from the smartwatch to assess the patient's stress level.

[0499] The patient's facial expressions and tone of voice are analyzed from camera and audio data and collected as feedback.

[0500] Example prompts to input to the generative AI model

[0501] "Patient A has a higher than normal heart rate and is not sleeping well at night. He is showing signs of stress and anxiety. Please suggest ways to address this."

[0502] "Based on Patient B's vital signs and dietary information today, please assess his cognitive status and suggest appropriate treatment options."

[0503] Through this series of processes, the system analyzes the patient's cognitive state in real time and promptly provides appropriate measures to caregivers, thereby reducing their burden and ensuring high-quality care. In addition, a feedback system continuously improves the analysis accuracy of the entire system.

[0504] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0505] Step 1:

[0506] Data collection

[0507] The server collects the following data using the data collection means:

[0508] Input: Patient's vital data (heart rate, blood pressure, sleep time), environmental data (temperature, humidity), nutritional data (images of meals), and audio data (daily conversation).

[0509] Specific operation: The server retrieves heart rate, blood pressure, and sleep time data from Patient A's smartwatch at 7:00 AM. It also retrieves the day's temperature and humidity using a weather API. It also uploads an image of breakfast from the patient's smartphone and records daily conversations using the microphone.

[0510] Output: Collected vital, environmental, nutritional and audio data.

[0511] Step 2:

[0512] Data Preprocessing

[0513] The server uses pre-processing means to prepare the collected data in a suitable format for analysis.

[0514] Input: Collected vital, environmental, nutritional, and audio data.

[0515] How it works: The server formats the heart rate data into time series data, converts the image of the breakfast into a format that is easy for AI to understand, and converts the audio data into text using speech recognition software.

[0516] Output: Preprocessed heart rate data, image data, and audio data converted to text.

[0517] Step 3:

[0518] Data analysis

[0519] The server uses data analysis means to input the preprocessed data into an AI model to analyze the patient's cognitive status.

[0520] Input: Preprocessed heart rate data, environmental data, nutrition data, and audio data.

[0521] Specific operations: Preprocessed heart rate data is input into an AI model to detect abnormal values. Image data of breakfast is input into an AI model to analyze nutritional balance. Text-converted voice data is input into an emotion analysis model to analyze changes in emotion.

[0522] Output: Analysis of the patient's cognitive status.

[0523] Step 4:

[0524] result generation

[0525] The server uses the result generation means to evaluate the cognitive state of the patient based on the analysis results and generates an appropriate course of action.

[0526] Input: Analysis results.

[0527] Specific behavior: Integrates the results of multiple analyses to assess the patient's cognitive state (e.g., stress level, lack of sleep, etc.), and uses a generative AI model to generate appropriate countermeasures (e.g., recommending a relaxing walk).

[0528] Output: Evaluation results and suggested solutions.

[0529] Step 5:

[0530] notification

[0531] The terminal uses the notification means to display the analysis results and countermeasures received from the server on the user device.

[0532] Input: Evaluation results and suggested actions.

[0533] What it does: Sends a detailed message to the caregiver's smartphone about the patient's cognitive status and how to address it. The notification might say, "Patient A's cognitive status has slightly deteriorated. This is thought to be due to stress. We recommend a 15-minute brisk walk."

[0534] Output: The notification message displayed on the user device.

[0535] Step 6:

[0536] Implementing countermeasures

[0537] The user (caregiver) takes the notified action.

[0538] Input: Notified suggested action.

[0539] Specific actions: The caregiver takes a walk around the neighborhood with Patient A. They say, "Let's take a 15-minute walk together," and then actually go for a walk.

[0540] Output: The outcome of the intervention (e.g., the patient's condition after the walk).

[0541] Step 7:

[0542] Feedback collection

[0543] The server collects the results of the actions taken using a feedback collection means.

[0544] Input: Results of the implemented measures (heart rate data, facial expression changes, etc.).

[0545] Specific operations: After the walk, the smartwatch collects heart rate data again to assess the patient's stress level, and analyzes the patient's facial expressions and tone of voice via camera and audio data to provide feedback.

[0546] Output: Collected feedback data, analysis results of improved AI model.

[0547] (Application example 1)

[0548] 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."

[0549] Until now, there has been no system that monitors a driver's fatigue and stress state in real time and provides appropriate countermeasures based on that information. This has made it difficult to improve driver safety and performance, potentially increasing the risk of traffic accidents. Therefore, the present invention aims to improve driver safety and comfort by analyzing the driver's vital signs, environmental data, and image data in real time and providing appropriate notifications and feedback according to the driver's condition.

[0550] 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.

[0551] In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, a means for collecting vital data, environmental data, and image data of the driver, a means for analyzing the data and generating recommendations based on the driver's condition, and a means for notifying the driver's device of the recommendations. This makes it possible to monitor the driver's fatigue and stress in real time and quickly provide appropriate measures.

[0552] The "data collection means" is a means for acquiring various data including vital data of the driver, environmental data, and image data.

[0553] "Preprocessing means" refers to means for converting collected data into a format that is easy to analyze.

[0554] "Data analysis means" means a means for using pre-processed data to assess the driver's condition and analyzing it using specific algorithms or generative AI models.

[0555] The "result generation means" is a means for generating specific recommendations or notifications for the driver based on the results of the data analysis.

[0556] "Notification means" refers to a means for transmitting generated recommendations and notifications to the driver's device.

[0557] The "feedback collection means" is a means for recording the driver's response to the provided countermeasures and the progress of the improvement, and for use in reanalysis and model improvement.

[0558] "Vital data" refers to biometric information such as the driver's heart rate, blood pressure, and sleep data.

[0559] "Environmental data" refers to data relating to climatic conditions and the surrounding environment, including, for example, temperature, humidity, and weather information.

[0560] "Image data" refers to images of the driver's face and footage of the inside of the vehicle, and is data used to analyze the driver's facial expression and posture.

[0561] A "generative AI model" is a pre-trained artificial intelligence model that infers the driver's state based on collected data and generates appropriate recommendations.

[0562] The present invention provides a system for analyzing a driver's vital data, environmental data, and image data in real time and providing appropriate countermeasures based on the driver's condition. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[0563] The server uses data collection means to collect driver vital data, weather data, and driver image data. This data collection is done through wearable devices such as smartwatches and fitness trackers, weather APIs, and in-car cameras. For example, heart rate, blood pressure, and sleep data are acquired.

[0564] The collected data is preprocessed by the server. The preprocessing means standardizes vital data such as heart rate and blood pressure, converts image data to grayscale, and resizes it to an appropriate size. Weather data is obtained using an API and converted into a format suitable for analysis.

[0565] The server then inputs the preprocessed data into a generative AI model for data analysis. The generative AI model is pre-trained to infer the driver's state and detect conditions such as fatigue, stress, and distraction. Based on the results of this analysis, a specific recommendation is generated for the driver. For example, a message such as "You need a break. Please pull over and take a break" is generated.

[0566] The generated results are sent to the driver's device, such as a smartphone or an in-vehicle display, using a notification means, allowing the driver to immediately check their own condition and take any necessary measures.

[0567] After the driver actually takes the recommended action, the server uses a feedback collection method to collect the driver's reaction data again. For example, heart rate and facial expression changes after a break are collected and used to retrain the AI ​​model. This feedback function further improves the system's analysis accuracy.

[0568] As a concrete example, the following scenario can be considered.

[0569] Data collected:

[0570] Heart rate: 85 bpm

[0571] Blood pressure: 130 / 85 mmHg

[0572] Sleep data: 6 hours of sleep

[0573] Weather data: 22°C, 65% humidity

[0574] Driver's facial image

[0575] Generated recommendations:

[0576] The driver is likely fatigued, so a notification is generated saying, "You need a break. Please stop and take a break."

[0577] An example of a prompt is as follows:

[0578] "Based on the collected data, a generative AI model will be used to analyze the driver's condition and generate a notification that a break is needed, if necessary.

[0579] In this way, the present invention monitors the driver's health condition in real time and helps prevent traffic accidents and health risks.

[0580] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0581] Step 1:

[0582] The server uses data collection means to collect the driver's vital signs, weather data, and image data of the driver. Specifically, it obtains heart rate, blood pressure, and sleep data from the wearable device, and obtains current weather data (temperature, humidity, etc.) using a weather API. It also captures images of the driver's face using the in-car camera. This allows it to collect a wide range of data on the driver's condition and environment.

[0583] Input: Vital data from wearable devices, weather data from weather API, image data of the driver from an in-car camera

[0584] Output: Raw data collected

[0585] Step 2:

[0586] The server uses preprocessing tools to convert the collected data into a format suitable for analysis. Heart rate and blood pressure are normalized to a standard scale, sleep data is converted into an index to evaluate sleep quality, weather data is formatted from the API, and image data is converted to grayscale and resized to a size suitable for analysis. This prepares the data in a unified format.

[0587] Input: Raw data collected

[0588] Output: Preprocessed data

[0589] Step 3:

[0590] The server uses data analysis means to input preprocessed data into the generative AI model, which analyzes the driver's condition. The generative AI model combines and analyzes heart rate, blood pressure, sleep data, weather data, and image data to estimate the driver's fatigue level, stress level, distraction state, etc. The AI ​​model is pre-trained and uses this data to predict the driver's condition with high accuracy.

[0591] Input: Preprocessed data

[0592] Output: Analysis results (driver's current status evaluation)

[0593] Step 4:

[0594] The server generates specific recommendations for the driver based on the analysis results using the result generation means. For example, a message such as "The driver is highly fatigued, so a 15-minute break is recommended." This recommendation serves as a specific guideline for action to ensure the driver's safety.

[0595] Input: Analysis results (driver's current status evaluation)

[0596] Output: Recommendations

[0597] Step 5:

[0598] The terminal uses notification methods to notify the driver of the generated recommendations. A message is displayed on the driver's smartphone or in-vehicle display, allowing the driver to immediately check the contents. This allows the driver to recognize their own condition and take any necessary measures promptly.

[0599] Input: Recommendations

[0600] Output: Notification to the driver

[0601] Step 6:

[0602] The user (driver) takes the appropriate action. For example, after checking the notification on the device, the user stops the car in a safe place and takes a break for the specified amount of time. This is expected to reduce the driver's fatigue and stress.

[0603] Input: Notification content

[0604] Output: Coping behavior

[0605] Step 7:

[0606] The server collects the results of the implemented measures using a feedback collection method. Feedback data such as heart rate, blood pressure, and changes in the driver's facial expressions after the break are collected again and used as training data for the generative AI model. This feedback improves the accuracy of the system.

[0607] Input: New vital data and image data after response actions

[0608] Output: Feedback data

[0609] Through the above series of processing steps, the system of the present invention can monitor the driver's health condition in real time and provide appropriate countermeasures promptly, thereby reducing the risk of traffic accidents and improving the safety and comfort of the driver.

[0610] 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.

[0611] The present invention is a system for supporting care of dementia patients, and includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion engine. The system aims to analyze a wide range of data, grasp the cognitive state of the patient in real time, and provide appropriate measures to caregivers.

[0612] First, the server collects vital data (e.g., heart rate, blood pressure), weather data (e.g., temperature, humidity), meal image data, and voice data through a data collection means. It also uses an emotion engine to recognize emotions from the voice data and facial expression data of users (caregivers and patients).

[0613] The server then uses preprocessing means to convert the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure into time-series data, converting food images into a format suitable for image recognition, and converting voice data into text. Furthermore, an emotion engine extracts emotions from the user's voice and facial expressions and records them as data.

[0614] The server then uses data analysis tools to input the preprocessed data into the AI ​​model to analyze the patient's cognitive state. The AI ​​model uses past learning data and analyzes current data to infer cognitive state and stress level. By including emotional data obtained from the emotion engine in this analysis, more detailed insights into the mental state of the patient and caregiver can be gained.

[0615] Based on the analysis results, the server uses the result generation means to determine an appropriate course of action. Specifically, if it determines that the cognitive state is slightly deteriorating, it generates a course of action to reduce stress, such as "recommending a walk to relax." It can also suggest a course of action that takes emotions into account by referring to the data obtained through the emotion engine.

[0616] The server then sends the generated results and advice on how to respond to the situation to the caregiver's device. The device (the caregiver's smartphone or tablet) displays the received notification in real time. An example message might read, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0617] The user (caregiver) then checks the notification received from the device and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the emotion engine also monitors changes in the caregiver's and patient's emotions and collects further data.

[0618] Finally, the user (caregiver) sends the results of the measures they took as feedback to the server. The server analyzes the collected feedback data and uses it to retrain the AI ​​model to improve its accuracy. In particular, since this feedback includes emotional data, it can make the next recommendation more accurate.

[0619] As a specific example, at 7:00 AM, the server collects Patient A's heart rate, sleep data, current temperature and humidity, and an image of their breakfast. At the same time, the emotion engine analyzes the caregiver's morning voice tone and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that Patient A's heart rate is higher than normal and that they are not getting enough sleep, suggesting stress. The result generation method generates a solution such as, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. We recommend a 15-minute light walk." The device then sends a notification to the caregiver, who then follows the recommendation and takes the walk, providing feedback on the results. This feedback data is collected and analyzed by the server to help further improve the system.

[0620] In this way, the system analyzes data from multiple perspectives and provides appropriate measures that take emotional data into consideration, thereby providing effective support for both caregivers and patients.

[0621] The processing flow will be explained below.

[0622] Step 1:

[0623] The server uses data collection tools to collect the patient's vital signs, weather data, meal image data, voice data, and user emotion data. For example, it obtains heart rate and blood pressure from a smartwatch and current temperature and humidity from a weather API. Meanwhile, the caregiver's smartphone camera takes pictures of the patient's breakfast and records the conversation using a voice recording device.

[0624] Step 2:

[0625] The server uses an emotion engine to analyze emotions from the voice and facial expression data of the caregiver and patient. For example, it analyzes emotions from voice and recognizes emotions such as stress and anxiety from the tone of the voice. It also analyzes the patient's facial expression from captured images to extract emotions such as smile and tension.

[0626] Step 3:

[0627] The server uses preprocessing means to convert the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure data into time-series data, converts food image data into a format suitable for image recognition, converts voice data into text, and quantifies and stores emotion data.

[0628] Step 4:

[0629] The server uses data analysis tools to input preprocessed data into the AI ​​model, which analyzes the patient's cognitive state. The AI ​​model suggests possible stress based on elevated heart rate and lack of sleep, and provides a complementary assessment of the degree of stress and anxiety based on the results of emotion analysis.

