System

Abstract

A system is provided.SOLUTION: A system, comprising: means for obtaining health data; means for preprocessing the obtained health data; means for training a machine learning model using the preprocessed health data; means for predicting a health risk of a user using the trained machine learning model; and means for notifying the user of the prediction result.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] Conventional medical systems often respond only after an illness has occurred, and early detection and prevention are insufficient. Furthermore, a lack of proactive health management is an issue for people with a family history of chronic illness. Therefore, there is a need for a system that can efficiently collect and analyze health data and provide appropriate health management guidance to users. [Means for solving the problem]

[0005] The present invention is a system including a means for acquiring health data, a means for preprocessing the acquired health data, a means for training a machine learning model using the preprocessed health data, a means for predicting a user's health risk using the trained machine learning model, and a means for notifying the user of the prediction result. The present invention further includes a means for imputing missing values ​​in the health data, a means for standardizing the imputed health data, a means for dividing a dataset into features and labels, and a means for dividing the dataset into training data and test data, thereby efficiently predicting a user's health risk and realizing early health management.

[0006] "Health Data" refers to data that includes physical and medical information about a user, such as age, weight, height, blood pressure, heart rate, and medical history.

[0007] "Preprocessing" refers to a series of processes performed on acquired health data, such as imputing missing values ​​and standardizing data, which are preparatory steps performed before analysis or training of machine learning models.

[0008] A "machine learning model" is an algorithm or network that uses large amounts of data to learn specific patterns and relationships and make predictions and classifications for unknown data.

[0009] "Training" is the process by which a machine learning model learns patterns and relationships using health data, a process that is carried out to improve the model's accuracy.

[0010] A means of predicting "health risk" is a method or process that uses a trained machine learning model to estimate a user's health status and future risk of disease.

[0011] "Notification" refers to the act of informing the user of the prediction results, and includes providing information that enables the user to take specific health management actions.

[0012] "Missing value imputation" is a process of filling in missing values ​​in health data in a specific way, and in many cases, the mean or median of the data is used.

[0013] "Standardization" is the process of converting data on different scales to a uniform scale by subtracting the mean of the data and dividing by the standard deviation.

[0014] A "feature" is an individual element or attribute of data that a machine learning model uses for prediction or classification, and in the case of health data, this includes age and weight.

[0015] A "label" is an output value or class corresponding to a feature, and is an indicator or target used by a machine learning model for training.

[0016] "Training data" is the portion of a dataset that a machine learning model uses to learn patterns.

[0017] "Test data" is a portion of a dataset used to evaluate the performance of a machine learning model; it is data that is separate from the training data and that the model has not learned. [Brief explanation of the drawings]

[0018] [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 illustrating 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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] The present invention provides a system for acquiring health data and predicting a user's health risk using a machine learning model. To implement this system, a server, a terminal, and a user work together.

[0040] Health data acquisition and preprocessing

[0041] The server acquires health data provided by the user. The acquired health data includes individual physical and medical information such as age, weight, height, blood pressure, heart rate, and medical history. The server then preprocesses this health data. Specifically, it imputes missing values ​​using the data's average value, and then standardizes the data. Standardization is the process of converting data to a certain scale.

[0042] Building and training machine learning models

[0043] After the health data is preprocessed, the server builds a machine learning model. The model is designed as a multi-layered neural network, with each layer having a different number of nodes. The built model is then trained using the health data. During this training process, the dataset is divided into features (such as age and weight) and labels (whether or not a person has a health risk), and then further divided into training data and test data. The model uses the training data to learn patterns and improve its prediction accuracy.

[0044] Predictions and Notifications

[0045] Using the trained model, the device inputs the user's health data and predicts health risks. Once the user's health data is input into the device, the device sends the data to a server, which then performs the prediction. The prediction results are returned to the device and ultimately notified to the user. This allows the user to understand their health risks and take proactive health management measures as needed.

[0046] Specific examples

[0047] For example, suppose a user has the following data: age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg. This user data is sent to the server via the device. The server preprocesses and standardizes this data, then inputs it into a trained machine learning model. The model predicts health risks based on this information and returns the prediction results to the device. The device then notifies the user of the prediction results. For example, if the health risk score is 0.7, the user is notified, "There is a health risk. We recommend that you seek medical advice." Conversely, if the score is 0.2, the user is notified, "Your health risk is low."

[0048] This system allows users to proactively manage their own health status and take appropriate measures in advance, enabling early detection and prevention that are difficult to achieve with conventional medical systems.

[0049] The processing flow will be explained below.

[0050] Step 1:

[0051] The server acquires health data. It reads physical and medical information provided by the user or medical institution from a file such as "health_data.csv."

[0052] Step 2:

[0053] The server preprocesses the acquired health data by imputing missing values ​​with the mean value of the data and then standardizing the data by subtracting the mean value and dividing by the standard deviation.

[0054] Step 3:

[0055] The server separates the preprocessed data into features and labels. Features include age, weight, height, blood pressure, etc., and labels include health risks (e.g., whether or not a disease is present).

[0056] Step 4:

[0057] The server splits the dataset into training data and test data, which are used to train the machine learning model and the test data, which are used to evaluate the model.

[0058] Step 5:

[0059] The server builds a machine learning model, which is designed as a multi-layer neural network with different numbers of nodes in each layer.

[0060] Step 6:

[0061] The server trains a machine learning model using the training data. Using the features and labels contained in the dataset, the model learns the patterns needed to predict health risks.

[0062] Step 7:

[0063] The device inputs the user's health data and sends it to the server. The user inputs health information such as age, weight, height, and blood pressure into the device.

[0064] Step 8:

[0065] The server receives the user's health data and uses a trained machine learning model to predict health risks, which are then calculated as a risk score.

[0066] Step 9:

[0067] The server sends the prediction results to the device, and generates a message based on the risk score to notify the user.

[0068] Step 10:

[0069] The user checks the prediction results on the device, receiving messages such as "There is a health risk. We recommend you consult a doctor" or "The health risk is low," and understanding their own health condition.

[0070] Example 1

[0071] 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."

[0072] With conventional health management systems, it was difficult for users to accurately assess their own health risks and take early action. Furthermore, complex processes such as data imputation, standardization, and training of machine learning models often required manual work, which was time-consuming and labor-intensive. In addition, delays in notification of prediction results meant that users were unable to respond promptly.

[0073] 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.

[0074] In this invention, the server includes a means for a user to input health data, a means for a terminal to transmit the health data to the server, a means for the server to acquire and preprocess the health data, a means for the server to train a machine learning model, a means for predicting the user's health risk using the trained machine learning model, and a means for notifying the user of the prediction result. This allows users to simply input their own health data, and the server automatically preprocesses and analyzes it, thereby enabling quick prediction and notification of health risks.

[0075] "Means for users to input health data" refers to the operating means by which users input health information such as their age, weight, height, blood pressure, heart rate, and medical history using a dedicated application or an input form on a device.

[0076] The "means by which the device transmits health data to the server" refers to the means by which the device transmits the input health data to the server using a data transmission method such as an API request.

[0077] The "means for the server to acquire and preprocess health data" refers to the means by which the server receives health data sent from the terminal and automatically performs preprocessing such as missing value completion and standardization of the data.

[0078] "Means for the server to train a machine learning model" refers to means for the server to use the preprocessed data to train a machine learning model, such as a multi-layer neural network, and learn health risk patterns from the data.

[0079] "Means for predicting a user's health risk using a trained machine learning model" means a means in which a server uses a trained model to automatically predict a user's health risk from input health data.

[0080] The "means for notifying the user of the prediction results" refers to a means by which the server or terminal visually or audibly notifies the user of the health risk prediction results obtained by the machine learning model.

[0081] "Means for completing missing values ​​in health data" refers to means for completing missing parts of health data acquired by the server using statistical methods, average value completion, etc.

[0082] "Means for standardizing imputed health data" refers to an operational means by which the server converts the imputed health data into a certain scale (e.g., a range of 0 to 1) in order to convert it into a format suitable for machine learning models.

[0083] "Means for dividing a dataset into features and labels" refers to the means by which the server classifies health data into features required for prediction (age, weight, etc.) and labels for the prediction target (presence or absence of health risk).

[0084] The "means for dividing into training data and test data" refers to the means by which the server divides the health data into data for training the model (learning data) and data for evaluating the model (test data).

[0085] MODE FOR CARRYING OUT THE INVENTION

[0086] The present invention provides a system for acquiring a user's health data and predicting health risks using a machine learning model. This system operates in cooperation with a server, a terminal, and a user.

[0087] The server retrieves the user's health data sent from the device. This health data includes information such as age, weight, height, blood pressure, heart rate, and medical history. The retrieved health data is preprocessed using "scikit-learn." Specifically, preprocessing involves filling in missing values ​​with the data's average value, etc., and then standardizing the data. Standardization is the process of converting the data to a certain scale to make it suitable for machine learning models.

[0088] The preprocessed data is then used to train a multi-layer neural network model built using TensorFlow. During the training process, the dataset is divided into features (e.g., age and weight) and labels (e.g., whether or not a person has a health risk), and then further divided into training data and test data. The server uses the training data to train the model, enabling it to predict health risks with high accuracy.

[0089] The device is responsible for transmitting health data entered by the user to the server. When the user enters health data through the device, the device sends the data to the server. The server uses the received data to predict health risks and returns the results to the device. The device notifies the user of the prediction results. This notification allows the user to understand their own health risks and proactively manage their health as necessary.