[0630] Step 5:

[0631] The server uses the result generation means to evaluate the patient's current cognitive state based on the analysis results obtained from the AI ​​model and generate appropriate measures to address the situation. For example, it generates specific advice such as, "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. We recommend a 15-minute light walk."

[0632] Step 6:

[0633] The server sends the generated results and recommended actions to the caregiver's device via a notification method. The device (caregiver's smartphone or tablet) displays this notification in real time. The caregiver receives a message saying, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0634] Step 7:

[0635] The user (caregiver) checks the notification and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the caregiver carefully observes the patient's reactions and records any changes in their emotions.

[0636] Step 8:

[0637] The user (caregiver) sends the results of the measures taken to the server through a feedback collection means. Specifically, the user records the patient's heart rate after the walk, changes in the patient's facial expression, and changes in the caregiver's own emotions using a smartphone and sends them to the server.

[0638] Step 9:

[0639] The server retrains the AI ​​model based on the data obtained through the feedback collection method to improve the accuracy of the next analysis. Based on the feedback data, the relationship between stress state and changes in heart rate and emotions is reevaluated and reflected in future countermeasures.

[0640] Through the above processing flow, the system can analyze the patient's cognitive state and emotions from multiple angles and provide caregivers with appropriate coping strategies in real time, thereby improving the quality of care and reducing the burden on caregivers.

[0641] Example 2

[0642] 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."

[0643] In providing care support for dementia patients, it is extremely important to understand the patient's cognitive state and emotional changes in real time and provide caregivers with appropriate measures. However, current systems have fragmented data collection and analysis, making it difficult to comprehensively understand the patient's overall situation. In addition, measures that take emotional data into consideration are insufficient, which hinders caregivers from providing optimal support.

[0644] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion recognition means. This makes it possible to collect biological data, environmental data, image data, and voice data of a patient from various sources, and through preprocessing and analysis, grasp the patient's cognitive state and emotions in real time and provide appropriate measures promptly.

[0645] The "data collection means" is a means for collecting various data such as patient biometric data, environmental data, image data, and voice data.

[0646] "Preprocessing means" refers to the process of converting collected data into a format suitable for analysis. Specifically, this involves generating time-series data, converting it into an image recognition format, and converting voice data into text.

[0647] The "data analysis means" is a means for inputting preprocessed data into the AI ​​model and analyzing it, which allows for the prediction of the patient's cognitive state and stress level.

[0648] The "result generation means" is a means for determining and generating appropriate measures based on the results of data analysis. Specifically, it suggests relaxation techniques, etc.

[0649] The "notification means" is a means for transmitting the generated countermeasure to the caregiver's device and displaying the content thereof.

[0650] The "feedback collection means" is a means for collecting the results of the measures taken by the caregiver and using them for subsequent analysis and to improve the accuracy of the AI ​​model.

[0651] "Emotion recognition means" is a means of extracting emotions from voice data and facial expression data and recording them as data. This data plays an important role in data analysis.

[0652] An "AI model" is an artificial intelligence model that analyzes collected data and generates recommendations, enabling highly accurate analysis using past data.

[0653] "Vital data" refers to biological information such as heart rate, blood pressure, and body temperature.

[0654] "Image data" refers to image information captured with a camera of food, facial expressions, etc.

[0655] "Audio data" refers to data that indicates acoustic information that records user speech and environmental sounds.

[0656] "Environmental data" refers to data relating to the external environment, such as temperature, humidity, and weather.

[0657] The present invention is a system for supporting care of dementia patients, which includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion recognition means. The system aims to analyze a wide range of data, grasp the cognitive state of the patient in real time, and provide appropriate measures to caregivers.

[0658] First, the server collects vital data (e.g., heart rate, blood pressure, body temperature, etc.), environmental data (e.g., temperature, humidity, etc.), image data (photos of meals and facial expressions), and voice data through a data collection means. Vital data is obtained from a wearable device (e.g., Fitbit or Apple Watch). Environmental data is acquired from a weather API (e.g., OpenWeatherMap API). Image data is collected through the caregiver's smartphone camera, and voice data is obtained through a dedicated app. In addition, an emotion recognition means is used to recognize emotions from the user's voice data and facial expression data.

[0659] Next, the server uses preprocessing means to convert and organize the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure data into time-series data, converts food images into a format suitable for image recognition (e.g., RGB pixel array), and converts audio data into text using speech recognition software (e.g., Google Cloud Speech-to-Text API). Furthermore, it uses emotion recognition means to extract emotions from voice and facial expressions and record them as data.

[0660] The server then uses data analysis tools to input the preprocessed data into an AI model (e.g., a model implemented in TensorFlow or PyTorch) to analyze the patient's cognitive state. The analysis involves analyzing past training data and using current data to estimate cognitive state and stress levels. Emotion data obtained from emotion recognition tools can also be incorporated into this analysis to provide detailed insights into the mental state of the patient and caregiver.

[0661] Based on the analysis results, the server uses a result generation method to determine an appropriate course of action. For example, if it determines that the cognitive state is slightly deteriorating, it will generate a suggestion such as "recommended walk to relax" as a measure to reduce stress. It is also possible to refer to emotional data and propose a course of action that takes emotions into consideration.

[0662] The server then sends the generated results and advice on how to respond to the situation to the caregiver's device, which then displays the received notification in real time. For example, a message might appear saying, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0663] The user (caregiver) then checks the notification received from the device and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the emotion recognition means monitors changes in the emotions of the caregiver and the patient and collects further data.

[0664] Finally, the user (caregiver) sends the results of the measures they took as feedback to the server. The server analyzes the collected feedback data and uses it to retrain the AI ​​model to improve its accuracy. In particular, since this feedback includes emotional data, it can make the next recommendation more accurate.

[0665] Examples:

[0666] At 7:00 AM, the server collects Patient A's heart rate, sleep data, current temperature and humidity, and an image of their breakfast. At the same time, the emotion recognition means analyzes the caregiver's morning tone of voice and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that Patient A's heart rate is higher than normal and that they are not getting enough sleep, suggesting stress. The result generation means generates a countermeasure such as, "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. A 15-minute light walk is recommended." The device then sends a notification to the caregiver, who then follows the recommendation and takes the walk, providing feedback on the results. This feedback data is collected and analyzed by the server to help further improve the system.

[0667] In this way, the system analyzes data from multiple perspectives and provides appropriate measures that take emotional data into consideration, thereby providing effective support for both caregivers and patients.

[0668] Example prompt sentence:

[0669] "Patient A's cognitive status has deteriorated slightly. This is likely due to stress. A 15-minute brisk walk is recommended."

[0670] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0671] Step 1: Data collection

[0672] The server collects various data through the data collection means.

[0673] Input: Patient vital data (heart rate, blood pressure, body temperature, etc.), environmental data (temperature, humidity, etc.), image data (photos of meals, images of facial expressions), voice data

[0674] Processing: Vital data is collected from wearable devices (e.g., Fitbit or Apple Watch). Environmental data is obtained from weather APIs (e.g., OpenWeatherMap API). Image data is collected through the caregiver's smartphone camera, and audio data is obtained through a dedicated app.

[0675] Output: A wide variety of raw data collected

[0676] Step 2: Data Preprocessing

[0677] The server uses pre-processing means to convert the collected data into a form suitable for analysis.

[0678] Input: Raw data collected

[0679] Processing: Heart rate and blood pressure data are converted into time-series data, and food images are converted into a format for image recognition (e.g., RGB pixel arrays). Audio data is converted into text using speech recognition software (e.g., Google Cloud Speech-to-Text API). Emotion recognition tools are used to extract emotions from voice and facial expressions and record them as data.

[0680] Output: Preprocessed datasets (time series data, formatted image data, text data, emotion data)

[0681] Step 3: Data analysis

[0682] The server uses data analysis means to input the preprocessed data into the AI ​​model and analyze the patient's cognitive status.

[0683] Input: Preprocessed datasets (time series data, formatted image data, text data, emotion data)

[0684] Processing: Data is fed into an AI model (e.g., a model implemented in TensorFlow or PyTorch) and current data is analyzed based on past training data. This allows for estimation of cognitive state and stress level. Emotion data obtained through emotion recognition methods is also included in the analysis.

[0685] Output: Analysis results (predictions of cognitive state, stress level, etc.)

[0686] Step 4: Result generation

[0687] The server determines and generates an appropriate countermeasure based on the analysis result using the result generation means.

[0688] Input: Analysis results (predicted results such as cognitive state, stress level, etc.)

[0689] Processing: If the cognitive state is deteriorating, it will determine measures to reduce stress, such as suggesting a relaxing walk or a relaxation session. It will also take into account emotional data to generate coping strategies.

[0690] Output: A message with the recommendations

[0691] Step 5: Notification Delivery

[0692] The server transmits the generated results and actions to the caregiver's device.

[0693] Input: Message with recommendations

[0694] Processing: The generated message is sent to the caregiver's device (smartphone or tablet).

[0695] Output: Notification message displayed on the caregiver's device

[0696] Step 6: Implement the recommendations

[0697] The user (caregiver) checks the notification and takes the recommended action.

[0698] Input: Notification message

[0699] Action: Review the notification and take the recommended action with the patient, for example, take a walk with the patient.

[0700] Output: Action taken

[0701] Step 7: Gather feedback

[0702] The user (caregiver) sends the results of the execution to the server as feedback.

[0703] Input: Execution results (e.g., changes in the patient's condition or emotions after the walk)

[0704] Processing: The results of the countermeasures are recorded via the device and sent to the server, which collects and analyzes the feedback data and retrains the AI ​​model.

[0705] Output: Feedback data is stored and used on the server

[0706] Throughout this entire process, the system analyzes the data from multiple angles and provides appropriate responses that take emotional data into account, thereby providing effective support for both caregivers and patients.

[0707] (Application example 2)

[0708] 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."

[0709] In factories and other workplaces, it is necessary to grasp the physical and mental health status of employees in real time and quickly provide appropriate countermeasures. However, currently there is a lack of systems that can accurately analyze workload and stress and suggest appropriate countermeasures. This presents a problem in that it is not possible to provide appropriate support to ensure employee efficiency and safety.

[0710] 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 a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion analysis means. This makes it possible to analyze the physical and mental health conditions of workers in real time and quickly provide appropriate countermeasures.

[0711] The "data collection means" is a system element that collects vital data, work environment data, work image data, and voice data of workers.

[0712] A "preprocessing means" is an element of the system that converts and prepares collected data into a form suitable for analysis.

[0713] "Data analysis means" refers to the system element that inputs preprocessed data, including emotional data, into the AI ​​model for analysis.

[0714] A "result generation means" is an element of the system that determines the appropriate course of action based on the results of data analysis.

[0715] The "notification means" is a system element that sends the generated results and how to deal with them to the worker's device.

[0716] The "feedback collection means" is a system element that sends the results of the implemented countermeasures to the server and collects them.

[0717] "Emotion analysis means" is an element of the system that recognizes and analyzes emotions from the worker's facial expressions and voice data.

[0718] The present invention provides a system for monitoring and analyzing the health status and stress levels of employees in the workplace in real time, which includes a data collection means, a pre-processing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and a sentiment analysis means.

[0719] First, the server collects the worker's vital data (e.g., heart rate, blood pressure), work environment data (e.g., temperature, humidity), work image data, and voice data through a data collection means. It also uses an emotion analysis means to recognize the worker's emotions from their facial expressions and voice data.

[0720] Next, the server uses preprocessing means to convert the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure into time-series data, converting work images into a format suitable for image recognition, and converting voice data into text. Furthermore, emotion analysis means extracts emotions from the worker's voice and facial expressions and records them as data.

[0721] The server then uses data analysis means to input the preprocessed data into the AI ​​model, which analyzes the worker's health condition and stress level. The AI ​​model uses past learning data and analyzes current data to estimate stress levels and fatigue levels. By including emotional data obtained by the emotion analysis means in this analysis, detailed insights into the worker's mental state can be gained.

[0722] Based on the analysis results, the server uses the result generation means to determine an appropriate response. Specifically, if the stress level is determined to be high, the server generates a response such as recommending a break. Using data obtained through the emotion analysis means, the server can also suggest a response that takes emotions into consideration.

[0723] Next, the server sends the generated results and countermeasures to the worker's device (terminal). The terminal displays the received notification in real time. An example message might be, "Worker A's stress level is high, so we recommend that he take a 10-minute break."

[0724] The user (worker) then checks the notification received from the device and takes the recommended action. For example, the worker takes a 10-minute break in the break room. During this action, the emotion analysis means also monitors changes in the worker's emotions and collects further data.

[0725] Finally, the user (worker) sends the results of the implemented measures to the server as feedback. The server analyzes the collected feedback data and uses it for retraining to improve the accuracy of the AI ​​model. In particular, since this feedback includes emotional data, it will make the next recommendation more accurate.

[0726] As a specific example, at 7:00 AM, the server collects worker A's heart rate, sleep data, and current work environment data, while the emotion analysis means simultaneously analyzes the worker's morning tone of voice and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that worker A's heart rate is higher than normal and that he or she is highly fatigued. The result generation means generates a countermeasure such as, "Worker A's stress level is high, so we recommend that he or she take a 10-minute break." The device then sends a notification to the worker, and the user (worker) follows the recommendation and takes a break, providing feedback on the results. This feedback data is collected and analyzed by the server and used to further improve the system.

[0727] This system analyzes data from multiple angles and provides appropriate countermeasures that take emotional data into consideration, thereby effectively supporting both worker health and work efficiency.

[0728] Example prompt sentence:

[0729] Please explain the system that collects and analyzes the heart rate, blood pressure, facial expressions and voice data of factory workers, and recommends appropriate breaks and ways to deal with stress when the worker's stress level is high.

[0730] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0731] Step 1:

[0732] The server uses data collection means to collect worker vital data (heart rate, blood pressure), work environment data (temperature, humidity), work image data, and voice data. In this step, the input is real-time data obtained from various sensors and cameras, and the output is raw data compiled by the data collection means.

[0733] Step 2:

[0734] The server uses preprocessing means to convert and organize the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure into time-series data, converts work images into a format for image recognition, and converts voice data into text. The input to this step is the raw data collected in step 1, and the output is preprocessed data.

[0735] Step 3:

[0736] The server uses emotion analysis means to recognize and extract emotions from the worker's facial and voice data. The emotion data is used for subsequent analysis. The input for this step is facial and voice data obtained from a camera and microphone, and the output is emotion data recognized by the emotion analysis means.

[0737] Step 4:

[0738] The server uses data analysis means to input the preprocessed data and emotional data into the AI ​​model to analyze the worker's health condition and stress level. The AI ​​model performs analysis based on past learning data and makes an evaluation based on current data. The inputs for this step are the preprocessed data and emotional data, and the output is the analysis results.