[0090] For example, a user has the following data:

[0091] Age: 25

[0092] Weight: 70kg

[0093] Height: 1.75m

[0094] Blood pressure: 120mmHg

[0095] When the user enters this data from the device, the device sends the data to the server. The server preprocesses the data and inputs it into a trained machine learning model to predict health risks. The prediction results are returned to the device, which then notifies the user. For example, if the health risk score is 0.7, the device will display the message "There is a health risk. We recommend that you seek medical advice." If the score is 0.2, the device will display the message "Your health risk is low."

[0096] This system allows users to understand their own health condition in real time and take appropriate measures early on, making early detection and prevention possible, something that was difficult to achieve with conventional medical systems.

[0097] An example of a prompt to input to a generative AI model would be:

[0098] "Predict the health risk of a user with the following health data: age: 25, weight: 70kg, height: 1.75m, blood pressure: 120mmHg."

[0099] This concrete example clearly demonstrates how users and related devices work together to implement the system, ensuring the smooth operation of a series of processes, including health data collection, pre-processing, model training, prediction, and result notification, throughout the system.

[0100] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0101] Step 1:

[0102] The user enters health data.

[0103] Specifically, the user uses a dedicated application or an input form on the device to enter health information such as their age, weight, height, blood pressure, heart rate, and medical history.

[0104] Inputs include health data such as age, weight, height, blood pressure, heart rate, and medical history.

[0105] As an output, the entered health data is displayed in the input field on the terminal.

[0106] Step 2:

[0107] The device sends the health data to the server.

[0108] Specifically, the device converts the data entered by the user into JSON format and makes an API request to the server.

[0109] The input is health data entered by the user into the terminal.

[0110] As output, health data in JSON format is sent to the server.

[0111] Step 3:

[0112] The server acquires and pre-processes the health data.

[0113] Specifically, the server complements missing values ​​in the acquired health data and standardizes the data.

[0114] The input is health data in JSON format sent from the device.

[0115] As an output, imputed and standardized health data is generated.

[0116] Step 4:

[0117] The server trains the machine learning model.

[0118] Specifically, the server uses the standardized data to train a model using a machine learning library (e.g., TensorFlow). The data is split into features and labels, and then further split into training data and test data.

[0119] The input is preprocessed health data.

[0120] The output is a trained machine learning model.

[0121] Step 5:

[0122] The device again sends the user's health data to the server and makes a prediction.

[0123] Specifically, the device again acquires the user's health data and sends it to the server, which uses this data to predict health risks.

[0124] The input is the user's health data.

[0125] As an output, the server returns the predicted health risk results.

[0126] Step 6:

[0127] The server predicts health risks and returns the results to the device.

[0128] Specifically, the server uses the trained model to analyze the health data, calculate a risk score, and send the result back to the device.

[0129] The input is the user's health data, again submitted.

[0130] As an output, a predicted health risk score is generated and sent back to the terminal.

[0131] Step 7:

[0132] The terminal notifies the user of the result.

[0133] Specifically, the device receives the prediction results from the server and displays them to the user, who is then notified of their health risk score and advice based on it.

[0134] The input is the predicted health risk score sent from the server.

[0135] As an output, the prediction results and advice are displayed on the terminal screen and notified to the user.

[0136] (Application example 1)

[0137] 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."

[0138] Conventional food delivery services do not suggest menus that take into account the user's health condition. This can lead to users with health risks choosing inappropriate meals, further worsening their health. Another problem is that it is difficult for users to understand their own health information and manage it themselves.

[0139] 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.

[0140] In this invention, the server includes means for acquiring health data, means for preprocessing the acquired health data, means for training a machine learning model using the preprocessed health data, means for predicting a user's health risk using the trained machine learning model, means for proposing a meal menu based on the health risk, and means for notifying the user of the prediction result. This makes it possible to propose a meal menu suitable for a user with health risks, allowing the user to select an appropriate meal while understanding their own health condition.

[0141] "Health Data" is physical and medical information about a user, such as age, weight, height, blood pressure, heart rate, and medical history.

[0142] "Preprocessing" refers to the process of complementing and standardizing missing values ​​in the acquired health data.

[0143] A "machine learning model" is an algorithm designed as a multi-layer neural network for predicting health risks based on health data.

[0144] "Training" is the process by which a machine learning model learns patterns based on health data and improves its prediction accuracy.

[0145] "Health risk" is an indicator that indicates the possibility that a user has a particular health problem and the degree of that risk.

[0146] "Meal menu suggestion" is the act of recommending appropriate meal choices based on the user's health risks.

[0147] "Notification" is the process of informing the user of prediction results and suggested meal menus.

[0148] "Missing value imputation" is the process of filling in missing information in a dataset using the average value of the data or other methods.

[0149] "Standardization" is the act of transforming data so that it is processed on a consistent scale.

[0150] A "feature" is an element used as input for a machine learning model, such as age or weight in a training dataset.

[0151] A "label" is correct data corresponding to input data, such as the presence or absence of a health risk predicted by a machine learning model.

[0152] "Training data" is a dataset used to train a machine learning model.

[0153] "Test data" is a dataset used to evaluate a trained machine learning model.

[0154] A "server" is a computer that acquires health data, preprocesses it, trains machine learning models, and performs prediction processing.

[0155] A "terminal" is a device through which a user inputs health data and receives prediction results and meal menu suggestions.

[0156] A system for realizing this invention involves a server and a terminal working together to acquire health data, preprocess it, train a machine learning model, make predictions, and provide notifications.

[0157] Server Processing

[0158] First, the server obtains health data from the user. This health data includes age, weight, height, blood pressure, heart rate, medical history, etc. The server then preprocesses this data. Specifically, it imputes missing values ​​using the data's average value, and then standardizes the data. Standardization is the process of converting data to a certain scale.

[0159] After the data is preprocessed, the server builds a machine learning model. This model is a multi-layered neural network, with each layer having a different number of nodes. The model is then trained using health data. During the training process, the dataset is divided into features (such as age and weight) and labels (whether or not a person has a health risk), and then further divided into training data and test data. The model uses this training data to learn patterns and improve its prediction accuracy.

[0160] Terminal handling

[0161] The device is used to input and transmit the user's health data. The user inputs their health data into the device, and the device transmits the data to the server. Once the server performs a prediction, the result is returned to the device.

[0162] Before notifying the user, a process is performed to suggest an appropriate meal menu based on the predicted health risk. For example, a balanced, low-calorie menu is suggested to a user with a high health risk, while a regular menu is suggested to a user with a low health risk.

[0163] Hardware and software used

[0164] Hardware: High-performance servers, users' smartphones

[0165] Software: Python, TensorFlow (machine learning library), NumPy, Scikit-learn (for data preprocessing)

[0166] Specific examples

[0167] For example, suppose a user enters the following health data: age: 29, weight: 68 kg, height: 1.80 m, blood pressure: 125 mmHg, heart rate: 70 bpm. This data is sent to the server via the device. The server preprocesses and standardizes the data, then inputs it into a trained machine learning model. The model predicts health risks based on this information and returns the prediction results to the device. The device then suggests a meal menu to the user based on the health risks, and finally notifies the user of the prediction results.

[0168] Prompt Sentence Examples

[0169] Based on the user's health data, suggest the following food menu:

[0170] Age: 29

[0171] Weight: 68kg

[0172] Height: 1.80m

[0173] Blood pressure: 125mmHg

[0174] Heart rate: 70 bpm

[0175] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0176] Step 1:

[0177] The user inputs health data into the device, including age, weight, height, blood pressure, heart rate, etc. This input data is temporarily stored inside the device.

[0178] Step 2:

[0179] The device sends the user's health data to the server. The input data is packaged in a data format such as JSON and sent to the server.

[0180] Step 3:

[0181] The server receives the received health data and imputes missing values. For example, if there are missing values ​​in the dataset, they are imputed with the average value of the data. At this point, data with the missing values ​​of the input data imputed is obtained.

[0182] Step 4:

[0183] The server standardizes the imputed health data. Specifically, it converts the data to a certain scale (e.g., mean 0, standard deviation 1), thereby obtaining standardized data.

[0184] Step 5:

[0185] The server uses the preprocessed health data to train a machine learning model. This process involves splitting the dataset into features (e.g., age, weight, height, etc.) and labels (e.g., whether or not a person has a health risk), and then splitting it into training data and test data. The model uses the training data to learn patterns.

[0186] Step 6:

[0187] The server inputs the user's health data and uses a trained machine learning model to predict health risks. Standardized user health data is input into the model, and a predicted result (e.g., a health risk score) is output.

[0188] Step 7:

[0189] The server then suggests appropriate meal plans based on the predicted health risk score: for a high risk score, a balanced, low-calorie menu is suggested; for a low risk score, a regular menu is suggested.

[0190] Step 8:

[0191] The prediction results and meal menu suggestions obtained from the server are sent back to the device, which receives the information and notifies the user. Ultimately, the user can check their own health risks and appropriate meal menus through the device.

[0192] Examples:

[0193] For example, a user enters the following health data: age: 29, weight: 68 kg, height: 1.80 m, blood pressure: 125 mmHg, heart rate: 70 bpm. This data is sent to the server in JSON format in step 2. The server imputes missing values ​​in step 3 and standardizes them in step 4. The preprocessed data is used to train a model in step 5, and health risk is predicted in step 6. For example, if the health risk score is 0.7, a balanced, low-calorie menu is suggested in step 7, and the user is notified in step 8.