[0739] Step 5:

[0740] The server uses the result generation means to determine an appropriate course of action based on the analysis results. For example, if the stress level is determined to be high, it generates a course of action such as recommending a break. The input of this step is the analysis result of step 4, and the output is a course of action.

[0741] Step 6:

[0742] The server uses the notification means to send the generated results and the countermeasures to the worker's terminal. The terminal displays the received notification to the worker in real time. The input of this step is information about the countermeasures, and the output is the notification sent to the worker's terminal.

[0743] Step 7:

[0744] The user (worker) checks the notification received from the terminal and takes the recommended action. For example, the user takes a specific action such as taking a 10-minute rest in the break room. The input of this step is the notification from the terminal, and the output is the action taken.

[0745] Step 8:

[0746] The user (operator) sends the results of the countermeasures they have taken as feedback to the server. The server analyzes the collected feedback data and uses it for retraining to improve the accuracy of the AI ​​model. The input of this step is the feedback information from the user, and the output is the data used for retraining.

[0747] 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.

[0748] 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.

[0749] 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.

[0750] [Third embodiment]

[0751] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0752] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0753] 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).

[0754] 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.

[0755] 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.

[0756] 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).

[0757] 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.

[0758] 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.

[0759] 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.

[0760] 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.

[0761] 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.

[0762] 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."

[0763] The present invention provides a system for analyzing a wide range of data, allowing caregivers to grasp the cognitive state of dementia patients in real time and take appropriate measures. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[0764] First, the server uses a data collection means to collect vital data, weather data, dietary image data, and voice data of the patient, thereby enabling comprehensive collection of a wide range of data related to the patient's daily life.

[0765] The server then uses preprocessing means to prepare the collected data in a format suitable for analysis, including converting heart rate data into time series data, food images into a data format for image recognition, and voice data into text.

[0766] The server then uses a data analysis means to input the pre-processed data into an AI model to analyze the patient's cognitive status, which is pre-trained to infer the cognitive status from the data.

[0767] Based on the analysis results, the server uses a result generation means to evaluate the patient's cognitive state and generate appropriate countermeasures. For example, if the patient's heart rate is high and stress is suspected, the server will suggest a countermeasure such as "recommending a walk to relax."

[0768] Next, the terminal uses a notification method to display the analysis results and countermeasures received from the server on the caregiver's device. For example, a message such as "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. A 15-minute light walk is recommended" may be displayed on a smartphone or tablet.

[0769] The user (caregiver) then implements the notified measures. For example, the caregiver may take a short walk with Patient A to reduce the patient's stress.

[0770] The server also uses a feedback collection method to collect the results of the implemented measures. For example, by sending the patient's heart rate after a walk or changes in facial expression back to the system, data can be provided to improve the analysis accuracy of the AI ​​model.

[0771] In this way, the system analyzes the patient's cognitive status in real time and promptly provides appropriate treatment to caregivers, thereby reducing the burden on caregivers and ensuring high-quality care.In addition, a feedback system continuously improves the accuracy of the analysis of the entire system.

[0772] Specific examples are given below.

[0773] Example 1:

[0774] At 7:00 AM, the server collects data on patient A's heart rate, blood pressure, sleep duration (from the smartwatch), current temperature and humidity (from the weather API), and an image of the patient's breakfast (from the smartphone).

[0775] Example 2:

[0776] The server inputs this data into the AI ​​model, and the AI ​​suggests that Patient A's heart rate is higher than normal and that he is not sleeping well at night, which may indicate stress or anxiety.

[0777] Example 3:

[0778] The device displays a message on the caregiver's smartphone such as, "Patient A's cognitive state has deteriorated slightly. This is thought to be due to stress. We recommend a 15-minute light walk."

[0779] Example 4:

[0780] The user (caregiver) checks the notification and tries to reduce stress by taking a walk around the neighborhood with Patient A.

[0781] Example 5:

[0782] The server again collects data on heart rate and changes in the patient's facial expression after the walk, and uses this data to retrain the AI ​​model.

[0783] Through this series of processes, the system constantly assesses the patient's cognitive status based on the latest information and provides appropriate advice to caregivers, thereby achieving higher quality care and reducing the burden on caregivers.

[0784] The processing flow will be explained below.

[0785] Step 1:

[0786] The server collects the patient's vital signs, weather data, meal image data, and audio data via a smartwatch, weather API, smartphone camera, audio recording device, etc. Specifically, at 7:00 AM, the server obtains Patient A's heart rate, blood pressure, sleep data, current temperature and humidity, and an image of his breakfast.

[0787] Step 2:

[0788] The server preprocesses the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure data into time-series data, converting food images into a format suitable for image recognition, and converting voice data into text. Specifically, the server organizes the heart rate data and checks for any abnormal values.

[0789] Step 3:

[0790] The server then inputs the preprocessed data into an AI model for analysis. The AI ​​model then analyzes the current data based on the previously learned data and predicts the patient's cognitive state. For example, if the patient's heart rate is higher than normal and they are not getting enough sleep, the AI ​​model will generate a result suggesting stress or anxiety.

[0791] Step 4:

[0792] The server evaluates the patient's cognitive state based on the analysis results obtained from the AI ​​model and generates appropriate treatment methods. For example, if stress is determined to be the cause, it generates specific advice such as "recommending a walk to relax."

[0793] Step 5:

[0794] The server sends the generated results and recommended actions to the caregiver's device. The device (the caregiver's smartphone or tablet) displays the results in real time. For example, a notification message might read, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0795] Step 6:

[0796] The user (caregiver) checks the notification and takes the recommended action, for example, taking a walk around the neighborhood with Patient A.

[0797] Step 7:

[0798] The user (caregiver) records the results of the measures taken and provides feedback to the server. Specifically, the user records the patient's heart rate and changes in facial expression after the walk using a smartphone and sends the results to the server.

[0799] Step 8:

[0800] The server retrains the AI ​​model based on the collected feedback data to improve future analysis accuracy, for example, by using the feedback data to re-learn the relationship between stress levels and heart rate.

[0801] Through this series of processes, the system constantly assesses the patient's cognitive status based on the latest information and provides appropriate advice to caregivers, thereby achieving higher quality care and reducing the burden on caregivers.

[0802] Example 1

[0803] 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."

[0804] In conventional dementia care, it has been difficult for caregivers to grasp the patient's cognitive state in real time and provide appropriate treatment promptly. Furthermore, there has been a lack of mechanisms for appropriately collecting feedback and utilizing it to improve the system's analysis accuracy. This has increased the burden on caregivers, making it difficult to provide high-quality care.

[0805] 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.

[0806] In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means. This allows the server to collect and analyze the patient's vital data, environmental data, nutritional data, and voice data in real time. The server also notifies the user device of the generated evaluation results and countermeasures, and recollects data after the countermeasures are implemented to improve the analytical accuracy of the AI ​​model. This reduces the burden on caregivers and enables the provision of high-quality care.

[0807] "Data collection tools" are devices and techniques for collecting patient vital, environmental, nutritional, and audio data.

[0808] "Preprocessing means" are devices and techniques used to convert collected data into a form suitable for analysis.

[0809] "Data analysis means" refers to the equipment and technology used to input pre-processed data into an AI model for analysis.

[0810] The "result generation means" refers to devices and technologies for evaluating the cognitive state of a patient based on the results of data analysis and generating appropriate treatment methods.

[0811] The "notification means" refers to the device and technology for notifying the generated evaluation results and countermeasures to the user device.

[0812] The "feedback collection means" refers to devices and techniques for collecting data based on the results of measures taken by users and improving the accuracy of the system's analysis.

[0813] "Vital data" refers to a patient's biological information such as heart rate, blood pressure, and sleep time.

[0814] "Environmental data" refers to data related to the external environment, such as temperature and humidity.

[0815] "Nutrition data" refers to data on the types, amounts, and nutrients of food analyzed from dietary image data.

[0816] "Audio data" refers to data recorded from patients' everyday conversations and voice information.

[0817] An "AI model" is a pre-trained artificial intelligence system that includes algorithms for inferring a patient's cognitive state from input data.

[0818] "User device" refers to an electronic device, such as a smartphone or tablet, used by a caregiver.

[0819] The present invention provides a system for analyzing a wide range of data, allowing caregivers to grasp the cognitive state of dementia patients in real time and take appropriate measures. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[0820] Data collection

[0821] The server uses the data collection means to collect the following data:

[0822] Vital data: Collect the patient's heart rate, blood pressure, and sleep duration through the smartwatch.

[0823] Environmental data: Use the weather API to obtain temperature and humidity.

[0824] Nutritional data: Photos of meals are taken from the patient's smartphone and converted into data for nutritional analysis.

[0825] Voice data: The patient's everyday conversations and voices are recorded using a microphone and sent to the server.

[0826] Examples:

[0827] At 7:00 AM, the server accesses Patient A's smartwatch and captures the patient's heart rate and blood pressure data for the day.

[0828] Get the day's temperature and humidity data from the weather API.

[0829] Images of the day's breakfast are automatically uploaded from the patient's smartphone.

[0830] Data Preprocessing

[0831] The server uses pre-processing means to prepare the collected data in a suitable format for analysis.

[0832] Convert heart rate data: Format heart rate data into time series data.

[0833] Nutrition data conversion: Convert images into a format that is easy for AI to interpret.

[0834] Voice to Text Conversion: We use voice recognition software (e.g., Google Speech-to-Text) to convert voice data into text.

[0835] Examples:

[0836] The server takes the heart rate data and converts it into time series data.

[0837] Image recognition software is used to identify the types and quantities of food in the breakfast image and prepare it in a format suitable for analysis.

[0838] Everyday conversations received via voice are converted into text and prepared for sentiment analysis.

[0839] Data analysis and result generation

[0840] The server uses a data analysis means to input the pre-processed data into an AI model to analyze the patient's cognitive status, which is pre-trained to infer the cognitive status from the data.

[0841] Based on the analysis results, the server uses the result generation means to assess the patient's cognitive state and generate an appropriate course of action.

[0842] Examples:

[0843] The pre-processed heart rate data is fed into an AI model to detect values ​​outside the normal range.

[0844] Image data of breakfast is input into an AI model and the nutritional balance is analyzed.

[0845] The text data obtained by speech recognition is input into a sentiment analysis model to detect changes in emotion.

[0846] notification

[0847] The terminal uses a notification means to display the analysis results and countermeasures received from the server on the user's (caregiver's) device.

[0848] Notification method: A message will appear on your smartphone or tablet.

[0849] Notification content: Includes analysis results and details of how to deal with the issue.

[0850] Examples:

[0851] The caregiver's smartphone will receive a notification saying, "Patient A's cognitive state has deteriorated slightly. This is thought to be due to stress. We recommend a 15-minute light walk."

[0852] Implementing countermeasures

[0853] The user (caregiver) takes the notified action.

[0854] Action to be taken: Take the notified action.

[0855] Specific action: For example, a caregiver takes a walk around the neighborhood with patient A.

[0856] Examples:

[0857] The caregiver asks Patient A, "Let's take a 15-minute walk together," and they actually go for a walk.

[0858] Feedback collection

[0859] The server uses a feedback collection means to collect the results of the actions taken.

[0860] Feedback items: Heart rate after a walk, changes in the patient's facial expression, daily reactions, etc.

[0861] Data reuse: Collected data can be used to retrain AI models to improve analysis accuracy.

[0862] Examples:

[0863] After the walk, heart rate data is collected again from the smartwatch to assess the patient's stress level.

[0864] The patient's facial expressions and tone of voice are analyzed from camera and audio data and collected as feedback.

[0865] Example prompts to input to the generative AI model

[0866] "Patient A has a higher than normal heart rate and is not sleeping well at night. He is showing signs of stress and anxiety. Please suggest ways to address this."

[0867] "Based on Patient B's vital signs and dietary information today, please assess his cognitive status and suggest appropriate treatment options."

[0868] Through this series of processes, the system analyzes the patient's cognitive state in real time and promptly provides appropriate measures to caregivers, thereby reducing their burden and ensuring high-quality care. In addition, a feedback system continuously improves the analysis accuracy of the entire system.

[0869] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0870] Step 1:

[0871] Data collection

[0872] The server collects the following data using the data collection means:

[0873] Input: Patient's vital data (heart rate, blood pressure, sleep time), environmental data (temperature, humidity), nutritional data (images of meals), and audio data (daily conversation).

[0874] Specific operation: The server retrieves heart rate, blood pressure, and sleep time data from Patient A's smartwatch at 7:00 AM. It also retrieves the day's temperature and humidity using a weather API. It also uploads an image of breakfast from the patient's smartphone and records daily conversations using the microphone.

[0875] Output: Collected vital, environmental, nutritional and audio data.

[0876] Step 2:

[0877] Data Preprocessing

[0878] The server uses pre-processing means to prepare the collected data in a suitable format for analysis.

[0879] Input: Collected vital, environmental, nutritional, and audio data.

[0880] How it works: The server formats the heart rate data into time series data, converts the image of the breakfast into a format that is easy for AI to understand, and converts the audio data into text using speech recognition software.

[0881] Output: Preprocessed heart rate data, image data, and audio data converted to text.

[0882] Step 3:

[0883] Data analysis

[0884] The server uses data analysis means to input the preprocessed data into an AI model to analyze the patient's cognitive status.

[0885] Input: Preprocessed heart rate data, environmental data, nutrition data, and audio data.

[0886] Specific operations: Preprocessed heart rate data is input into an AI model to detect abnormal values. Image data of breakfast is input into an AI model to analyze nutritional balance. Text-converted voice data is input into an emotion analysis model to analyze changes in emotion.

[0887] Output: Analysis of the patient's cognitive status.

[0888] Step 4:

[0889] result generation

[0890] The server uses the result generation means to evaluate the cognitive state of the patient based on the analysis results and generates an appropriate course of action.

[0891] Input: Analysis results.

[0892] Specific behavior: Integrates the results of multiple analyses to assess the patient's cognitive state (e.g., stress level, lack of sleep, etc.), and uses a generative AI model to generate appropriate countermeasures (e.g., recommending a relaxing walk).

[0893] Output: Evaluation results and suggested solutions.

[0894] Step 5:

[0895] notification

[0896] The terminal uses the notification means to display the analysis results and countermeasures received from the server on the user device.

[0897] Input: Evaluation results and suggested actions.

[0898] What it does: Sends a detailed message to the caregiver's smartphone about the patient's cognitive status and how to address it. The notification might say, "Patient A's cognitive status has slightly deteriorated. This is thought to be due to stress. We recommend a 15-minute brisk walk."

[0899] Output: The notification message displayed on the user device.

[0900] Step 6:

[0901] Implementing countermeasures

[0902] The user (caregiver) takes the notified action.

[0903] Input: Notified suggested action.

[0904] Specific actions: The caregiver takes a walk around the neighborhood with Patient A. They say, "Let's take a 15-minute walk together," and then actually go for a walk.

[0905] Output: The outcome of the intervention (e.g., the patient's condition after the walk).

[0906] Step 7:

[0907] Feedback collection

[0908] The server collects the results of the actions taken using a feedback collection means.

[0909] Input: Results of the implemented measures (heart rate data, facial expression changes, etc.).

[0910] Specific operations: After the walk, the smartwatch collects heart rate data again to assess the patient's stress level, and analyzes the patient's facial expressions and tone of voice via camera and audio data to provide feedback.

[0911] Output: Collected feedback data, analysis results of improved AI model.

[0912] (Application example 1)

[0913] 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."

[0914] Until now, there has been no system that monitors a driver's fatigue and stress state in real time and provides appropriate countermeasures based on that information. This has made it difficult to improve driver safety and performance, potentially increasing the risk of traffic accidents. Therefore, the present invention aims to improve driver safety and comfort by analyzing the driver's vital signs, environmental data, and image data in real time and providing appropriate notifications and feedback according to the driver's condition.

[0915] 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.

[0916] In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, a means for collecting vital data, environmental data, and image data of the driver, a means for analyzing the data and generating recommendations based on the driver's condition, and a means for notifying the driver's device of the recommendations. This makes it possible to monitor the driver's fatigue and stress in real time and quickly provide appropriate countermeasures.

[0917] The "data collection means" is a means for acquiring various data including vital data of the driver, environmental data, and image data.

[0918] "Preprocessing means" refers to means for converting collected data into a format that is easy to analyze.

[0919] "Data analysis means" means a means for using pre-processed data to assess the driver's condition and analyzing it using specific algorithms or generative AI models.

[0920] The "result generation means" is a means for generating specific recommendations or notifications for the driver based on the results of the data analysis.

[0921] "Notification means" refers to a means for transmitting generated recommendations and notifications to the driver's device.

[0922] The "feedback collection means" is a means for recording the driver's response to the provided countermeasures and the progress of the improvement, and for use in reanalysis and model improvement.

[0923] "Vital data" refers to biometric information such as the driver's heart rate, blood pressure, and sleep data.

[0924] "Environmental data" refers to data relating to climatic conditions and the surrounding environment, including, for example, temperature, humidity, and weather information.

[0925] "Image data" refers to images of the driver's face and footage of the inside of the vehicle, and is data used to analyze the driver's facial expression and posture.

[0926] A "generative AI model" is a pre-trained artificial intelligence model that infers the driver's state based on collected data and generates appropriate recommendations.

[0927] The present invention provides a system for analyzing a driver's vital data, environmental data, and image data in real time and providing appropriate countermeasures based on the driver's condition. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[0928] The server uses data collection means to collect driver vital data, weather data, and driver image data. This data collection is done through wearable devices such as smartwatches and fitness trackers, weather APIs, and in-car cameras. For example, heart rate, blood pressure, and sleep data are acquired.

[0929] The collected data is preprocessed by the server. The preprocessing means standardizes vital data such as heart rate and blood pressure, converts image data to grayscale, and resizes it to an appropriate size. Weather data is obtained using an API and converted into a format suitable for analysis.

[0930] The server then inputs the preprocessed data into a generative AI model for data analysis. The generative AI model is pre-trained to infer the driver's state and detect conditions such as fatigue, stress, and distraction. Based on the results of this analysis, a specific recommendation is generated for the driver. For example, a message such as "You need a break. Please pull over and take a break" is generated.

[0931] The generated results are sent to the driver's device, such as a smartphone or an in-vehicle display, using a notification means, allowing the driver to immediately check their own condition and take any necessary measures.

[0932] After the driver actually takes the recommended action, the server uses a feedback collection method to collect the driver's reaction data again. For example, heart rate and facial expression changes after a break are collected and used to retrain the AI ​​model. This feedback function further improves the system's analysis accuracy.

[0933] As a concrete example, the following scenario can be considered.

[0934] Data collected:

[0935] Heart rate: 85 bpm

[0936] Blood pressure: 130 / 85 mmHg

[0937] Sleep data: 6 hours of sleep

[0938] Weather data: 22°C, 65% humidity

[0939] Driver's facial image

[0940] Generated recommendations:

[0941] The driver is likely fatigued, so a notification is generated saying, "You need a break. Please stop and take a break."

[0942] An example of a prompt is as follows:

[0943] "Based on the collected data, a generative AI model will be used to analyze the driver's condition and generate a notification that a break is needed, if necessary.

[0944] In this way, the present invention monitors the driver's health condition in real time and helps prevent traffic accidents and health risks.

[0945] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0946] Step 1:

[0947] The server uses data collection means to collect the driver's vital signs, weather data, and image data of the driver. Specifically, it obtains heart rate, blood pressure, and sleep data from the wearable device, and obtains current weather data (temperature, humidity, etc.) using a weather API. It also captures images of the driver's face using the in-car camera. This allows it to collect a wide range of data on the driver's condition and environment.

[0948] Input: Vital data from wearable devices, weather data from weather API, image data of the driver from an in-car camera

[0949] Output: Raw data collected

[0950] Step 2:

[0951] The server uses preprocessing tools to convert the collected data into a format suitable for analysis. Heart rate and blood pressure are normalized to a standard scale, sleep data is converted into an index to evaluate sleep quality, weather data is formatted from the API, and image data is converted to grayscale and resized to a size suitable for analysis. This prepares the data in a unified format.

[0952] Input: Raw data collected

[0953] Output: Preprocessed data

[0954] Step 3:

[0955] The server uses data analysis means to input preprocessed data into the generative AI model, which analyzes the driver's condition. The generative AI model combines and analyzes heart rate, blood pressure, sleep data, weather data, and image data to estimate the driver's fatigue level, stress level, distraction state, etc. The AI ​​model is pre-trained and uses this data to predict the driver's condition with high accuracy.

[0956] Input: Preprocessed data

[0957] Output: Analysis results (driver's current status evaluation)

[0958] Step 4:

[0959] The server generates specific recommendations for the driver based on the analysis results using the result generation means. For example, a message such as "The driver is highly fatigued, so a 15-minute break is recommended." This recommendation serves as a specific guideline for action to ensure the driver's safety.

[0960] Input: Analysis results (driver's current status evaluation)

[0961] Output: Recommendations

[0962] Step 5:

[0963] The terminal uses notification means to notify the driver of the generated recommendations. A message is displayed on the driver's smartphone or in-vehicle display, allowing the driver to immediately check the contents. This allows the driver to recognize their own condition and take any necessary measures promptly.

[0964] Input: Recommendations

[0965] Output: Notification to the driver

[0966] Step 6:

[0967] The user (driver) takes the appropriate action. For example, after checking the notification on the device, the user stops the car in a safe place and takes a break for the specified amount of time. This is expected to reduce the driver's fatigue and stress.

[0968] Input: Notification content

[0969] Output: Coping behavior

[0970] Step 7:

[0971] The server collects the results of the implemented measures using a feedback collection method. Feedback data such as heart rate, blood pressure, and changes in the driver's facial expressions after the break are collected again and used as training data for the generative AI model. This feedback improves the accuracy of the system.

[0972] Input: New vital data and image data after response actions

[0973] Output: Feedback data

[0974] Through the above series of processing steps, the system of the present invention can monitor the driver's health condition in real time and provide appropriate countermeasures promptly, thereby reducing the risk of traffic accidents and improving the safety and comfort of the driver.

[0975] 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.

[0976] The present invention is a system for supporting care of dementia patients, and includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion engine. The system aims to analyze a wide range of data, grasp the cognitive state of the patient in real time, and provide appropriate measures to caregivers.

[0977] First, the server collects vital data (e.g., heart rate, blood pressure), weather data (e.g., temperature, humidity), meal image data, and voice data through a data collection means. It also uses an emotion engine to recognize emotions from the voice data and facial expression data of users (caregivers and patients).

[0978] The server then uses preprocessing means to convert the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure into time-series data, converting food images into a format suitable for image recognition, and converting voice data into text. Furthermore, an emotion engine extracts emotions from the user's voice and facial expressions and records them as data.

[0979] The server then uses data analysis tools to input the preprocessed data into the AI ​​model to analyze the patient's cognitive state. The AI ​​model uses past learning data and analyzes current data to infer cognitive state and stress level. By including emotional data obtained from the emotion engine in this analysis, more detailed insights into the mental state of the patient and caregiver can be gained.

[0980] Based on the analysis results, the server uses the result generation means to determine an appropriate course of action. Specifically, if it determines that the cognitive state is slightly deteriorating, it generates a course of action to reduce stress, such as "recommending a walk to relax." It can also suggest a course of action that takes emotions into account by referring to the data obtained through the emotion engine.

[0981] The server then sends the generated results and advice on how to respond to the situation to the caregiver's device. The device (the caregiver's smartphone or tablet) displays the received notification in real time. An example message might read, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0982] The user (caregiver) then checks the notification received from the device and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the emotion engine also monitors changes in the caregiver's and patient's emotions and collects further data.

[0983] Finally, the user (caregiver) sends the results of the measures they took as feedback to the server. The server analyzes the collected feedback data and uses it to retrain the AI ​​model to improve its accuracy. In particular, since this feedback includes emotional data, it can make the next recommendation more accurate.

[0984] As a specific example, at 7:00 AM, the server collects Patient A's heart rate, sleep data, current temperature and humidity, and an image of their breakfast. At the same time, the emotion engine analyzes the caregiver's morning voice tone and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that Patient A's heart rate is higher than normal and that they are not getting enough sleep, suggesting stress. The result generation method generates a solution such as, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. We recommend a 15-minute light walk." The device then sends a notification to the caregiver, who then follows the recommendation and takes the walk, providing feedback on the results. This feedback data is collected and analyzed by the server to help further improve the system.

[0985] In this way, the system analyzes data from multiple perspectives and provides appropriate measures that take emotional data into consideration, thereby providing effective support for both caregivers and patients.

[0986] The processing flow will be explained below.

[0987] Step 1:

[0988] The server uses data collection tools to collect the patient's vital signs, weather data, meal image data, voice data, and user emotion data. For example, it obtains heart rate and blood pressure from a smartwatch and current temperature and humidity from a weather API. Meanwhile, the caregiver's smartphone camera takes pictures of the patient's breakfast and records the conversation using a voice recording device.

[0989] Step 2:

[0990] The server uses an emotion engine to analyze emotions from the voice and facial expression data of the caregiver and patient. For example, it analyzes emotions from voice and recognizes emotions such as stress and anxiety from the tone of the voice. It also analyzes the patient's facial expression from captured images to extract emotions such as smile and tension.

[0991] Step 3:

[0992] The server uses preprocessing means to convert the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure data into time-series data, converts food image data into a format suitable for image recognition, converts voice data into text, and quantifies and stores emotion data.

[0993] Step 4:

[0994] The server uses data analysis tools to input preprocessed data into the AI ​​model, which analyzes the patient's cognitive state. The AI ​​model suggests possible stress based on elevated heart rate and lack of sleep, and provides a complementary assessment of the degree of stress and anxiety based on the results of emotion analysis.

[0995] Step 5:

[0996] The server uses the result generation means to evaluate the patient's current cognitive state based on the analysis results obtained from the AI ​​model and generate appropriate measures to address the situation. For example, it generates specific advice such as, "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. We recommend a 15-minute light walk."

[0997] Step 6:

[0998] The server sends the generated results and recommended actions to the caregiver's device via a notification method. The device (caregiver's smartphone or tablet) displays this notification in real time. The caregiver receives a message saying, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[0999] Step 7:

[1000] The user (caregiver) checks the notification and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the caregiver carefully observes the patient's reactions and records any changes in their emotions.

[1001] Step 8:

[1002] The user (caregiver) sends the results of the measures taken to the server through a feedback collection means. Specifically, the user records the patient's heart rate after the walk, changes in the patient's facial expression, and changes in the caregiver's own emotions using a smartphone and sends them to the server.

[1003] Step 9:

[1004] The server retrains the AI ​​model based on the data obtained through the feedback collection method to improve the accuracy of the next analysis. Based on the feedback data, the relationship between stress state and changes in heart rate and emotions is reevaluated and reflected in future countermeasures.

[1005] Through the above processing flow, the system can analyze the patient's cognitive state and emotions from multiple angles and provide caregivers with appropriate coping strategies in real time, thereby improving the quality of care and reducing the burden on caregivers.

[1006] Example 2

[1007] 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."

[1008] In providing care support for dementia patients, it is extremely important to understand the patient's cognitive state and emotional changes in real time and provide caregivers with appropriate measures. However, current systems have fragmented data collection and analysis, making it difficult to comprehensively understand the patient's overall situation. In addition, measures that take emotional data into consideration are insufficient, which hinders caregivers from providing optimal support.

[1009] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion recognition means. This makes it possible to collect biological data, environmental data, image data, and voice data of a patient from various sources, and through preprocessing and analysis, grasp the patient's cognitive state and emotions in real time and provide appropriate measures promptly.

[1010] The "data collection means" is a means for collecting various data such as patient biometric data, environmental data, image data, and voice data.

[1011] "Preprocessing means" refers to the process of converting collected data into a format suitable for analysis. Specifically, this involves generating time-series data, converting it into an image recognition format, and converting voice data into text.

[1012] The "data analysis means" is a means for inputting preprocessed data into the AI ​​model and analyzing it, which allows for the prediction of the patient's cognitive state and stress level.

[1013] The "result generation means" is a means for determining and generating appropriate measures based on the results of data analysis. Specifically, it suggests relaxation techniques, etc.

[1014] The "notification means" is a means for transmitting the generated countermeasure to the caregiver's device and displaying the content thereof.

[1015] The "feedback collection means" is a means for collecting the results of the measures taken by the caregiver and using them for subsequent analysis and to improve the accuracy of the AI ​​model.

[1016] "Emotion recognition means" is a means of extracting emotions from voice data and facial expression data and recording them as data. This data plays an important role in data analysis.

[1017] An "AI model" is an artificial intelligence model that analyzes collected data and generates recommendations, enabling highly accurate analysis using past data.

[1018] "Vital data" refers to biological information such as heart rate, blood pressure, and body temperature.