[0194] Example prompt sentence:

[0195] Based on the user's health data, suggest the following food menu:

[0196] Age: 29

[0197] Weight: 68kg

[0198] Height: 1.80m

[0199] Blood pressure: 125mmHg

[0200] Heart rate: 70 bpm

[0201] 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.

[0202] The present invention combines a system that acquires health data and predicts a user's health risks using machine learning models with an emotion engine that recognizes the user's emotions. This system improves the accuracy of health risk predictions and provides customized feedback according to the user's emotional state.

[0203] Health data acquisition and preprocessing

[0204] The server acquires health data provided by users, including individual physical and medical information such as age, weight, height, blood pressure, heart rate, medical history, etc. After acquiring the data, the server performs preprocessing, including imputing and standardizing missing values.

[0205] Building and training machine learning models

[0206] The server separates the preprocessed data into features (e.g., age, weight, height, etc.) and labels (e.g., whether or not there is a health risk). The server then splits the dataset into training data and test data and builds a machine learning model. This model is designed as a multi-layer neural network and is trained to learn specific patterns.

[0207] Introducing the Emotion Engine

[0208] The emotion engine recognizes the user's emotional state. This engine collects and analyzes the user's emotional data using techniques such as voice analysis, facial expression analysis, or text analysis. The emotion engine then sends the results to the server.

[0209] Combining health risk prediction and emotions

[0210] The server combines the emotion data obtained from the emotion engine with the preprocessed health data to predict the user's health risk using a machine learning model. Through this process, the emotional state is reflected as additional information in the health risk assessment.

[0211] Notification and feedback of prediction results

[0212] The server sends the prediction results to the device, which then notifies the user. The generated message takes into account the user's emotional state, providing more personalized feedback. For example, even if the risk score is high, if the user is feeling depressed, an encouraging message such as "You need to be careful about your current health condition, but don't panic and consult a specialist" will be displayed.

[0213] Specific examples

[0214] For example, suppose a user has data on age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg, and the emotion engine detects stress from the user's tone of voice. This data is sent from the device to the server. The server preprocesses this information and inputs it into a trained machine learning model. The model predicts health risks and generates feedback based on the results and emotion data. For example, if the model predicts "moderate health risk" and the emotion engine detects "high stress," the device will notify the user, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[0215] This system allows for more detailed and individualized management of the user's health status, enabling early and appropriate treatment. It also takes into account the user's emotional state, providing customized support for the user, leading to more effective health management.

[0216] The processing flow will be explained below.

[0217] Step 1:

[0218] The server acquires health data, reading physical and medical information provided by users and medical institutions from files such as "health_data.csv."

[0219] Step 2:

[0220] The server preprocesses the acquired health data. Missing values ​​are filled with the mean value of the data, and then the data is standardized. Standardization is the process of subtracting the mean value of the data and dividing by the standard deviation.

[0221] Step 3:

[0222] The server separates the preprocessed data into features (age, weight, height, blood pressure, etc.) and labels (presence or absence of health risks), preparing the data in a format suitable for model training.

[0223] Step 4:

[0224] The server splits the dataset into training data and test data, which are used to train the machine learning model and evaluate the model's performance.

[0225] Step 5:

[0226] The server builds a machine learning model, designed as a multi-layer neural network with different numbers of nodes in each layer.

[0227] Step 6:

[0228] The server trains the machine learning model using the training data. Using the features and labels, the model learns the patterns needed to predict health risks.

[0229] Step 7:

[0230] The device inputs the user's health data and sends it to the server. For example, the user inputs health information such as age, weight, height, and blood pressure into the device.

[0231] Step 8:

[0232] The emotion engine recognizes the user's emotional state, collects emotion data using voice analysis, facial expression analysis, or text analysis, and sends the analysis results to the server.

[0233] Step 9:

[0234] The server combines the emotion data obtained from the emotion engine with the pre-processed health data and predicts the user's health risk using a trained machine learning model.

[0235] Step 10:

[0236] The server sends the prediction results to the device, and a customized message is generated based on the prediction results, taking into account the user's emotional state.

[0237] Step 11:

[0238] The device will notify the user of the prediction results, for example, by displaying a message such as, "Your current health condition requires attention, but please do not rush into anything and consult a specialist."

[0239] Step 12:

[0240] Users can check the prediction results on their device and take medical advice or lifestyle changes as needed, thus receiving personalized health advice that takes into account their emotional state.

[0241] Example 2

[0242] 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."

[0243] Many conventional health management systems collect user health data and predict health risks using machine learning models. However, they lack the ability to predict health risks taking into account the user's emotional state and provide feedback based on the results. This makes it difficult to improve prediction accuracy and provide customized responses to users.

[0244] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0245] In this invention, the server includes means for acquiring health data, means for preprocessing the acquired health data, means for training a machine learning model using the preprocessed health data, means for predicting a user's health risk using the trained machine learning model, an emotion engine for recognizing the user's emotional state, means for predicting health risk by combining the emotion data obtained from the emotion engine with the preprocessed health data, and means for notifying the user of the prediction result and providing customized feedback according to the user's emotional state. This enables more accurate health risk prediction and personalized feedback that takes the user's emotional state into consideration.

[0246] "Health Data" refers collectively to a user's individual physical and medical information, such as age, weight, height, blood pressure, heart rate, and medical history.

[0247] "Preprocessing" refers to the process of filling in missing values ​​and standardizing the acquired health data.

[0248] A "machine learning model" is an algorithmic structure that learns patterns based on data and makes predictions and classifications.

[0249] An "emotion engine" is a system that recognizes and analyzes user emotional data using techniques such as voice analysis, facial expression analysis, and text analysis.

[0250] "Emotion data" is information about the user's emotional state, specifically data about the user's emotions such as stress, joy, sadness, etc.

[0251] "Feedback" refers to messages of instruction or advice provided to the user based on the prediction results.

[0252] "Standardization" refers to the process of converting data to a certain scale to improve the efficiency of model training.

[0253] "Notification" refers to the act of informing the user of the prediction results, particularly via a terminal.

[0254] The present invention combines a system that acquires health data and predicts a user's health risks using machine learning models with an emotion engine that recognizes the user's emotions. This system improves the accuracy of health risk predictions and provides customized feedback according to the user's emotional state.

[0255] Health data acquisition and preprocessing

[0256] The server receives health data provided by users. This data includes physical and medical information such as age, weight, height, blood pressure, heart rate, and medical history. Users enter this information using a terminal, which then sends the data to the server. The server then performs preprocessing, including imputing missing values ​​and standardizing the data. Python's Pandas library is used for imputing missing values, and Scikit-learn's StandardScaler is used for standardizing.

[0257] Building and training machine learning models

[0258] The server builds a machine learning model using the preprocessed health data. The data is first divided into features and labels, and then split into training data and test data. A multi-layer neural network model is designed using TensorFlow and the model is trained using the training data.

[0259] Introducing the Emotion Engine

[0260] The emotion engine is used to recognize the user's emotional state. This engine collects and analyzes user emotion data using techniques such as speech analysis, facial expression analysis, and text analysis. For example, it uses the Google Cloud Speech-to-Text API to convert speech to text and performs sentiment analysis of the text using the Natural Language Toolkit (NLTK).

[0261] Combining health risk prediction and emotions

[0262] The server combines the emotion data obtained from the emotion engine with the preprocessed health data and uses a machine learning model to predict the user's health risk. By incorporating the emotion data into the health risk assessment, the accuracy of the prediction is improved.

[0263] Notification and feedback of prediction results

[0264] The server sends the prediction results to the device, which then generates a personalized message based on the predicted health risk and the user's emotional state and notifies the user. For example, even if the risk score is high, if the user is in a state of high stress, a message such as "You need to be careful about your current health condition, but don't rush and consult a specialist" will be displayed.

[0265] Specific examples

[0266] For example, suppose a user has data on age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg, and the emotion engine detects stress from the user's tone of voice. This data is sent from the device to the server. The server preprocesses this information and inputs it into a trained machine learning model. The model predicts health risks and generates feedback based on the results and emotion data. For example, if the model predicts "moderate health risk" and the emotion engine detects "high stress," the device will notify the user, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[0267] Example prompts for generative AI models

[0268] "Please predict the health risk of a user with the following data: age 25, weight 70kg, height 1.75m, and blood pressure 120mmHg. Also, high stress has been detected from the user's voice tone. Please generate a feedback message based on this."

[0269] This system allows users to manage their health condition in more detail and individually than ever before, enabling early and appropriate treatment. It also takes into account the user's emotional state, providing customized support for the user and realizing more effective health management.

[0270] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0271] Step 1:

[0272] The server collects health data provided by the user. The user uses a device to input information such as age, weight, height, blood pressure, heart rate, and medical history, and the device then sends this data to the server. The input health data is then aggregated on the server.

[0273] Step 2:

[0274] The server preprocesses the acquired health data. Specifically, it uses Python's Pandas library to impute missing values ​​in the data and standardizes the data using Scikit-learn's StandardScaler. The input is the acquired health data, and the output is the imputed and standardized health data.

[0275] Step 3:

[0276] The server uses the preprocessed data to build and train a machine learning model. It separates the data into features and labels, and splits it into training data and test data. It designs a multilayer neural network model using TensorFlow and trains the model using the training data. The input is the preprocessed health data, and the output is a trained machine learning model.

[0277] Step 4:

[0278] The emotion engine recognizes the user's emotional state. The device collects the user's voice data and converts it into text using voice analysis (for example, Google Cloud Speech-to-Text API). The text data is sent to a server, which performs emotion analysis using a library such as NLTK. The input is voice data, and the output is emotion data (for example, "high stress").

[0279] Step 5:

[0280] The server combines the emotion data with preprocessed health data and inputs it into a machine learning model to predict health risks. Including information about emotional states improves the accuracy of predictions. The inputs are emotion data and preprocessed health data, and the output is health risk prediction results.

[0281] Step 6:

[0282] The server sends the prediction results to the device. The device then generates a personalized message based on the predicted health risk and the user's emotional state, and notifies the user. For example, a message such as "There is a moderate health risk. Pay attention to stress management and try relaxation techniques" is displayed. The input is the health risk prediction result and emotional data, and the output is a notification message to the user.

[0283] This series of processes allows for more detailed management of the user's health status and provides feedback tailored to individual situations, enabling early and appropriate action to be taken.

[0284] (Application example 2)

[0285] 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."

[0286] Conventional health management systems predict health risks based solely on a user's health data, without taking the user's emotional state into account, resulting in poor prediction accuracy. Furthermore, because customized feedback based on the user's emotional state is not provided, it is difficult to provide appropriate advice. Therefore, there is a need for a system that can more accurately predict a user's health risks and provide feedback that takes their emotional state into account.

[0287] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring health data, means for pre-processing the acquired health data, means for training a machine learning model using the pre-processed health data, means for acquiring emotion data, means for analyzing the acquired emotion data, means for training a machine learning model by combining the pre-processed health data and the emotion data, means for predicting a user's health risk using the trained machine learning model, and means for notifying the user of the prediction result together with feedback customized based on the user's emotional state. This enables highly accurate health risk prediction that takes the user's emotional state into account and the provision of appropriate feedback that is individually customized.

[0288] "Health Data" refers to physical and medical information about a User, such as age, weight, height, blood pressure, heart rate, and medical history.

[0289] "Preprocessing" refers to the process of filling in missing values ​​in the acquired data and standardizing it to improve data quality.

[0290] A "machine learning model" is an algorithm that learns patterns and rules from massive data sets and makes predictions and classifications for new data.

[0291] "Emotional Data" refers to information related to a user's emotional state collected through speech analysis, facial expression analysis, and text analysis.

[0292] "Emotion engine" refers to a system that recognizes and analyzes a user's emotional state using voice analysis, facial expression analysis, or text analysis.

[0293] "Feedback" refers to advice or information provided to the user based on the system's predicted health risks.

[0294] "Customized feedback" refers to individually tailored advice or information that is tailored based on the user's emotional state.

[0295] "Features" refer to specific attributes or aspects of data that are input into a machine learning model, and include information that serves as the basis for classification and prediction.

[0296] "Labels" refer to the correct outcomes or categories of data used to train machine learning models.

[0297] "Missing value imputation" refers to the process of filling in missing data points with appropriate values.

[0298] "Standardization" refers to the process of standardizing the scale and distribution of data.

[0299] "Notification" refers to a message or information sent from the system to a user.

[0300] This invention is a system that acquires health data and predicts a user's health risks using a machine learning model. By combining it with an emotion engine that recognizes the user's emotional state, it provides even more accurate health risk predictions and personalized feedback.

[0301] System configuration

[0302] The system consists of the following main components:

[0303] 1. Means of collecting health data

[0304] The server obtains the user's physical and medical information, such as age, weight, height, blood pressure, heart rate, and medical history, from their smartphone or wearable device.

[0305] 2. Data preprocessing methods

[0306] The server preprocesses the acquired health data to improve the data quality, specifically by imputing missing values ​​and standardizing the data, using the Python Pandas library.

[0307] 3. Machine learning model training methods

[0308] The preprocessed data is used to train a machine learning model, here using a multi-layer neural network (MLPClassifier) ​​and the Scikit-learn library.

[0309] 4. Means of acquiring emotional data

[0310] The emotion engine is used to obtain emotion data by analyzing the user's voice, facial expression, and text. For example, the open source library LibROSA is used for voice data analysis.

[0311] 5. Sentiment Data Analysis Methods

[0312] We use natural language processing (NLP) techniques to analyze the acquired emotional data and recognize the user's emotional state. In particular, we use the SpaCy library to analyze text data.

[0313] 6. Machine learning model (emotion data combination) training method

[0314] The machine learning model is retrained by combining health data and emotion data as new features, which enables even more accurate prediction of health risks.

[0315] 7. Health risk prediction tools

[0316] A trained machine learning model is used to predict the user's health risks, and the predictions are made in real time, with the results sent from the server to the device.

[0317] 8. Feedback Notification Methods

[0318] The server notifies the user of the prediction results along with customized feedback based on the user's emotional state. For example, if the user is in a high stress state, the server provides feedback such as "Try some relaxation techniques."

[0319] Usage example

[0320] For example, a 30-year-old employee connects to the system using a smartphone and enters their health data. At the same time, their voice is also collected through the smartphone's microphone. The server preprocesses this data and analyzes their emotional data. The machine learning model combines this information to predict the employee's health risk and provide feedback. Specifically, the message displayed might read, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[0321] Example prompts for generative AI models

[0322] You can provide input to a generative AI model using prompt statements like the following:

[0323] This article introduces the functionality of a system that predicts health risks based on an employee's health data and emotional state. For example, consider an employee who is 30 years old, weighs 75 kg, is 1.80 m tall, has a blood pressure of 130 mmHg, and a heart rate of 80, and is in a high-stress situation. By inputting this information into the system, it can predict health risks and provide appropriate feedback. It also asks for specific prompts.

[0324] This system can predict users' health risks with greater accuracy than before and provide customized feedback that takes into account their emotional state, enabling more effective health management.

[0325] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0326] Step 1:

[0327] The server acquires health data from the user. The input includes information such as the user's age, weight, height, blood pressure, heart rate, and medical history. This data is collected from smartphones and wearable devices. The output is the acquired raw health data.

[0328] Step 2:

[0329] The server preprocesses the acquired health data. Here, missing values ​​in the data are imputed and standardized. Missing values ​​are generally imputed with the mean or median, and standardization is usually done by converting each feature to the same scale, usually with a mean of 0 and a standard deviation of 1. The input is the raw health data acquired in step 1, and the output is preprocessed health data.

[0330] Step 3:

[0331] The server trains a machine learning model using the preprocessed health data. It uses a multi-layer neural network (MLPClassifier) ​​and leverages the Scikit-learn library. The input includes the preprocessed data and its corresponding labels, and the output is a trained model.

[0332] Step 4:

[0333] The server uses an emotion engine to obtain the user's emotional data. Emotional data is collected through voice analysis, facial expression analysis, and text analysis. Specifically, data is collected from the smartphone's microphone and camera. The input includes the user's voice and image data, and the analyzed emotional data is obtained as the output.

[0334] Step 5:

[0335] The server analyzes the acquired emotional data. The emotion engine uses natural language processing (NLP) technology to recognize the user's emotional state from voice and text data. Specifically, it uses the SpaCy library. The input is emotional data, and the output is the emotional state resulting from the analysis.

[0336] Step 6:

[0337] The server combines the preprocessed health data and emotion data to create new features and retrain the machine learning model, enabling highly accurate health risk prediction that takes emotional states into account. The input includes the preprocessed health data and emotion data, and the output is the retrained model.

[0338] Step 7:

[0339] The server uses a trained machine learning model to predict the user's health risk. This prediction is performed in real time and uses the model obtained in step 6. The input includes current health data and emotion data, and the output is a predicted health risk result.

[0340] Step 8:

[0341] The server notifies the user of the prediction result along with customized feedback based on the user's emotional state. For example, if the risk is high and stress is high, a message such as "Try relaxation techniques" is displayed. The input includes the predicted health risk result and the user's emotional state, and the output is a customized feedback message.

[0342] 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.

[0343] 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.

[0344] 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.

[0345] [Second embodiment]

[0346] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0347] 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.

[0348] 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).

[0349] 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.

[0350] 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.

[0351] 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).

[0352] 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.

[0353] 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.

[0354] 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.

[0355] 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.

[0356] 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.

[0357] 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."

[0358] The present invention provides a system for acquiring health data and predicting a user's health risk using a machine learning model. To implement this system, a server, a terminal, and a user work together.

[0359] Health data acquisition and preprocessing

[0360] The server acquires health data provided by the user. The acquired health data includes individual physical and medical information such as age, weight, height, blood pressure, heart rate, and medical history. The server then preprocesses this health data. Specifically, it imputes missing values ​​using the data's average value, and then standardizes the data. Standardization is the process of converting data to a certain scale.

[0361] Building and training machine learning models

[0362] After the health data is preprocessed, the server builds a machine learning model. The model is designed as a multi-layered neural network, with each layer having a different number of nodes. The built model is then trained using the health data. During this training process, the dataset is divided into features (such as age and weight) and labels (whether or not a person has a health risk), and then further divided into training data and test data. The model uses the training data to learn patterns and improve its prediction accuracy.

[0363] Predictions and Notifications

[0364] Using the trained model, the device inputs the user's health data and predicts health risks. Once the user's health data is input into the device, the device sends the data to a server, which then performs the prediction. The prediction results are returned to the device and ultimately notified to the user. This allows the user to understand their health risks and take proactive health management measures as needed.