[1019] "Image data" refers to image information captured with a camera of food, facial expressions, etc.

[1020] "Audio data" refers to data that indicates acoustic information that records user speech and environmental sounds.

[1021] "Environmental data" refers to data relating to the external environment, such as temperature, humidity, and weather.

[1022] The present invention is a system for supporting care of dementia patients, which includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion recognition means. The system aims to analyze a wide range of data, grasp the cognitive state of the patient in real time, and provide appropriate measures to caregivers.

[1023] First, the server collects vital data (e.g., heart rate, blood pressure, body temperature, etc.), environmental data (e.g., temperature, humidity, etc.), image data (photos of meals and facial expressions), and voice data through a data collection means. Vital data is obtained from a wearable device (e.g., Fitbit or Apple Watch). Environmental data is acquired from a weather API (e.g., OpenWeatherMap API). Image data is collected through the caregiver's smartphone camera, and voice data is obtained through a dedicated app. In addition, an emotion recognition means is used to recognize emotions from the user's voice data and facial expression data.

[1024] Next, the server uses preprocessing means to convert and organize the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure data into time-series data, converts food images into a format suitable for image recognition (e.g., RGB pixel array), and converts audio data into text using speech recognition software (e.g., Google Cloud Speech-to-Text API). Furthermore, it uses emotion recognition means to extract emotions from voice and facial expressions and record them as data.

[1025] The server then uses data analysis tools to input the preprocessed data into an AI model (e.g., a model implemented in TensorFlow or PyTorch) to analyze the patient's cognitive state. The analysis involves analyzing past training data and using current data to estimate cognitive state and stress levels. Emotion data obtained from emotion recognition tools can also be incorporated into this analysis to provide detailed insights into the mental state of the patient and caregiver.

[1026] Based on the analysis results, the server uses a result generation method to determine an appropriate course of action. For example, if it determines that the cognitive state is slightly deteriorating, it will generate a suggestion such as "recommended walk to relax" as a measure to reduce stress. It is also possible to refer to emotional data and propose a course of action that takes emotions into consideration.

[1027] The server then sends the generated results and advice on how to respond to the situation to the caregiver's device, which then displays the received notification in real time. For example, a message might appear saying, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[1028] The user (caregiver) then checks the notification received from the device and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the emotion recognition means monitors changes in the emotions of the caregiver and the patient and collects further data.

[1029] Finally, the user (caregiver) sends the results of the measures they took as feedback to the server. The server analyzes the collected feedback data and uses it to retrain the AI ​​model to improve its accuracy. In particular, since this feedback includes emotional data, it can make the next recommendation more accurate.

[1030] Examples:

[1031] At 7:00 AM, the server collects Patient A's heart rate, sleep data, current temperature and humidity, and an image of their breakfast. At the same time, the emotion recognition means analyzes the caregiver's morning tone of voice and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that Patient A's heart rate is higher than normal and that they are not getting enough sleep, suggesting stress. The result generation means generates a countermeasure such as, "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. We recommend a 15-minute light walk." The device then sends a notification to the caregiver, who then follows the recommendation and takes the walk, providing feedback on the results. This feedback data is collected and analyzed by the server to help further improve the system.

[1032] In this way, the system analyzes data from multiple perspectives and provides appropriate measures that take emotional data into consideration, thereby providing effective support for both caregivers and patients.

[1033] Example prompt sentence:

[1034] "Patient A's cognitive status has deteriorated slightly. This is likely due to stress. A 15-minute brisk walk is recommended."

[1035] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1036] Step 1: Data collection

[1037] The server collects various data through the data collection means.

[1038] Input: Patient vital data (heart rate, blood pressure, body temperature, etc.), environmental data (temperature, humidity, etc.), image data (photos of meals, images of facial expressions), voice data

[1039] Processing: Vital data is collected from wearable devices (e.g., Fitbit or Apple Watch). Environmental data is obtained from weather APIs (e.g., OpenWeatherMap API). Image data is collected through the caregiver's smartphone camera, and audio data is obtained through a dedicated app.

[1040] Output: A wide variety of raw data collected

[1041] Step 2: Data Preprocessing

[1042] The server uses pre-processing means to convert the collected data into a form suitable for analysis.

[1043] Input: Raw data collected

[1044] Processing: Heart rate and blood pressure data are converted into time-series data, and food images are converted into a format for image recognition (e.g., RGB pixel arrays). Audio data is converted into text using speech recognition software (e.g., Google Cloud Speech-to-Text API). Emotion recognition tools are used to extract emotions from voice and facial expressions and record them as data.

[1045] Output: Preprocessed datasets (time series data, formatted image data, text data, emotion data)

[1046] Step 3: Data analysis

[1047] The server uses data analysis means to input the preprocessed data into the AI ​​model and analyze the patient's cognitive status.

[1048] Input: Preprocessed datasets (time series data, formatted image data, text data, emotion data)

[1049] Processing: Data is fed into an AI model (e.g., a model implemented in TensorFlow or PyTorch) and current data is analyzed based on past training data. This allows for estimation of cognitive state and stress level. Emotion data obtained through emotion recognition methods is also included in the analysis.

[1050] Output: Analysis results (predictions of cognitive state, stress level, etc.)

[1051] Step 4: Result Generation

[1052] The server determines and generates an appropriate countermeasure based on the analysis result using the result generation means.

[1053] Input: Analysis results (predicted results such as cognitive state, stress level, etc.)

[1054] Processing: If the cognitive state is deteriorating, it will determine measures to reduce stress, such as suggesting a relaxing walk or a relaxation session. It will also take into account emotional data to generate coping strategies.

[1055] Output: A message with the recommendations

[1056] Step 5: Notification Delivery

[1057] The server transmits the generated results and actions to the caregiver's device.

[1058] Input: Message with recommendations

[1059] Processing: The generated message is sent to the caregiver's device (smartphone or tablet).

[1060] Output: Notification message displayed on the caregiver's device

[1061] Step 6: Implement the recommendations

[1062] The user (caregiver) checks the notification and takes the recommended action.

[1063] Input: Notification message

[1064] Action: Review the notification and take the recommended action with the patient, for example, take a walk with the patient.

[1065] Output: Action taken

[1066] Step 7: Gather feedback

[1067] The user (caregiver) sends the results of the execution to the server as feedback.

[1068] Input: Execution results (e.g., changes in the patient's condition or emotions after the walk)

[1069] Processing: The results of the countermeasures are recorded via the device and sent to the server, which collects and analyzes the feedback data and retrains the AI ​​model.

[1070] Output: Feedback data is stored and used on the server

[1071] Throughout this entire process, the system analyzes the data from multiple angles and provides appropriate responses that take emotional data into account, thereby providing effective support for both caregivers and patients.

[1072] (Application example 2)

[1073] 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."

[1074] In factories and other workplaces, it is necessary to grasp the physical and mental health status of employees in real time and quickly provide appropriate countermeasures. However, currently there is a lack of systems that can accurately analyze workload and stress and suggest appropriate countermeasures. This presents a problem in that it is not possible to provide appropriate support to ensure employee efficiency and safety.

[1075] 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 a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion analysis means. This makes it possible to analyze the physical and mental health conditions of workers in real time and quickly provide appropriate countermeasures.

[1076] The "data collection means" is a system element that collects vital data, work environment data, work image data, and voice data of workers.

[1077] A "preprocessing means" is an element of the system that converts and prepares collected data into a form suitable for analysis.

[1078] "Data analysis means" refers to the system element that inputs preprocessed data, including emotional data, into the AI ​​model for analysis.

[1079] A "result generation means" is an element of the system that determines the appropriate course of action based on the results of data analysis.

[1080] The "notification means" is a system element that sends the generated results and how to deal with them to the worker's device.

[1081] The "feedback collection means" is a system element that sends the results of the implemented countermeasures to the server and collects them.

[1082] "Emotion analysis means" is an element of the system that recognizes and analyzes emotions from the worker's facial expressions and voice data.

[1083] The present invention provides a system for monitoring and analyzing the health status and stress levels of employees in the workplace in real time, which includes a data collection means, a pre-processing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and a sentiment analysis means.

[1084] First, the server collects the worker's vital data (e.g., heart rate, blood pressure), work environment data (e.g., temperature, humidity), work image data, and voice data through a data collection means. It also uses an emotion analysis means to recognize the worker's emotions from their facial expressions and voice data.

[1085] Next, the server uses preprocessing means to convert the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure into time-series data, converting work images into a format suitable for image recognition, and converting voice data into text. Furthermore, emotion analysis means extracts emotions from the worker's voice and facial expressions and records them as data.

[1086] The server then uses data analysis means to input the preprocessed data into the AI ​​model, which analyzes the worker's health condition and stress level. The AI ​​model uses past learning data and analyzes current data to estimate stress levels and fatigue levels. By including emotional data obtained by the emotion analysis means in this analysis, detailed insights into the worker's mental state can be gained.

[1087] Based on the analysis results, the server uses the result generation means to determine an appropriate response. Specifically, if the stress level is determined to be high, the server generates a response such as recommending a break. Using data obtained through the emotion analysis means, the server can also suggest a response that takes emotions into consideration.

[1088] Next, the server sends the generated results and countermeasures to the worker's device (terminal). The terminal displays the received notification in real time. An example message might be, "Worker A's stress level is high, so we recommend that he take a 10-minute break."

[1089] The user (worker) then checks the notification received from the device and takes the recommended action. For example, the worker takes a 10-minute break in the break room. During this action, the emotion analysis means also monitors changes in the worker's emotions and collects further data.

[1090] Finally, the user (worker) sends the results of the implemented measures to the server as feedback. The server analyzes the collected feedback data and uses it for retraining to improve the accuracy of the AI ​​model. In particular, since this feedback includes emotional data, it will make the next recommendation more accurate.

[1091] As a specific example, at 7:00 AM, the server collects worker A's heart rate, sleep data, and current work environment data, while the emotion analysis means simultaneously analyzes the worker's morning tone of voice and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that worker A's heart rate is higher than normal and that he or she is highly fatigued. The result generation means generates a countermeasure such as, "Worker A's stress level is high, so we recommend that he or she take a 10-minute break." The device then sends a notification to the worker, and the user (worker) follows the recommendation and takes a break, providing feedback on the results. This feedback data is collected and analyzed by the server and used to further improve the system.

[1092] This system analyzes data from multiple angles and provides appropriate countermeasures that take emotional data into consideration, thereby effectively supporting both worker health and work efficiency.

[1093] Example prompt sentence:

[1094] Please explain the system that collects and analyzes the heart rate, blood pressure, facial expressions and voice data of factory workers, and recommends appropriate breaks and ways to deal with stress when the worker's stress level is high.

[1095] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1096] Step 1:

[1097] The server uses data collection means to collect worker vital data (heart rate, blood pressure), work environment data (temperature, humidity), work image data, and voice data. In this step, the input is real-time data obtained from various sensors and cameras, and the output is raw data compiled by the data collection means.

[1098] Step 2:

[1099] The server uses preprocessing means to convert and organize the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure into time-series data, converts work images into a format for image recognition, and converts voice data into text. The input to this step is the raw data collected in step 1, and the output is preprocessed data.

[1100] Step 3:

[1101] The server uses emotion analysis means to recognize and extract emotions from the worker's facial and voice data. The emotion data is used for subsequent analysis. The input for this step is facial and voice data obtained from a camera and microphone, and the output is emotion data recognized by the emotion analysis means.

[1102] Step 4:

[1103] The server uses data analysis means to input the preprocessed data and emotional data into the AI ​​model to analyze the worker's health condition and stress level. The AI ​​model performs analysis based on past learning data and makes an evaluation based on current data. The inputs for this step are the preprocessed data and emotional data, and the output is the analysis results.

[1104] Step 5:

[1105] The server uses the result generation means to determine an appropriate course of action based on the analysis results. For example, if the stress level is determined to be high, it generates a course of action such as recommending a break. The input of this step is the analysis result of step 4, and the output is a course of action.

[1106] Step 6:

[1107] The server uses the notification means to send the generated results and the countermeasures to the worker's terminal. The terminal displays the received notification to the worker in real time. The input of this step is information about the countermeasures, and the output is the notification sent to the worker's terminal.

[1108] Step 7:

[1109] The user (worker) checks the notification received from the terminal and takes the recommended action. For example, the user takes a specific action such as taking a 10-minute rest in the break room. The input of this step is the notification from the terminal, and the output is the action taken.

[1110] Step 8:

[1111] The user (operator) sends the results of the countermeasures they have taken as feedback to the server. The server analyzes the collected feedback data and uses it for retraining to improve the accuracy of the AI ​​model. The input of this step is the feedback information from the user, and the output is the data used for retraining.

[1112] 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.

[1113] 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.

[1114] 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.

[1115] [Fourth embodiment]

[1116] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1117] 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.

[1118] 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).

[1119] 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.

[1120] 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.

[1121] 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).

[1122] 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.

[1123] 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.

[1124] 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.

[1125] 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.

[1126] 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.

[1127] 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.

[1128] 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."

[1129] The present invention provides a system for analyzing a wide range of data, allowing caregivers to grasp the cognitive state of dementia patients in real time and take appropriate measures. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[1130] First, the server uses a data collection means to collect vital data, weather data, dietary image data, and voice data of the patient, thereby enabling comprehensive collection of a wide range of data related to the patient's daily life.

[1131] The server then uses preprocessing means to prepare the collected data in a format suitable for analysis, including converting heart rate data into time series data, food images into a data format for image recognition, and voice data into text.

[1132] The server then uses a data analysis means to input the pre-processed data into an AI model to analyze the patient's cognitive status, which is pre-trained to infer the cognitive status from the data.

[1133] Based on the analysis results, the server uses a result generation means to evaluate the patient's cognitive state and generate appropriate countermeasures. For example, if the patient's heart rate is high and stress is suspected, the server will suggest a countermeasure such as "recommending a walk to relax."

[1134] Next, the terminal uses a notification method to display the analysis results and countermeasures received from the server on the caregiver's device. For example, a message such as "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. A 15-minute light walk is recommended" may be displayed on a smartphone or tablet.

[1135] The user (caregiver) then implements the notified measures. For example, the caregiver may take a short walk with Patient A to reduce the patient's stress.

[1136] The server also uses a feedback collection method to collect the results of the implemented measures. For example, by sending the patient's heart rate after a walk or changes in facial expression back to the system, data can be provided to improve the analysis accuracy of the AI ​​model.

[1137] In this way, the system analyzes the patient's cognitive status in real time and promptly provides appropriate treatment to caregivers, thereby reducing the burden on caregivers and ensuring high-quality care.In addition, a feedback system continuously improves the accuracy of the analysis of the entire system.