[0365] Specific examples

[0366] For example, suppose a user has the following data: age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg. This user data is sent to the server via the device. The server preprocesses and standardizes this data, then inputs it into a trained machine learning model. The model predicts health risks based on this information and returns the prediction results to the device. The device then notifies the user of the prediction results. For example, if the health risk score is 0.7, the user is notified, "There is a health risk. We recommend that you seek medical advice." Conversely, if the score is 0.2, the user is notified, "Your health risk is low."

[0367] This system allows users to proactively manage their own health status and take appropriate measures in advance, enabling early detection and prevention that are difficult to achieve with conventional medical systems.

[0368] The processing flow will be explained below.

[0369] Step 1:

[0370] The server acquires health data. It reads physical and medical information provided by the user or medical institution from a file such as "health_data.csv."

[0371] Step 2:

[0372] The server preprocesses the acquired health data by imputing missing values ​​with the mean value of the data and then standardizing the data by subtracting the mean value and dividing by the standard deviation.

[0373] Step 3:

[0374] The server separates the preprocessed data into features and labels. Features include age, weight, height, blood pressure, etc., and labels include health risks (e.g., whether or not a disease is present).

[0375] Step 4:

[0376] The server splits the dataset into training data and test data, which are used to train the machine learning model and the test data, which are used to evaluate the model.

[0377] Step 5:

[0378] The server builds a machine learning model, which is designed as a multi-layer neural network with different numbers of nodes in each layer.

[0379] Step 6:

[0380] The server trains a machine learning model using the training data. Using the features and labels contained in the dataset, the model learns the patterns needed to predict health risks.

[0381] Step 7:

[0382] The device inputs the user's health data and sends it to the server. The user inputs health information such as age, weight, height, and blood pressure into the device.

[0383] Step 8:

[0384] The server receives the user's health data and uses a trained machine learning model to predict health risks, which are then calculated as a risk score.

[0385] Step 9:

[0386] The server sends the prediction results to the device, and generates a message based on the risk score to notify the user.

[0387] Step 10:

[0388] The user checks the prediction results on the device, receiving messages such as "There is a health risk. We recommend you consult a doctor" or "The health risk is low," and understanding their own health condition.

[0389] Example 1

[0390] 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."

[0391] With conventional health management systems, it was difficult for users to accurately assess their own health risks and take early action. Furthermore, complex processes such as data imputation, standardization, and training of machine learning models often required manual work, which was time-consuming and labor-intensive. In addition, delays in notification of prediction results meant that users were unable to respond promptly.

[0392] 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.

[0393] In this invention, the server includes a means for a user to input health data, a means for a terminal to transmit the health data to the server, a means for the server to acquire and preprocess the health data, a means for the server to train a machine learning model, a means for predicting the user's health risk using the trained machine learning model, and a means for notifying the user of the prediction result. This allows users to simply input their own health data, and the server automatically preprocesses and analyzes it, thereby enabling quick prediction and notification of health risks.

[0394] "Means for users to input health data" refers to the operating means by which users input health information such as their age, weight, height, blood pressure, heart rate, and medical history using a dedicated application or an input form on a device.

[0395] The "means by which the device transmits health data to the server" refers to the means by which the device transmits the input health data to the server using a data transmission method such as an API request.

[0396] The "means for the server to acquire and preprocess health data" refers to the means by which the server receives health data sent from the terminal and automatically performs preprocessing such as missing value completion and standardization of the data.

[0397] "Means for the server to train a machine learning model" refers to means for the server to use the preprocessed data to train a machine learning model, such as a multi-layer neural network, and learn health risk patterns from the data.

[0398] "Means for predicting a user's health risk using a trained machine learning model" means a means in which a server uses a trained model to automatically predict a user's health risk from input health data.

[0399] The "means for notifying the user of the prediction results" refers to a means by which the server or terminal visually or audibly notifies the user of the health risk prediction results obtained by the machine learning model.

[0400] "Means for completing missing values ​​in health data" refers to means for completing missing parts of health data acquired by the server using statistical methods, average value completion, etc.

[0401] "Means for standardizing imputed health data" refers to an operational means by which the server converts the imputed health data into a certain scale (e.g., a range of 0 to 1) in order to convert it into a format suitable for machine learning models.

[0402] "Means for dividing a dataset into features and labels" refers to the means by which the server classifies health data into features required for prediction (age, weight, etc.) and labels for the prediction target (presence or absence of health risk).

[0403] The "means for dividing into training data and test data" refers to the means by which the server divides the health data into data for training the model (learning data) and data for evaluating the model (test data).

[0404] MODE FOR CARRYING OUT THE INVENTION

[0405] The present invention provides a system for acquiring a user's health data and predicting health risks using a machine learning model. This system operates in cooperation with a server, a terminal, and a user.

[0406] The server retrieves the user's health data sent from the device. This health data includes information such as age, weight, height, blood pressure, heart rate, and medical history. The retrieved health data is preprocessed using "scikit-learn." Specifically, preprocessing involves filling in missing values ​​with the data's average value, etc., and then standardizing the data. Standardization is the process of converting the data to a certain scale to make it suitable for machine learning models.

[0407] The preprocessed data is then used to train a multi-layer neural network model built using TensorFlow. During the training process, the dataset is divided into features (e.g., age and weight) and labels (e.g., whether or not a person has a health risk), and then further divided into training data and test data. The server uses the training data to train the model, enabling it to predict health risks with high accuracy.

[0408] The device is responsible for transmitting health data entered by the user to the server. When the user enters health data through the device, the device sends the data to the server. The server uses the received data to predict health risks and returns the results to the device. The device notifies the user of the prediction results. This notification allows the user to understand their own health risks and proactively manage their health as necessary.

[0409] For example, a user has the following data:

[0410] Age: 25

[0411] Weight: 70kg

[0412] Height: 1.75m

[0413] Blood pressure: 120mmHg

[0414] When the user enters this data from the device, the device sends the data to the server. The server preprocesses the data and inputs it into a trained machine learning model to predict health risks. The prediction results are returned to the device, which then notifies the user. For example, if the health risk score is 0.7, the device will display the message "There is a health risk. We recommend that you seek medical advice." If the score is 0.2, the device will display the message "Your health risk is low."

[0415] This system allows users to understand their own health condition in real time and take appropriate measures early on, making early detection and prevention possible, something that was difficult to achieve with conventional medical systems.

[0416] An example of a prompt to input to a generative AI model would be:

[0417] "Predict the health risk of a user with the following health data: age: 25, weight: 70kg, height: 1.75m, blood pressure: 120mmHg."

[0418] This concrete example clearly demonstrates how users and related devices work together to implement the system, ensuring the smooth operation of a series of processes, including health data collection, pre-processing, model training, prediction, and result notification, throughout the system.

[0419] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0420] Step 1:

[0421] The user enters health data.

[0422] Specifically, the user uses a dedicated application or an input form on the device to enter health information such as their age, weight, height, blood pressure, heart rate, and medical history.

[0423] Inputs include health data such as age, weight, height, blood pressure, heart rate, and medical history.

[0424] As an output, the entered health data is displayed in the input field on the terminal.

[0425] Step 2:

[0426] The device sends the health data to the server.

[0427] Specifically, the device converts the data entered by the user into JSON format and makes an API request to the server.

[0428] The input is health data entered by the user into the terminal.

[0429] As output, health data in JSON format is sent to the server.

[0430] Step 3:

[0431] The server acquires and pre-processes the health data.

[0432] Specifically, the server complements missing values ​​in the acquired health data and standardizes the data.

[0433] The input is health data in JSON format sent from the device.

[0434] As an output, imputed and standardized health data is generated.

[0435] Step 4:

[0436] The server trains the machine learning model.

[0437] Specifically, the server uses the standardized data to train a model using a machine learning library (e.g., TensorFlow). The data is split into features and labels, and then further split into training data and test data.

[0438] The input is preprocessed health data.

[0439] The output is a trained machine learning model.

[0440] Step 5:

[0441] The device again sends the user's health data to the server and makes a prediction.

[0442] Specifically, the device again acquires the user's health data and sends it to the server, which uses this data to predict health risks.

[0443] The input is the user's health data.

[0444] As an output, the server returns the predicted health risk results.

[0445] Step 6:

[0446] The server predicts health risks and returns the results to the device.

[0447] Specifically, the server uses the trained model to analyze the health data, calculate a risk score, and send the result back to the device.

[0448] The input is the user's health data, again submitted.

[0449] As an output, a predicted health risk score is generated and sent back to the terminal.

[0450] Step 7:

[0451] The terminal notifies the user of the result.

[0452] Specifically, the device receives the prediction results from the server and displays them to the user, who is then notified of their health risk score and advice based on it.

[0453] The input is the predicted health risk score sent from the server.

[0454] As an output, the prediction results and advice are displayed on the terminal screen and notified to the user.

[0455] (Application example 1)

[0456] 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."

[0457] Conventional food delivery services do not suggest menus that take into account the user's health condition. This can lead to users with health risks choosing inappropriate meals, further worsening their health. Another problem is that it is difficult for users to understand their own health information and manage it themselves.

[0458] 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.

[0459] In this invention, the server includes means for acquiring health data, means for preprocessing the acquired health data, means for training a machine learning model using the preprocessed health data, means for predicting a user's health risk using the trained machine learning model, means for proposing a meal menu based on the health risk, and means for notifying the user of the prediction result. This makes it possible to propose a meal menu suitable for a user with health risks, allowing the user to select an appropriate meal while understanding their own health condition.

[0460] "Health Data" is physical and medical information about a user, such as age, weight, height, blood pressure, heart rate, and medical history.

[0461] "Preprocessing" refers to the process of complementing and standardizing missing values ​​in the acquired health data.

[0462] A "machine learning model" is an algorithm designed as a multi-layer neural network for predicting health risks based on health data.

[0463] "Training" is the process by which a machine learning model learns patterns based on health data and improves its prediction accuracy.

[0464] "Health risk" is an indicator that indicates the possibility that a user has a particular health problem and the degree of that risk.

[0465] "Meal menu suggestion" is the act of recommending appropriate meal choices based on the user's health risks.

[0466] "Notification" is the process of informing the user of prediction results and suggested meal menus.

[0467] "Missing value imputation" is the process of filling in missing information in a dataset using the average value of the data or other methods.

[0468] "Standardization" is the act of transforming data so that it is processed on a consistent scale.

[0469] A "feature" is an element used as input for a machine learning model, such as age or weight in a training dataset.

[0470] A "label" is correct data corresponding to input data, such as the presence or absence of a health risk predicted by a machine learning model.

[0471] "Training data" is a dataset used to train a machine learning model.

[0472] "Test data" is a dataset used to evaluate a trained machine learning model.

[0473] A "server" is a computer that acquires health data, preprocesses it, trains machine learning models, and performs prediction processing.

[0474] A "terminal" is a device through which a user inputs health data and receives prediction results and meal menu suggestions.

[0475] A system for realizing this invention involves a server and a terminal working together to acquire health data, preprocess it, train a machine learning model, make predictions, and provide notifications.

[0476] Server Processing

[0477] First, the server obtains health data from the user. This health data includes age, weight, height, blood pressure, heart rate, medical history, etc. The server then preprocesses this data. Specifically, it imputes missing values ​​using the data's average value, and then standardizes the data. Standardization is the process of converting data to a certain scale.

[0478] After the data is preprocessed, the server builds a machine learning model. This model is a multi-layered neural network, with each layer having a different number of nodes. The model is then trained using health data. During the training process, the dataset is divided into features (such as age and weight) and labels (whether or not a person has a health risk), and then further divided into training data and test data. The model uses this training data to learn patterns and improve its prediction accuracy.

[0479] Terminal handling

[0480] The device is used to input and transmit the user's health data. The user inputs their health data into the device, and the device transmits the data to the server. Once the server performs a prediction, the result is returned to the device.

[0481] Before notifying the user, a process is performed to suggest an appropriate meal menu based on the predicted health risk. For example, a balanced, low-calorie menu is suggested to a user with a high health risk, while a regular menu is suggested to a user with a low health risk.

[0482] Hardware and software used

[0483] Hardware: High-performance servers, users' smartphones

[0484] Software: Python, TensorFlow (machine learning library), NumPy, Scikit-learn (for data preprocessing)

[0485] Specific examples

[0486] For example, suppose a user enters the following health data: age: 29, weight: 68 kg, height: 1.80 m, blood pressure: 125 mmHg, heart rate: 70 bpm. This data is sent to the server via the device. The server preprocesses and standardizes the data, then inputs it into a trained machine learning model. The model predicts health risks based on this information and returns the prediction results to the device. The device then suggests a meal menu to the user based on the health risks, and finally notifies the user of the prediction results.

[0487] Prompt Sentence Examples

[0488] Based on the user's health data, suggest the following food menu:

[0489] Age: 29

[0490] Weight: 68kg

[0491] Height: 1.80m

[0492] Blood pressure: 125mmHg

[0493] Heart rate: 70 bpm

[0494] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0495] Step 1:

[0496] The user inputs health data into the device, including age, weight, height, blood pressure, heart rate, etc. This input data is temporarily stored inside the device.

[0497] Step 2:

[0498] The device sends the user's health data to the server. The input data is packaged in a data format such as JSON and sent to the server.

[0499] Step 3:

[0500] The server receives the received health data and imputes missing values. For example, if there are missing values ​​in the dataset, they are imputed with the average value of the data. At this point, data with the missing values ​​of the input data imputed is obtained.

[0501] Step 4:

[0502] The server standardizes the imputed health data. Specifically, it converts the data to a certain scale (e.g., mean 0, standard deviation 1), thereby obtaining standardized data.

[0503] Step 5:

[0504] The server uses the preprocessed health data to train a machine learning model. This process involves splitting the dataset into features (e.g., age, weight, height, etc.) and labels (e.g., whether or not a person has a health risk), and then splitting it into training data and test data. The model uses the training data to learn patterns.

[0505] Step 6:

[0506] The server inputs the user's health data and uses a trained machine learning model to predict health risks. Standardized user health data is input into the model, and a predicted result (e.g., a health risk score) is output.

[0507] Step 7:

[0508] The server then suggests appropriate meal plans based on the predicted health risk score: for a high risk score, a balanced, low-calorie menu is suggested; for a low risk score, a regular menu is suggested.

[0509] Step 8:

[0510] The prediction results and meal menu suggestions obtained from the server are sent back to the device, which receives the information and notifies the user. Ultimately, the user can check their own health risks and appropriate meal menus through the device.

[0511] Examples:

[0512] For example, a user enters the following health data: age: 29, weight: 68 kg, height: 1.80 m, blood pressure: 125 mmHg, heart rate: 70 bpm. This data is sent to the server in JSON format in step 2. The server imputes missing values ​​in step 3 and standardizes them in step 4. The preprocessed data is used to train a model in step 5, and health risk is predicted in step 6. For example, if the health risk score is 0.7, a balanced, low-calorie menu is suggested in step 7, and the user is notified in step 8.

[0513] Example prompt sentence:

[0514] Based on the user's health data, suggest the following food menu:

[0515] Age: 29

[0516] Weight: 68kg

[0517] Height: 1.80m

[0518] Blood pressure: 125mmHg

[0519] Heart rate: 70 bpm

[0520] 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.

[0521] The present invention combines a system that acquires health data and predicts a user's health risks using machine learning models with an emotion engine that recognizes the user's emotions. This system improves the accuracy of health risk predictions and provides customized feedback according to the user's emotional state.

[0522] Health data acquisition and preprocessing

[0523] The server acquires health data provided by users, including individual physical and medical information such as age, weight, height, blood pressure, heart rate, medical history, etc. After acquiring the data, the server performs preprocessing, including imputing and standardizing missing values.

[0524] Building and training machine learning models

[0525] The server separates the preprocessed data into features (e.g., age, weight, height, etc.) and labels (e.g., whether or not there is a health risk). The server then splits the dataset into training data and test data and builds a machine learning model. This model is designed as a multi-layer neural network and is trained to learn specific patterns.

[0526] Introducing the Emotion Engine

[0527] The emotion engine recognizes the user's emotional state. This engine collects and analyzes the user's emotional data using techniques such as voice analysis, facial expression analysis, or text analysis. The emotion engine then sends the results to the server.

[0528] Combining health risk prediction and emotions

[0529] The server combines the emotion data obtained from the emotion engine with the preprocessed health data to predict the user's health risk using a machine learning model. Through this process, the emotional state is reflected as additional information in the health risk assessment.

[0530] Notification and feedback of prediction results

[0531] The server sends the prediction results to the device, which then notifies the user. The generated message takes into account the user's emotional state, providing more personalized feedback. For example, even if the risk score is high, if the user is feeling depressed, an encouraging message such as "You need to be careful about your current health condition, but don't panic and consult a specialist" will be displayed.

[0532] Specific examples

[0533] For example, suppose a user has data on age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg, and the emotion engine detects stress from the user's tone of voice. This data is sent from the device to the server. The server preprocesses this information and inputs it into a trained machine learning model. The model predicts health risks and generates feedback based on the results and emotion data. For example, if the model predicts "moderate health risk" and the emotion engine detects "high stress," the device will notify the user, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[0534] This system allows for more detailed and individualized management of the user's health status, enabling early and appropriate treatment. It also takes into account the user's emotional state, providing customized support for the user, leading to more effective health management.

[0535] The processing flow will be explained below.

[0536] Step 1:

[0537] The server acquires health data, reading physical and medical information provided by users and medical institutions from files such as "health_data.csv."

[0538] Step 2:

[0539] The server preprocesses the acquired health data. Missing values ​​are filled with the mean value of the data, and then the data is standardized. Standardization is the process of subtracting the mean value of the data and dividing by the standard deviation.

[0540] Step 3:

[0541] The server separates the preprocessed data into features (age, weight, height, blood pressure, etc.) and labels (presence or absence of health risks), preparing the data in a format suitable for model training.

[0542] Step 4:

[0543] The server splits the dataset into training data and test data, which are used to train the machine learning model and evaluate the model's performance.

[0544] Step 5:

[0545] The server builds a machine learning model, designed as a multi-layer neural network with different numbers of nodes in each layer.

[0546] Step 6:

[0547] The server trains the machine learning model using the training data. Using the features and labels, the model learns the patterns needed to predict health risks.

[0548] Step 7:

[0549] The device inputs the user's health data and sends it to the server. For example, the user inputs health information such as age, weight, height, and blood pressure into the device.

[0550] Step 8:

[0551] The emotion engine recognizes the user's emotional state, collects emotion data using voice analysis, facial expression analysis, or text analysis, and sends the analysis results to the server.

[0552] Step 9:

[0553] The server combines the emotion data obtained from the emotion engine with the pre-processed health data and predicts the user's health risk using a trained machine learning model.