[1138] Specific examples are given below.

[1139] Example 1:

[1140] At 7:00 AM, the server collects data on patient A's heart rate, blood pressure, sleep duration (from the smartwatch), current temperature and humidity (from the weather API), and an image of the patient's breakfast (from the smartphone).

[1141] Example 2:

[1142] The server inputs this data into the AI ​​model, and the AI ​​suggests that Patient A's heart rate is higher than normal and that he is not sleeping well at night, which may indicate stress or anxiety.

[1143] Example 3:

[1144] The device displays a message on the caregiver's smartphone such as, "Patient A's cognitive state has deteriorated slightly. This is thought to be due to stress. We recommend a 15-minute light walk."

[1145] Example 4:

[1146] The user (caregiver) checks the notification and tries to reduce stress by taking a walk around the neighborhood with Patient A.

[1147] Example 5:

[1148] The server again collects data on heart rate and changes in the patient's facial expression after the walk, and uses this data to retrain the AI ​​model.

[1149] Through this series of processes, the system constantly assesses the patient's cognitive status based on the latest information and provides appropriate advice to caregivers, thereby achieving higher quality care and reducing the burden on caregivers.

[1150] The processing flow will be explained below.

[1151] Step 1:

[1152] The server collects the patient's vital signs, weather data, meal image data, and audio data via a smartwatch, weather API, smartphone camera, audio recording device, etc. Specifically, at 7:00 AM, the server obtains Patient A's heart rate, blood pressure, sleep data, current temperature and humidity, and an image of his breakfast.

[1153] Step 2:

[1154] The server preprocesses the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure data into time-series data, converting food images into a format suitable for image recognition, and converting voice data into text. Specifically, the server organizes the heart rate data and checks for any abnormal values.

[1155] Step 3:

[1156] The server then inputs the preprocessed data into an AI model for analysis. The AI ​​model then analyzes the current data based on the previously learned data and predicts the patient's cognitive state. For example, if the patient's heart rate is higher than normal and they are not getting enough sleep, the AI ​​model will generate a result suggesting stress or anxiety.

[1157] Step 4:

[1158] The server evaluates the patient's cognitive state based on the analysis results obtained from the AI ​​model and generates appropriate treatment methods. For example, if stress is determined to be the cause, it generates specific advice such as "recommending a walk to relax."

[1159] Step 5:

[1160] The server sends the generated results and recommended actions to the caregiver's device. The device (the caregiver's smartphone or tablet) displays the results in real time. For example, a notification message might read, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[1161] Step 6:

[1162] The user (caregiver) checks the notification and takes the recommended action, for example, taking a walk around the neighborhood with Patient A.

[1163] Step 7:

[1164] The user (caregiver) records the results of the measures taken and provides feedback to the server. Specifically, the user records the patient's heart rate and changes in facial expression after the walk using a smartphone and sends the results to the server.

[1165] Step 8:

[1166] The server retrains the AI ​​model based on the collected feedback data to improve future analysis accuracy, for example, by using the feedback data to re-learn the relationship between stress levels and heart rate.

[1167] Through this series of processes, the system constantly assesses the patient's cognitive status based on the latest information and provides appropriate advice to caregivers, thereby achieving higher quality care and reducing the burden on caregivers.

[1168] Example 1

[1169] 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."

[1170] In conventional dementia care, it has been difficult for caregivers to grasp the patient's cognitive state in real time and provide appropriate treatment promptly. Furthermore, there has been a lack of mechanisms for appropriately collecting feedback and utilizing it to improve the system's analysis accuracy. This has increased the burden on caregivers, making it difficult to provide high-quality care.

[1171] 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.

[1172] In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means. This allows the server to collect and analyze the patient's vital data, environmental data, nutritional data, and voice data in real time. The server also notifies the user device of the generated evaluation results and countermeasures, and recollects data after the countermeasures are implemented to improve the analytical accuracy of the AI ​​model. This reduces the burden on caregivers and enables the provision of high-quality care.

[1173] "Data collection tools" are devices and techniques for collecting patient vital, environmental, nutritional, and audio data.

[1174] "Preprocessing means" are devices and techniques used to convert collected data into a form suitable for analysis.

[1175] "Data analysis means" refers to the equipment and technology used to input pre-processed data into an AI model for analysis.

[1176] The "result generation means" refers to devices and technologies for evaluating the cognitive state of a patient based on the results of data analysis and generating appropriate treatment methods.

[1177] The "notification means" refers to the device and technology for notifying the generated evaluation results and countermeasures to the user device.

[1178] The "feedback collection means" refers to devices and techniques for collecting data based on the results of measures taken by users and improving the accuracy of the system's analysis.

[1179] "Vital data" refers to a patient's biological information such as heart rate, blood pressure, and sleep time.

[1180] "Environmental data" refers to data related to the external environment, such as temperature and humidity.

[1181] "Nutrition data" refers to data on the types, amounts, and nutrients of food analyzed from dietary image data.

[1182] "Audio data" refers to data recorded from patients' everyday conversations and voice information.

[1183] An "AI model" is a pre-trained artificial intelligence system that includes algorithms for inferring a patient's cognitive state from input data.

[1184] "User device" refers to an electronic device, such as a smartphone or tablet, used by a caregiver.

[1185] The present invention provides a system for analyzing a wide range of data, allowing caregivers to grasp the cognitive state of dementia patients in real time and take appropriate measures. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[1186] Data collection

[1187] The server uses the data collection means to collect the following data:

[1188] Vital data: Collect the patient's heart rate, blood pressure, and sleep duration through the smartwatch.

[1189] Environmental data: Use the weather API to obtain temperature and humidity.

[1190] Nutritional data: Photos of meals are taken from the patient's smartphone and converted into data for nutritional analysis.

[1191] Voice data: The patient's everyday conversations and voices are recorded using a microphone and sent to the server.

[1192] Examples:

[1193] At 7:00 AM, the server accesses Patient A's smartwatch and captures the patient's heart rate and blood pressure data for the day.

[1194] Get the day's temperature and humidity data from the weather API.

[1195] Images of the day's breakfast are automatically uploaded from the patient's smartphone.

[1196] Data Preprocessing

[1197] The server uses pre-processing means to prepare the collected data in a suitable format for analysis.

[1198] Convert heart rate data: Format heart rate data into time series data.

[1199] Nutrition data conversion: Convert images into a format that is easy for AI to interpret.

[1200] Voice to Text Conversion: We use voice recognition software (e.g., Google Speech-to-Text) to convert voice data into text.

[1201] Examples:

[1202] The server takes the heart rate data and converts it into time series data.

[1203] Image recognition software is used to identify the types and quantities of food in the breakfast image and prepare it in a format suitable for analysis.

[1204] Everyday conversations received via voice are converted into text and prepared for sentiment analysis.

[1205] Data analysis and result generation

[1206] The server uses a data analysis means to input the pre-processed data into an AI model to analyze the patient's cognitive status, which is pre-trained to infer the cognitive status from the data.

[1207] Based on the analysis results, the server uses the result generation means to assess the patient's cognitive state and generate an appropriate course of action.

[1208] Examples:

[1209] The pre-processed heart rate data is fed into an AI model to detect values ​​outside the normal range.

[1210] Image data of breakfast is input into an AI model and the nutritional balance is analyzed.

[1211] The text data obtained by speech recognition is input into a sentiment analysis model to detect changes in emotion.

[1212] notification

[1213] The terminal uses a notification means to display the analysis results and countermeasures received from the server on the user's (caregiver's) device.

[1214] Notification method: A message will appear on your smartphone or tablet.

[1215] Notification content: Includes analysis results and details of how to deal with the issue.

[1216] Examples:

[1217] The caregiver's smartphone will receive a notification saying, "Patient A's cognitive state has deteriorated slightly. This is thought to be due to stress. We recommend a 15-minute light walk."

[1218] Implementing countermeasures

[1219] The user (caregiver) takes the notified action.

[1220] Action to be taken: Take the notified action.

[1221] Specific action: For example, a caregiver takes a walk around the neighborhood with patient A.

[1222] Examples:

[1223] The caregiver asks Patient A, "Let's take a 15-minute walk together," and they actually go for a walk.

[1224] Feedback collection

[1225] The server uses a feedback collection means to collect the results of the actions taken.

[1226] Feedback items: Heart rate after a walk, changes in the patient's facial expression, daily reactions, etc.

[1227] Data reuse: Collected data can be used to retrain AI models to improve analysis accuracy.

[1228] Examples:

[1229] After the walk, heart rate data is collected again from the smartwatch to assess the patient's stress level.

[1230] The patient's facial expressions and tone of voice are analyzed from camera and audio data and collected as feedback.

[1231] Example prompts to input to the generative AI model

[1232] "Patient A has a higher than normal heart rate and is not sleeping well at night. He is showing signs of stress and anxiety. Please suggest ways to address this."

[1233] "Based on Patient B's vital signs and dietary information today, please assess his cognitive status and suggest appropriate treatment options."

[1234] Through this series of processes, the system analyzes the patient's cognitive state in real time and promptly provides appropriate measures to caregivers, thereby reducing their burden and ensuring high-quality care. In addition, a feedback system continuously improves the analysis accuracy of the entire system.

[1235] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1236] Step 1:

[1237] Data collection

[1238] The server collects the following data using the data collection means:

[1239] Input: Patient's vital data (heart rate, blood pressure, sleep time), environmental data (temperature, humidity), nutritional data (images of meals), and audio data (daily conversation).

[1240] Specific operation: The server retrieves heart rate, blood pressure, and sleep time data from Patient A's smartwatch at 7:00 AM. It also retrieves the day's temperature and humidity using a weather API. It also uploads an image of breakfast from the patient's smartphone and records daily conversations using the microphone.

[1241] Output: Collected vital, environmental, nutritional and audio data.

[1242] Step 2:

[1243] Data Preprocessing

[1244] The server uses pre-processing means to prepare the collected data in a suitable format for analysis.

[1245] Input: Collected vital, environmental, nutritional, and audio data.

[1246] How it works: The server formats the heart rate data into time series data, converts the image of the breakfast into a format that is easy for AI to understand, and converts the audio data into text using speech recognition software.

[1247] Output: Preprocessed heart rate data, image data, and audio data converted to text.

[1248] Step 3:

[1249] Data analysis

[1250] The server uses data analysis means to input the preprocessed data into an AI model to analyze the patient's cognitive status.

[1251] Input: Preprocessed heart rate data, environmental data, nutrition data, and audio data.

[1252] Specific operations: Preprocessed heart rate data is input into an AI model to detect abnormal values. Image data of breakfast is input into an AI model to analyze nutritional balance. Text-converted voice data is input into an emotion analysis model to analyze changes in emotion.

[1253] Output: Analysis of the patient's cognitive status.

[1254] Step 4:

[1255] result generation

[1256] The server uses the result generation means to evaluate the cognitive state of the patient based on the analysis results and generates an appropriate course of action.

[1257] Input: Analysis results.

[1258] Specific behavior: Integrates the results of multiple analyses to assess the patient's cognitive state (e.g., stress level, lack of sleep, etc.), and uses a generative AI model to generate appropriate countermeasures (e.g., recommending a relaxing walk).

[1259] Output: Evaluation results and suggested solutions.

[1260] Step 5:

[1261] notification

[1262] The terminal uses the notification means to display the analysis results and countermeasures received from the server on the user device.

[1263] Input: Evaluation results and suggested actions.

[1264] What it does: Sends a detailed message to the caregiver's smartphone about the patient's cognitive status and how to address it. The notification might say, "Patient A's cognitive status has slightly deteriorated. This is thought to be due to stress. We recommend a 15-minute brisk walk."

[1265] Output: The notification message displayed on the user device.

[1266] Step 6:

[1267] Implementing countermeasures

[1268] The user (caregiver) takes the notified action.

[1269] Input: Notified suggested action.

[1270] Specific actions: The caregiver takes a walk around the neighborhood with Patient A. They say, "Let's take a 15-minute walk together," and then actually go for a walk.

[1271] Output: The outcome of the intervention (e.g., the patient's condition after the walk).

[1272] Step 7:

[1273] Feedback collection

[1274] The server collects the results of the actions taken using a feedback collection means.

[1275] Input: Results of the implemented measures (heart rate data, facial expression changes, etc.).

[1276] Specific operations: After the walk, the smartwatch collects heart rate data again to assess the patient's stress level, and analyzes the patient's facial expressions and tone of voice via camera and audio data to provide feedback.

[1277] Output: Collected feedback data, analysis results of improved AI model.

[1278] (Application example 1)

[1279] 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."

[1280] Until now, there has been no system that monitors a driver's fatigue and stress state in real time and provides appropriate countermeasures based on that information. This has made it difficult to improve driver safety and performance, potentially increasing the risk of traffic accidents. Therefore, the present invention aims to improve driver safety and comfort by analyzing the driver's vital signs, environmental data, and image data in real time and providing appropriate notifications and feedback according to the driver's condition.

[1281] 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.

[1282] In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, a means for collecting vital data, environmental data, and image data of the driver, a means for analyzing the data and generating recommendations based on the driver's condition, and a means for notifying the driver's device of the recommendations. This makes it possible to monitor the driver's fatigue and stress in real time and quickly provide appropriate countermeasures.

[1283] The "data collection means" is a means for acquiring various data including vital data of the driver, environmental data, and image data.

[1284] "Preprocessing means" refers to means for converting collected data into a format that is easy to analyze.

[1285] "Data analysis means" means a means for using pre-processed data to assess the driver's condition and analyzing it using specific algorithms or generative AI models.

[1286] The "result generation means" is a means for generating specific recommendations or notifications for the driver based on the results of the data analysis.

[1287] "Notification means" refers to a means for transmitting generated recommendations and notifications to the driver's device.

[1288] The "feedback collection means" is a means for recording the driver's response to the provided countermeasures and the progress of the improvement, and for use in reanalysis and model improvement.

[1289] "Vital data" refers to biometric information such as the driver's heart rate, blood pressure, and sleep data.

[1290] "Environmental data" refers to data relating to climatic conditions and the surrounding environment, including, for example, temperature, humidity, and weather information.

[1291] "Image data" refers to images of the driver's face and footage of the inside of the vehicle, and is data used to analyze the driver's facial expression and posture.

[1292] A "generative AI model" is a pre-trained artificial intelligence model that infers the driver's state based on collected data and generates appropriate recommendations.

[1293] The present invention provides a system for analyzing a driver's vital data, environmental data, and image data in real time and providing appropriate countermeasures based on the driver's condition. The system includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, and a feedback collection means.