[0554] Step 10:

[0555] The server sends the prediction results to the device, and a customized message is generated based on the prediction results, taking into account the user's emotional state.

[0556] Step 11:

[0557] The device will notify the user of the prediction results, for example, by displaying a message such as, "Your current health condition requires attention, but please do not rush into anything and consult a specialist."

[0558] Step 12:

[0559] Users can check the prediction results on their device and take medical advice or lifestyle changes as needed, thus receiving personalized health advice that takes into account their emotional state.

[0560] Example 2

[0561] 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."

[0562] Many conventional health management systems collect user health data and predict health risks using machine learning models. However, they lack the ability to predict health risks taking into account the user's emotional state and provide feedback based on the results. This makes it difficult to improve prediction accuracy and provide customized responses to users.

[0563] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0564] In this invention, the server includes means for acquiring health data, means for preprocessing the acquired health data, means for training a machine learning model using the preprocessed health data, means for predicting a user's health risk using the trained machine learning model, an emotion engine for recognizing the user's emotional state, means for predicting health risk by combining the emotion data obtained from the emotion engine with the preprocessed health data, and means for notifying the user of the prediction result and providing customized feedback according to the user's emotional state. This enables more accurate health risk prediction and personalized feedback that takes the user's emotional state into consideration.

[0565] "Health Data" refers collectively to a user's individual physical and medical information, such as age, weight, height, blood pressure, heart rate, and medical history.

[0566] "Preprocessing" refers to the process of filling in missing values ​​and standardizing the acquired health data.

[0567] A "machine learning model" is an algorithmic structure that learns patterns based on data and makes predictions and classifications.

[0568] An "emotion engine" is a system that recognizes and analyzes user emotional data using techniques such as voice analysis, facial expression analysis, and text analysis.

[0569] "Emotion data" is information about the user's emotional state, specifically data about the user's emotions such as stress, joy, sadness, etc.

[0570] "Feedback" refers to messages of instruction or advice provided to the user based on the prediction results.

[0571] "Standardization" refers to the process of converting data to a certain scale to improve the efficiency of model training.

[0572] "Notification" refers to the act of informing the user of the prediction results, particularly via a terminal.

[0573] The present invention combines a system that acquires health data and predicts a user's health risks using machine learning models with an emotion engine that recognizes the user's emotions. This system improves the accuracy of health risk predictions and provides customized feedback according to the user's emotional state.

[0574] Health data acquisition and preprocessing

[0575] The server receives health data provided by users. This data includes physical and medical information such as age, weight, height, blood pressure, heart rate, and medical history. Users enter this information using a terminal, which then sends the data to the server. The server then performs preprocessing, including imputing missing values ​​and standardizing the data. Python's Pandas library is used for imputing missing values, and Scikit-learn's StandardScaler is used for standardizing.

[0576] Building and training machine learning models

[0577] The server builds a machine learning model using the preprocessed health data. The data is first divided into features and labels, and then split into training data and test data. A multi-layer neural network model is designed using TensorFlow and the model is trained using the training data.

[0578] Introducing the Emotion Engine

[0579] The emotion engine is used to recognize the user's emotional state. This engine collects and analyzes user emotion data using techniques such as speech analysis, facial expression analysis, and text analysis. For example, it uses the Google Cloud Speech-to-Text API to convert speech to text and performs sentiment analysis of the text using the Natural Language Toolkit (NLTK).

[0580] Combining health risk prediction and emotions

[0581] The server combines the emotion data obtained from the emotion engine with the preprocessed health data and uses a machine learning model to predict the user's health risk. By incorporating the emotion data into the health risk assessment, the accuracy of the prediction is improved.

[0582] Notification and feedback of prediction results

[0583] The server sends the prediction results to the device, which then generates a personalized message based on the predicted health risk and the user's emotional state and notifies the user. For example, even if the risk score is high, if the user is in a state of high stress, a message such as "You need to be careful about your current health condition, but don't rush and consult a specialist" will be displayed.

[0584] Specific examples

[0585] For example, suppose a user has data on age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg, and the emotion engine detects stress from the user's tone of voice. This data is sent from the device to the server. The server preprocesses this information and inputs it into a trained machine learning model. The model predicts health risks and generates feedback based on the results and emotion data. For example, if the model predicts "moderate health risk" and the emotion engine detects "high stress," the device will notify the user, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[0586] Example prompts for generative AI models

[0587] "Please predict the health risk of a user with the following data: age 25, weight 70kg, height 1.75m, and blood pressure 120mmHg. Also, high stress has been detected from the user's voice tone. Please generate a feedback message based on this."

[0588] This system allows users to manage their health condition in more detail and individually than ever before, enabling early and appropriate treatment. It also takes into account the user's emotional state, providing customized support for the user and realizing more effective health management.

[0589] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0590] Step 1:

[0591] The server collects health data provided by the user. The user uses a device to input information such as age, weight, height, blood pressure, heart rate, and medical history, and the device then sends this data to the server. The input health data is then aggregated on the server.

[0592] Step 2:

[0593] The server preprocesses the acquired health data. Specifically, it uses Python's Pandas library to impute missing values ​​in the data and standardizes the data using Scikit-learn's StandardScaler. The input is the acquired health data, and the output is the imputed and standardized health data.

[0594] Step 3:

[0595] The server uses the preprocessed data to build and train a machine learning model. It separates the data into features and labels, and splits it into training data and test data. It designs a multilayer neural network model using TensorFlow and trains the model using the training data. The input is the preprocessed health data, and the output is a trained machine learning model.

[0596] Step 4:

[0597] The emotion engine recognizes the user's emotional state. The device collects the user's voice data and converts it into text using voice analysis (for example, Google Cloud Speech-to-Text API). The text data is sent to a server, which performs emotion analysis using a library such as NLTK. The input is voice data, and the output is emotion data (for example, "high stress").

[0598] Step 5:

[0599] The server combines the emotion data with preprocessed health data and inputs it into a machine learning model to predict health risks. Including information about emotional states improves the accuracy of predictions. The inputs are emotion data and preprocessed health data, and the output is health risk prediction results.

[0600] Step 6:

[0601] The server sends the prediction results to the device. The device then generates a personalized message based on the predicted health risk and the user's emotional state, and notifies the user. For example, a message such as "There is a moderate health risk. Pay attention to stress management and try relaxation techniques" is displayed. The input is the health risk prediction result and emotional data, and the output is a notification message to the user.

[0602] This series of processes allows for more detailed management of the user's health status and provides feedback tailored to individual situations, enabling early and appropriate action to be taken.

[0603] (Application example 2)

[0604] 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."

[0605] Conventional health management systems predict health risks based solely on a user's health data, without taking the user's emotional state into account, resulting in poor prediction accuracy. Furthermore, because customized feedback based on the user's emotional state is not provided, it is difficult to provide appropriate advice. Therefore, there is a need for a system that can more accurately predict a user's health risks and provide feedback that takes their emotional state into account.

[0606] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring health data, means for pre-processing the acquired health data, means for training a machine learning model using the pre-processed health data, means for acquiring emotion data, means for analyzing the acquired emotion data, means for training a machine learning model by combining the pre-processed health data and the emotion data, means for predicting a user's health risk using the trained machine learning model, and means for notifying the user of the prediction result together with feedback customized based on the user's emotional state. This enables highly accurate health risk prediction that takes the user's emotional state into account and the provision of appropriate feedback that is individually customized.

[0607] "Health Data" refers to physical and medical information about a User, such as age, weight, height, blood pressure, heart rate, and medical history.

[0608] "Preprocessing" refers to the process of filling in missing values ​​in the acquired data and standardizing it to improve data quality.

[0609] A "machine learning model" is an algorithm that learns patterns and rules from massive data sets and makes predictions and classifications for new data.

[0610] "Emotional Data" refers to information related to a user's emotional state collected through speech analysis, facial expression analysis, and text analysis.

[0611] "Emotion engine" refers to a system that recognizes and analyzes a user's emotional state using voice analysis, facial expression analysis, or text analysis.

[0612] "Feedback" refers to advice or information provided to the user based on the system's predicted health risks.

[0613] "Customized feedback" refers to individually tailored advice or information that is tailored based on the user's emotional state.

[0614] "Features" refer to specific attributes or aspects of data that are input into a machine learning model, and include information that serves as the basis for classification and prediction.

[0615] "Labels" refer to the correct outcomes or categories of data used to train machine learning models.

[0616] "Missing value imputation" refers to the process of filling in missing data points with appropriate values.

[0617] "Standardization" refers to the process of standardizing the scale and distribution of data.

[0618] "Notification" refers to a message or information sent from the system to a user.

[0619] This invention is a system that acquires health data and predicts a user's health risks using a machine learning model. By combining it with an emotion engine that recognizes the user's emotional state, it provides even more accurate health risk predictions and personalized feedback.

[0620] System configuration

[0621] The system consists of the following main components:

[0622] 1. Means of collecting health data

[0623] The server obtains the user's physical and medical information, such as age, weight, height, blood pressure, heart rate, and medical history, from their smartphone or wearable device.

[0624] 2. Data preprocessing methods

[0625] The server preprocesses the acquired health data to improve the data quality, specifically by imputing missing values ​​and standardizing the data, using the Python Pandas library.

[0626] 3. Machine learning model training methods

[0627] The preprocessed data is used to train a machine learning model, here using a multi-layer neural network (MLPClassifier) ​​and the Scikit-learn library.