[1294] The server uses data collection means to collect driver vital data, weather data, and driver image data. This data collection is done through wearable devices such as smartwatches and fitness trackers, weather APIs, and in-car cameras. For example, heart rate, blood pressure, and sleep data are acquired.

[1295] The collected data is preprocessed by the server. The preprocessing means standardizes vital data such as heart rate and blood pressure, converts image data to grayscale, and resizes it to an appropriate size. Weather data is obtained using an API and converted into a format suitable for analysis.

[1296] The server then inputs the preprocessed data into a generative AI model for data analysis. The generative AI model is pre-trained to infer the driver's state and detect conditions such as fatigue, stress, and distraction. Based on the results of this analysis, a specific recommendation is generated for the driver. For example, a message such as "You need a break. Please pull over and take a break" is generated.

[1297] The generated results are sent to the driver's device, such as a smartphone or an in-vehicle display, using a notification means, allowing the driver to immediately check their own condition and take any necessary measures.

[1298] After the driver actually takes the recommended action, the server uses a feedback collection method to collect the driver's reaction data again. For example, heart rate and facial expression changes after a break are collected and used to retrain the AI ​​model. This feedback function further improves the system's analysis accuracy.

[1299] As a concrete example, the following scenario can be considered.

[1300] Data collected:

[1301] Heart rate: 85 bpm

[1302] Blood pressure: 130 / 85 mmHg

[1303] Sleep data: 6 hours of sleep

[1304] Weather data: 22°C, 65% humidity

[1305] Driver's facial image

[1306] Generated recommendations:

[1307] The driver is likely fatigued, so a notification is generated saying, "You need a break. Please stop and take a break."

[1308] An example of a prompt is as follows:

[1309] "Based on the collected data, a generative AI model will be used to analyze the driver's condition and generate a notification that a break is needed, if necessary.

[1310] In this way, the present invention monitors the driver's health condition in real time and helps prevent traffic accidents and health risks.

[1311] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1312] Step 1:

[1313] The server uses data collection means to collect the driver's vital signs, weather data, and image data of the driver. Specifically, it obtains heart rate, blood pressure, and sleep data from the wearable device, and obtains current weather data (temperature, humidity, etc.) using a weather API. It also captures images of the driver's face using the in-car camera. This allows it to collect a wide range of data on the driver's condition and environment.

[1314] Input: Vital data from wearable devices, weather data from weather API, image data of the driver from an in-car camera

[1315] Output: Raw data collected

[1316] Step 2:

[1317] The server uses preprocessing tools to convert the collected data into a format suitable for analysis. Heart rate and blood pressure are normalized to a standard scale, sleep data is converted into an index to evaluate sleep quality, weather data is formatted from the API, and image data is converted to grayscale and resized to a size suitable for analysis. This prepares the data in a unified format.

[1318] Input: Raw data collected

[1319] Output: Preprocessed data

[1320] Step 3:

[1321] The server uses data analysis means to input preprocessed data into the generative AI model, which analyzes the driver's condition. The generative AI model combines and analyzes heart rate, blood pressure, sleep data, weather data, and image data to estimate the driver's fatigue level, stress level, distraction state, etc. The AI ​​model is pre-trained and uses this data to predict the driver's condition with high accuracy.

[1322] Input: Preprocessed data

[1323] Output: Analysis results (driver's current status evaluation)

[1324] Step 4:

[1325] The server generates specific recommendations for the driver based on the analysis results using the result generation means. For example, a message such as "The driver is highly fatigued, so a 15-minute break is recommended." This recommendation serves as a specific guideline for action to ensure the driver's safety.

[1326] Input: Analysis results (driver's current status evaluation)

[1327] Output: Recommendations

[1328] Step 5:

[1329] The terminal uses notification methods to notify the driver of the generated recommendations. A message is displayed on the driver's smartphone or in-vehicle display, allowing the driver to immediately check the contents. This allows the driver to recognize their own condition and take any necessary measures promptly.

[1330] Input: Recommendations

[1331] Output: Notification to the driver

[1332] Step 6:

[1333] The user (driver) takes the appropriate action. For example, after checking the notification on the device, the user stops the car in a safe place and takes a break for the specified amount of time. This is expected to reduce the driver's fatigue and stress.

[1334] Input: Notification content

[1335] Output: Coping behavior

[1336] Step 7:

[1337] The server collects the results of the implemented measures using a feedback collection method. Feedback data such as heart rate, blood pressure, and changes in the driver's facial expressions after the break are collected again and used as training data for the generative AI model. This feedback improves the accuracy of the system.

[1338] Input: New vital data and image data after response actions

[1339] Output: Feedback data

[1340] Through the above series of processing steps, the system of the present invention can monitor the driver's health condition in real time and provide appropriate countermeasures promptly, thereby reducing the risk of traffic accidents and improving the safety and comfort of the driver.

[1341] 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.

[1342] The present invention is a system for supporting care of dementia patients, and includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion engine. The system aims to analyze a wide range of data, grasp the cognitive state of the patient in real time, and provide appropriate measures to caregivers.

[1343] First, the server collects vital data (e.g., heart rate, blood pressure), weather data (e.g., temperature, humidity), meal image data, and voice data through a data collection means. It also uses an emotion engine to recognize emotions from the voice data and facial expression data of users (caregivers and patients).

[1344] The server then uses preprocessing means to convert the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure into time-series data, converting food images into a format suitable for image recognition, and converting voice data into text. Furthermore, an emotion engine extracts emotions from the user's voice and facial expressions and records them as data.

[1345] The server then uses data analysis tools to input the preprocessed data into the AI ​​model to analyze the patient's cognitive state. The AI ​​model uses past learning data and analyzes current data to infer cognitive state and stress level. By including emotional data obtained from the emotion engine in this analysis, more detailed insights into the mental state of the patient and caregiver can be gained.

[1346] Based on the analysis results, the server uses the result generation means to determine an appropriate course of action. Specifically, if it determines that the cognitive state is slightly deteriorating, it generates a course of action to reduce stress, such as "recommending a walk to relax." It can also suggest a course of action that takes emotions into account by referring to the data obtained through the emotion engine.

[1347] The server then sends the generated results and advice on how to respond to the situation to the caregiver's device. The device (the caregiver's smartphone or tablet) displays the received notification in real time. An example message might read, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[1348] The user (caregiver) then checks the notification received from the device and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the emotion engine also monitors changes in the caregiver's and patient's emotions and collects further data.

[1349] Finally, the user (caregiver) sends the results of the measures they took as feedback to the server. The server analyzes the collected feedback data and uses it to retrain the AI ​​model to improve its accuracy. In particular, since this feedback includes emotional data, it can make the next recommendation more accurate.

[1350] As a specific example, at 7:00 AM, the server collects Patient A's heart rate, sleep data, current temperature and humidity, and an image of their breakfast. At the same time, the emotion engine analyzes the caregiver's morning voice tone and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that Patient A's heart rate is higher than normal and that they are not getting enough sleep, suggesting stress. The result generation method generates a solution such as, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. We recommend a 15-minute light walk." The device then sends a notification to the caregiver, who then follows the recommendation and takes the walk, providing feedback on the results. This feedback data is collected and analyzed by the server to help further improve the system.

[1351] In this way, the system analyzes data from multiple perspectives and provides appropriate measures that take emotional data into consideration, thereby providing effective support for both caregivers and patients.

[1352] The processing flow will be explained below.

[1353] Step 1:

[1354] The server uses data collection tools to collect the patient's vital signs, weather data, meal image data, voice data, and user emotion data. For example, it obtains heart rate and blood pressure from a smartwatch and current temperature and humidity from a weather API. Meanwhile, the caregiver's smartphone camera takes pictures of the patient's breakfast and records the conversation using a voice recording device.

[1355] Step 2:

[1356] The server uses an emotion engine to analyze emotions from the voice and facial expression data of the caregiver and patient. For example, it analyzes emotions from voice and recognizes emotions such as stress and anxiety from the tone of the voice. It also analyzes the patient's facial expression from captured images to extract emotions such as smile and tension.

[1357] Step 3:

[1358] The server uses preprocessing means to convert the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure data into time-series data, converts food image data into a format suitable for image recognition, converts voice data into text, and quantifies and stores emotion data.

[1359] Step 4:

[1360] The server uses data analysis tools to input preprocessed data into the AI ​​model, which analyzes the patient's cognitive state. The AI ​​model suggests possible stress based on elevated heart rate and lack of sleep, and provides a complementary assessment of the degree of stress and anxiety based on the results of emotion analysis.

[1361] Step 5:

[1362] The server uses the result generation means to evaluate the patient's current cognitive state based on the analysis results obtained from the AI ​​model and generate appropriate measures to address the situation. For example, it generates specific advice such as, "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. We recommend a 15-minute light walk."

[1363] Step 6:

[1364] The server sends the generated results and recommended actions to the caregiver's device via a notification method. The device (caregiver's smartphone or tablet) displays this notification in real time. The caregiver receives a message saying, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[1365] Step 7:

[1366] The user (caregiver) checks the notification and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the caregiver carefully observes the patient's reactions and records any changes in their emotions.

[1367] Step 8:

[1368] The user (caregiver) sends the results of the measures taken to the server through a feedback collection means. Specifically, the user records the patient's heart rate after the walk, changes in the patient's facial expression, and changes in the caregiver's own emotions using a smartphone and sends them to the server.

[1369] Step 9:

[1370] The server retrains the AI ​​model based on the data obtained through the feedback collection method to improve the accuracy of the next analysis. Based on the feedback data, the relationship between stress state and changes in heart rate and emotions is reevaluated and reflected in future countermeasures.

[1371] Through the above processing flow, the system can analyze the patient's cognitive state and emotions from multiple angles and provide caregivers with appropriate coping strategies in real time, thereby improving the quality of care and reducing the burden on caregivers.

[1372] Example 2

[1373] 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."

[1374] In providing care support for dementia patients, it is extremely important to understand the patient's cognitive state and emotional changes in real time and provide caregivers with appropriate measures. However, current systems have fragmented data collection and analysis, making it difficult to comprehensively understand the patient's overall situation. In addition, measures that take emotional data into consideration are insufficient, which hinders caregivers from providing optimal support.

[1375] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion recognition means. This makes it possible to collect biological data, environmental data, image data, and voice data of a patient from various sources, and through preprocessing and analysis, grasp the patient's cognitive state and emotions in real time and provide appropriate measures promptly.

[1376] The "data collection means" is a means for collecting various data such as patient biometric data, environmental data, image data, and voice data.

[1377] "Preprocessing means" refers to the means of converting collected data into a format suitable for analysis. Specifically, it generates time-series data, converts it into an image recognition format, converts voice data into text, etc.

[1378] The "data analysis means" is a means for inputting preprocessed data into the AI ​​model and analyzing it, which allows for the prediction of the patient's cognitive state and stress level.

[1379] The "result generation means" is a means for determining and generating appropriate measures based on the results of data analysis. Specifically, it suggests relaxation techniques, etc.

[1380] The "notification means" is a means for transmitting the generated countermeasure to the caregiver's device and displaying the content thereof.

[1381] The "feedback collection means" is a means for collecting the results of the measures taken by the caregiver and using them for subsequent analysis and to improve the accuracy of the AI ​​model.

[1382] "Emotion recognition means" is a means of extracting emotions from voice data and facial expression data and recording them as data. This data plays an important role in data analysis.

[1383] An "AI model" is an artificial intelligence model that analyzes collected data and generates recommendations, enabling highly accurate analysis using past data.

[1384] "Vital data" refers to biological information such as heart rate, blood pressure, and body temperature.

[1385] "Image data" refers to image information captured with a camera of food, facial expressions, etc.

[1386] "Audio data" refers to data that indicates acoustic information that records user speech and environmental sounds.

[1387] "Environmental data" refers to data relating to the external environment, such as temperature, humidity, and weather.

[1388] The present invention is a system for supporting care of dementia patients, which includes a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion recognition means. The system aims to analyze a wide range of data, grasp the cognitive state of the patient in real time, and provide appropriate measures to caregivers.

[1389] First, the server collects vital data (e.g., heart rate, blood pressure, body temperature, etc.), environmental data (e.g., temperature, humidity, etc.), image data (photos of meals and facial expressions), and voice data through a data collection means. Vital data is obtained from a wearable device (e.g., Fitbit or Apple Watch). Environmental data is acquired from a weather API (e.g., OpenWeatherMap API). Image data is collected through the caregiver's smartphone camera, and voice data is obtained through a dedicated app. In addition, an emotion recognition means is used to recognize emotions from the user's voice data and facial expression data.

[1390] Next, the server uses preprocessing means to convert and organize the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure data into time-series data, converts food images into a format suitable for image recognition (e.g., RGB pixel array), and converts audio data into text using speech recognition software (e.g., Google Cloud Speech-to-Text API). Furthermore, it uses emotion recognition means to extract emotions from voice and facial expressions and record them as data.

[1391] The server then uses data analysis tools to input the preprocessed data into an AI model (e.g., a model implemented in TensorFlow or PyTorch) to analyze the patient's cognitive state. The analysis involves analyzing past training data and using current data to estimate cognitive state and stress levels. Emotion data obtained from emotion recognition tools can also be incorporated into this analysis to provide detailed insights into the mental state of the patient and caregiver.

[1392] Based on the analysis results, the server uses a result generation method to determine an appropriate course of action. For example, if it determines that the cognitive state is slightly deteriorating, it will generate a suggestion such as "recommended walk to relax" as a measure to reduce stress. It is also possible to refer to emotional data and propose a course of action that takes emotions into consideration.

[1393] The server then sends the generated results and advice on how to respond to the situation to the caregiver's device, which then displays the received notification in real time. For example, a message might appear saying, "Patient A's cognitive status has slightly deteriorated. Stress is thought to be the cause. A 15-minute light walk is recommended."

[1394] The user (caregiver) then checks the notification received from the device and takes the recommended action. For example, the caregiver takes a walk around the neighborhood with Patient A. During this process, the emotion recognition means monitors changes in the emotions of the caregiver and the patient and collects further data.

[1395] Finally, the user (caregiver) sends the results of the measures they took as feedback to the server. The server analyzes the collected feedback data and uses it to retrain the AI ​​model to improve its accuracy. In particular, since this feedback includes emotional data, it can make the next recommendation more accurate.

[1396] Examples:

[1397] At 7:00 AM, the server collects Patient A's heart rate, sleep data, current temperature and humidity, and an image of their breakfast. At the same time, the emotion recognition means analyzes the caregiver's morning tone of voice and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that Patient A's heart rate is higher than normal and that they are not getting enough sleep, suggesting stress. The result generation means generates a countermeasure such as, "Patient A's cognitive state has deteriorated slightly. Stress is thought to be the cause. We recommend a 15-minute light walk." The device then sends a notification to the caregiver, who then follows the recommendation and takes the walk, providing feedback on the results. This feedback data is collected and analyzed by the server to help further improve the system.

[1398] In this way, the system analyzes data from multiple perspectives and provides appropriate measures that take emotional data into consideration, thereby providing effective support for both caregivers and patients.

[1399] Example prompt sentence:

[1400] "Patient A's cognitive status has deteriorated slightly. This is likely due to stress. A 15-minute brisk walk is recommended."

[1401] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1402] Step 1: Data collection

[1403] The server collects various data through the data collection means.

[1404] Input: Patient vital data (heart rate, blood pressure, body temperature, etc.), environmental data (temperature, humidity, etc.), image data (photos of meals, images of facial expressions), voice data

[1405] Processing: Vital data is collected from wearable devices (e.g., Fitbit or Apple Watch). Environmental data is obtained from weather APIs (e.g., OpenWeatherMap API). Image data is collected through the caregiver's smartphone camera, and audio data is obtained through a dedicated app.

[1406] Output: A wide variety of raw data collected

[1407] Step 2: Data Preprocessing

[1408] The server uses pre-processing means to convert the collected data into a form suitable for analysis.

[1409] Input: Raw data collected

[1410] Processing: Heart rate and blood pressure data are converted into time-series data, and food images are converted into a format for image recognition (e.g., RGB pixel arrays). Audio data is converted into text using speech recognition software (e.g., Google Cloud Speech-to-Text API). Emotion recognition tools are used to extract emotions from voice and facial expressions and record them as data.

[1411] Output: Preprocessed datasets (time series data, formatted image data, text data, emotion data)

[1412] Step 3: Data analysis

[1413] The server uses data analysis means to input the preprocessed data into the AI ​​model and analyze the patient's cognitive status.

[1414] Input: Preprocessed datasets (time series data, formatted image data, text data, emotion data)

[1415] Processing: Data is fed into an AI model (e.g., a model implemented in TensorFlow or PyTorch) and current data is analyzed based on past training data. This allows for estimation of cognitive state and stress level. Emotion data obtained through emotion recognition methods is also included in the analysis.

[1416] Output: Analysis results (predictions of cognitive state, stress level, etc.)

[1417] Step 4: Result generation

[1418] The server determines and generates an appropriate countermeasure based on the analysis result using the result generation means.

[1419] Input: Analysis results (predicted results such as cognitive state, stress level, etc.)

[1420] Processing: If the cognitive state is deteriorating, it will determine measures to reduce stress, such as suggesting a relaxing walk or a relaxation session. It will also take into account emotional data to generate coping strategies.

[1421] Output: A message with the recommendations

[1422] Step 5: Notification Delivery

[1423] The server transmits the generated results and actions to the caregiver's device.

[1424] Input: Message with recommendations

[1425] Processing: The generated message is sent to the caregiver's device (smartphone or tablet).

[1426] Output: Notification message displayed on the caregiver's device

[1427] Step 6: Implement the recommendations

[1428] The user (caregiver) checks the notification and takes the recommended action.

[1429] Input: Notification message

[1430] Action: Review the notification and take the recommended action with the patient, for example, take a walk with the patient.

[1431] Output: Action taken

[1432] Step 7: Gather feedback

[1433] The user (caregiver) sends the results of the execution to the server as feedback.

[1434] Input: Execution results (e.g., changes in the patient's condition or emotions after the walk)

[1435] Processing: The results of the countermeasures are recorded via the device and sent to the server, which collects and analyzes the feedback data and retrains the AI ​​model.

[1436] Output: Feedback data is stored and used on the server

[1437] Throughout this entire process, the system analyzes the data from multiple angles and provides appropriate responses that take emotional data into account, thereby providing effective support for both caregivers and patients.

[1438] (Application example 2)

[1439] 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."

[1440] In factories and other workplaces, it is necessary to grasp the physical and mental health status of employees in real time and quickly provide appropriate countermeasures. However, currently there is a lack of systems that can accurately analyze workload and stress and suggest appropriate countermeasures. This presents a problem in that it is not possible to provide appropriate support to ensure employee efficiency and safety.

[1441] 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 a data collection means, a preprocessing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and an emotion analysis means. This makes it possible to analyze the physical and mental health conditions of workers in real time and quickly provide appropriate countermeasures.

[1442] The "data collection means" is a system element that collects vital data, work environment data, work image data, and voice data of workers.

[1443] A "preprocessing means" is an element of the system that converts and prepares collected data into a form suitable for analysis.

[1444] "Data analysis means" refers to the system element that inputs preprocessed data, including emotional data, into the AI ​​model for analysis.

[1445] A "result generation means" is an element of the system that determines the appropriate course of action based on the results of data analysis.

[1446] The "notification means" is a system element that sends the generated results and how to deal with them to the worker's device.

[1447] The "feedback collection means" is a system element that sends the results of the implemented countermeasures to the server and collects them.

[1448] "Emotion analysis means" is an element of the system that recognizes and analyzes emotions from the worker's facial expressions and voice data.

[1449] The present invention provides a system for monitoring and analyzing the health status and stress levels of employees in the workplace in real time, which includes a data collection means, a pre-processing means, a data analysis means, a result generation means, a notification means, a feedback collection means, and a sentiment analysis means.

[1450] First, the server collects the worker's vital data (e.g., heart rate, blood pressure), work environment data (e.g., temperature, humidity), work image data, and voice data through a data collection means. It also uses an emotion analysis means to recognize the worker's emotions from their facial expressions and voice data.

[1451] Next, the server uses preprocessing means to convert the collected data into a format suitable for analysis. This includes converting heart rate and blood pressure into time-series data, converting work images into a format suitable for image recognition, and converting voice data into text. Furthermore, emotion analysis means extracts emotions from the worker's voice and facial expressions and records them as data.

[1452] The server then uses data analysis means to input the preprocessed data into the AI ​​model, which analyzes the worker's health condition and stress level. The AI ​​model uses past learning data and analyzes current data to estimate stress levels and fatigue levels. By including emotional data obtained by the emotion analysis means in this analysis, detailed insights into the worker's mental state can be gained.

[1453] Based on the analysis results, the server uses the result generation means to determine an appropriate response. Specifically, if the stress level is determined to be high, the server generates a response such as recommending a break. Using data obtained through the emotion analysis means, the server can also suggest a response that takes emotions into consideration.

[1454] Next, the server sends the generated results and countermeasures to the worker's device (terminal). The terminal displays the received notification in real time. An example message might be, "Worker A's stress level is high, so we recommend that he take a 10-minute break."

[1455] The user (worker) then checks the notification received from the device and takes the recommended action. For example, the worker takes a 10-minute break in the break room. During this action, the emotion analysis means also monitors changes in the worker's emotions and collects further data.

[1456] Finally, the user (worker) sends the results of the implemented measures to the server as feedback. The server analyzes the collected feedback data and uses it for retraining to improve the accuracy of the AI ​​model. In particular, since this feedback includes emotional data, it will make the next recommendation more accurate.

[1457] As a specific example, at 7:00 AM, the server collects worker A's heart rate, sleep data, and current work environment data, while the emotion analysis means simultaneously analyzes the worker's morning tone of voice and facial expression. After data preprocessing, the AI ​​model analyzes this data and determines that worker A's heart rate is higher than normal and that he or she is highly fatigued. The result generation means generates a countermeasure such as, "Worker A's stress level is high, so we recommend that he or she take a 10-minute break." The device then sends a notification to the worker, and the user (worker) follows the recommendation and takes a break, providing feedback on the results. This feedback data is collected and analyzed by the server and used to further improve the system.

[1458] This system analyzes data from multiple angles and provides appropriate countermeasures that take emotional data into consideration, thereby effectively supporting both worker health and work efficiency.

[1459] Example prompt sentence:

[1460] Please explain the system that collects and analyzes the heart rate, blood pressure, facial expressions and voice data of factory workers, and recommends appropriate breaks and ways to deal with stress when the worker's stress level is high.

[1461] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1462] Step 1:

[1463] The server uses data collection means to collect worker vital data (heart rate, blood pressure), work environment data (temperature, humidity), work image data, and voice data. In this step, the input is real-time data obtained from various sensors and cameras, and the output is raw data compiled by the data collection means.

[1464] Step 2:

[1465] The server uses preprocessing means to convert and organize the collected data into a format suitable for analysis. Specifically, it converts heart rate and blood pressure into time-series data, converts work images into a format for image recognition, and converts voice data into text. The input to this step is the raw data collected in step 1, and the output is preprocessed data.

[1466] Step 3:

[1467] The server uses emotion analysis means to recognize and extract emotions from the worker's facial and voice data. The emotion data is used for subsequent analysis. The input for this step is facial and voice data obtained from a camera and microphone, and the output is emotion data recognized by the emotion analysis means.

[1468] Step 4:

[1469] The server uses data analysis means to input the preprocessed data and emotional data into the AI ​​model to analyze the worker's health condition and stress level. The AI ​​model performs analysis based on past learning data and makes an evaluation based on current data. The inputs for this step are the preprocessed data and emotional data, and the output is the analysis results.

[1470] Step 5:

[1471] The server uses the result generation means to determine an appropriate course of action based on the analysis results. For example, if the stress level is determined to be high, it generates a course of action such as recommending a break. The input of this step is the analysis result of step 4, and the output is a course of action.

[1472] Step 6:

[1473] The server uses the notification means to send the generated results and the countermeasures to the worker's terminal. The terminal displays the received notification to the worker in real time. The input of this step is information about the countermeasures, and the output is the notification sent to the worker's terminal.

[1474] Step 7:

[1475] The user (worker) checks the notification received from the terminal and takes the recommended action. For example, the user takes a specific action such as taking a 10-minute rest in the break room. The input of this step is the notification from the terminal, and the output is the action taken.

[1476] Step 8:

[1477] The user (operator) sends the results of the countermeasures they have taken as feedback to the server. The server analyzes the collected feedback data and uses it for retraining to improve the accuracy of the AI ​​model. The input of this step is the feedback information from the user, and the output is the data used for retraining.

[1478] 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.

[1479] 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.

[1480] 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.

[1481] 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.

[1482] 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.

[1483] 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.

[1484] 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).

[1485] 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.

[1486] 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."

[1487] 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.

[1488] 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).

[1489] 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.

[1490] 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.

[1491] 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.

[1492] 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.

[1493] 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.

[1494] 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.

[1495] 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.

[1496] 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.

[1497] 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.

[1498] 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.

[1499] The following is further disclosed regarding the above embodiment.

[1500] (Claim 1)

[1501] data collection means;

[1502] A pre-processing means;

[1503] data analysis means;

[1504] result generating means;

[1505] Notification means;

[1506] A system including a means of collecting feedback.

[1507] (Claim 2)

[1508] 2. The system according to claim 1, wherein the data collection means is means for collecting vital data, meteorological data, dietary image data, and audio data of the patient.

[1509] (Claim 3)

[1510] 2. The system according to claim 1, wherein the data analysis means is a means for inputting preprocessed data into an AI model and analyzing it.

[1511] "Example 1"

[1512] (Claim 1)

[1513] data collection means;

[1514] A pre-processing means;

[1515] data analysis means;

[1516] result generating means;

[1517] Notification means;

[1518] a feedback collection means;

[1519] means for collecting, preprocessing, and analyzing the patient's vital data, environmental data, nutritional data, and audio data to generate assessment results and measures;

[1520] means for notifying the generated evaluation results and countermeasures to the user device;

[1521] A system that includes a means to re-collect data after users have implemented countermeasures and use it to retrain the AI ​​model.

[1522] (Claim 2)

[1523] 2. The system of claim 1, wherein the data collection means is means for collecting vital data, environmental data, nutritional image data, and audio data of the patient.

[1524] (Claim 3)

[1525] The system according to claim 1, wherein the data analysis means is a means for inputting preprocessed data into an AI model for analysis, and further comprises a means for proposing countermeasures based on the generated analysis results.

[1526] "Application Example 1"

[1527] (Claim 1)

[1528] data collection means;

[1529] A pre-processing means;

[1530] data analysis means;

[1531] result generating means;

[1532] Notification means;

[1533] a feedback collection means;

[1534] means for collecting vital data, environmental data, and image data of a driver;

[1535] means for analyzing the data to generate recommendations based on the driver's state;

[1536] The system includes a means for communicating the recommendations to the driver's device.

[1537] (Claim 2)

[1538] 2. The system according to claim 1, wherein the data collection means is means for collecting vital data of the driver, weather data, and image data of the driver.

[1539] (Claim 3)

[1540] 2. The system according to claim 1, wherein the data analysis means is a means for inputting preprocessed data into a generative AI model and analyzing it.

[1541] "Example 2: Combining Emotion Engines"

[1542] (Claim 1)

[1543] data collection means;

[1544] A pre-processing means;

[1545] data analysis means;

[1546] result generating means;

[1547] Notification means;

[1548] a feedback collection means;

[1549] A system including an emotion recognition means.

[1550] (Claim 2)

[1551] 2. The system of claim 1, wherein the data collection means is means for collecting biometric data, environmental data, image data, and audio data of a patient.

[1552] (Claim 3)

[1553] 2. The system according to claim 1, wherein the data analysis means is a means for inputting the emotion data obtained from the emotion recognition means and the preprocessed data into an AI model and analyzing them.

[1554] "Application example 2 when combining emotion engines"

[1555] (Claim 1)

[1556] data collection means;

[1557] A pre-processing means;

[1558] data analysis means;

[1559] result generating means;

[1560] Notification means;

[1561] a feedback collection means;

[1562] A system including a means for sentiment analysis.

[1563] (Claim 2)

[1564] 2. The system according to claim 1, wherein the data collection means is means for collecting vital data, work environment data, work image data, and voice data of the worker.

[1565] (Claim 3)

[1566] 2. The system according to claim 1, wherein the data analysis means is a means for inputting preprocessed data including emotion data into an AI model and analyzing the data. [Explanation of symbols]

[1567] 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. data collection means; A pre-processing means; data analysis means; result generating means; Notification means; A system including a means of collecting feedback.

2. 2. The system according to claim 1, wherein said data collection means is means for collecting vital data, meteorological data, dietary image data, and audio data of a patient.

3. 2. The system according to claim 1, wherein the data analysis means is a means for inputting preprocessed data into an AI model and analyzing the data.

Images