[0628] 4. Means of acquiring emotional data

[0629] The emotion engine is used to obtain emotion data by analyzing the user's voice, facial expression, and text. For example, the open source library LibROSA is used for voice data analysis.

[0630] 5. Sentiment Data Analysis Methods

[0631] We use natural language processing (NLP) techniques to analyze the acquired emotional data and recognize the user's emotional state. In particular, we use the SpaCy library to analyze text data.

[0632] 6. Machine learning model (emotion data combination) training method

[0633] The machine learning model is retrained by combining health data and emotion data as new features, which enables even more accurate prediction of health risks.

[0634] 7. Health risk prediction tools

[0635] A trained machine learning model is used to predict the user's health risks, and the predictions are made in real time, with the results sent from the server to the device.

[0636] 8. Feedback Notification Methods

[0637] The server notifies the user of the prediction results along with customized feedback based on the user's emotional state. For example, if the user is in a high stress state, the server provides feedback such as "Try some relaxation techniques."

[0638] Usage example

[0639] For example, a 30-year-old employee connects to the system using a smartphone and enters their health data. At the same time, their voice is also collected through the smartphone's microphone. The server preprocesses this data and analyzes their emotional data. The machine learning model combines this information to predict the employee's health risk and provide feedback. Specifically, the message displayed might read, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[0640] Example prompts for generative AI models

[0641] You can provide input to a generative AI model using prompt statements like the following:

[0642] This article introduces the functionality of a system that predicts health risks based on an employee's health data and emotional state. For example, consider an employee who is 30 years old, weighs 75 kg, is 1.80 m tall, has a blood pressure of 130 mmHg, and a heart rate of 80, and is in a high-stress situation. By inputting this information into the system, it can predict health risks and provide appropriate feedback. It also asks for specific prompts.

[0643] This system can predict users' health risks with greater accuracy than before and provide customized feedback that takes into account their emotional state, enabling more effective health management.

[0644] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0645] Step 1:

[0646] The server acquires health data from the user. The input includes information such as the user's age, weight, height, blood pressure, heart rate, and medical history. This data is collected from smartphones and wearable devices. The output is the acquired raw health data.

[0647] Step 2:

[0648] The server preprocesses the acquired health data. Here, missing values ​​in the data are imputed and standardized. Missing values ​​are generally imputed with the mean or median, and standardization is usually done by converting each feature to the same scale, usually with a mean of 0 and a standard deviation of 1. The input is the raw health data acquired in step 1, and the output is preprocessed health data.

[0649] Step 3:

[0650] The server trains a machine learning model using the preprocessed health data. It uses a multi-layer neural network (MLPClassifier) ​​and leverages the Scikit-learn library. The input includes the preprocessed data and its corresponding labels, and the output is a trained model.

[0651] Step 4:

[0652] The server uses an emotion engine to obtain the user's emotional data. Emotional data is collected through voice analysis, facial expression analysis, and text analysis. Specifically, data is collected from the smartphone's microphone and camera. The input includes the user's voice and image data, and the analyzed emotional data is obtained as the output.

[0653] Step 5:

[0654] The server analyzes the acquired emotional data. The emotion engine uses natural language processing (NLP) technology to recognize the user's emotional state from voice and text data. Specifically, it uses the SpaCy library. The input is emotional data, and the output is the emotional state resulting from the analysis.

[0655] Step 6:

[0656] The server combines the preprocessed health data and emotion data to create new features and retrain the machine learning model, enabling highly accurate health risk prediction that takes emotional states into account. The input includes the preprocessed health data and emotion data, and the output is the retrained model.

[0657] Step 7:

[0658] The server uses a trained machine learning model to predict the user's health risk. This prediction is performed in real time and uses the model obtained in step 6. The input includes current health data and emotion data, and the output is a predicted health risk result.

[0659] Step 8:

[0660] The server notifies the user of the prediction result along with customized feedback based on the user's emotional state. For example, if the risk is high and stress is high, a message such as "Try relaxation techniques" is displayed. The input includes the predicted health risk result and the user's emotional state, and the output is a customized feedback message.

[0661] 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.

[0662] 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.

[0663] 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.

[0664] [Third embodiment]

[0665] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0666] 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.

[0667] 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).

[0668] 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.

[0669] 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.

[0670] 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).

[0671] 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.

[0672] 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.

[0673] 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.

[0674] 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.

[0675] 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.

[0676] 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."

[0677] The present invention provides a system for acquiring health data and predicting a user's health risk using a machine learning model. To implement this system, a server, a terminal, and a user work together.

[0678] Health data acquisition and preprocessing

[0679] The server acquires health data provided by the user. The acquired health data includes individual physical and medical information such as age, weight, height, blood pressure, heart rate, and medical history. The server then preprocesses this health data. Specifically, it imputes missing values ​​using the data's average value, and then standardizes the data. Standardization is the process of converting data to a certain scale.

[0680] Building and training machine learning models

[0681] After the health data is preprocessed, the server builds a machine learning model. The model is designed as a multi-layered neural network, with each layer having a different number of nodes. The built model is then trained using the health data. During this training process, the dataset is divided into features (such as age and weight) and labels (whether or not a person has a health risk), and then further divided into training data and test data. The model uses the training data to learn patterns and improve its prediction accuracy.

[0682] Predictions and Notifications

[0683] Using the trained model, the device inputs the user's health data and predicts health risks. Once the user's health data is input into the device, the device sends the data to a server, which then performs the prediction. The prediction results are returned to the device and ultimately notified to the user. This allows the user to understand their health risks and take proactive health management measures as needed.

[0684] Specific examples

[0685] For example, suppose a user has the following data: age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg. This user data is sent to the server via the device. The server preprocesses and standardizes this data, then inputs it into a trained machine learning model. The model predicts health risks based on this information and returns the prediction results to the device. The device then notifies the user of the prediction results. For example, if the health risk score is 0.7, the user is notified, "There is a health risk. We recommend that you seek medical advice." Conversely, if the score is 0.2, the user is notified, "Your health risk is low."

[0686] This system allows users to proactively manage their own health status and take appropriate measures in advance, enabling early detection and prevention that are difficult to achieve with conventional medical systems.

[0687] The processing flow will be explained below.

[0688] Step 1:

[0689] The server acquires health data. It reads physical and medical information provided by the user or medical institution from a file such as "health_data.csv."

[0690] Step 2:

[0691] The server preprocesses the acquired health data by imputing missing values ​​with the mean value of the data and then standardizing the data by subtracting the mean value and dividing by the standard deviation.

[0692] Step 3:

[0693] The server separates the preprocessed data into features and labels. Features include age, weight, height, blood pressure, etc., and labels include health risks (e.g., whether or not a disease is present).

[0694] Step 4:

[0695] The server splits the dataset into training data and test data, which are used to train the machine learning model and the test data, which are used to evaluate the model.

[0696] Step 5:

[0697] The server builds a machine learning model, which is designed as a multi-layer neural network with different numbers of nodes in each layer.

[0698] Step 6:

[0699] The server trains a machine learning model using the training data. Using the features and labels contained in the dataset, the model learns the patterns needed to predict health risks.

[0700] Step 7:

[0701] The device inputs the user's health data and sends it to the server. The user inputs health information such as age, weight, height, and blood pressure into the device.

[0702] Step 8:

[0703] The server receives the user's health data and uses a trained machine learning model to predict health risks, which are then calculated as a risk score.

[0704] Step 9:

[0705] The server sends the prediction results to the device, and generates a message based on the risk score to notify the user.

[0706] Step 10:

[0707] The user checks the prediction results on the device, receiving messages such as "There is a health risk. We recommend you consult a doctor" or "The health risk is low," and understanding their own health condition.

[0708] Example 1

[0709] 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."

[0710] With conventional health management systems, it was difficult for users to accurately assess their own health risks and take early action. Furthermore, complex processes such as data imputation, standardization, and training of machine learning models often required manual work, which was time-consuming and labor-intensive. In addition, delays in notification of prediction results meant that users were unable to respond promptly.

[0711] 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.

[0712] In this invention, the server includes a means for a user to input health data, a means for a terminal to transmit the health data to the server, a means for the server to acquire and preprocess the health data, a means for the server to train a machine learning model, a means for predicting the user's health risk using the trained machine learning model, and a means for notifying the user of the prediction result. This allows users to simply input their own health data, and the server automatically preprocesses and analyzes it, thereby enabling quick prediction and notification of health risks.

[0713] "Means for users to input health data" refers to the operating means by which users input health information such as their age, weight, height, blood pressure, heart rate, and medical history using a dedicated application or an input form on a device.

[0714] The "means by which the device transmits health data to the server" refers to the means by which the device transmits the input health data to the server using a data transmission method such as an API request.

[0715] The "means for the server to acquire and preprocess health data" refers to the means by which the server receives health data sent from the terminal and automatically performs preprocessing such as missing value completion and standardization of the data.

[0716] "Means for the server to train a machine learning model" refers to means for the server to use the preprocessed data to train a machine learning model, such as a multi-layer neural network, and learn health risk patterns from the data.

[0717] "Means for predicting a user's health risk using a trained machine learning model" means a means in which a server uses a trained model to automatically predict a user's health risk from input health data.

[0718] The "means for notifying the user of the prediction results" refers to a means by which the server or terminal visually or audibly notifies the user of the health risk prediction results obtained by the machine learning model.

[0719] "Means for completing missing values ​​in health data" refers to means for completing missing parts of health data acquired by the server using statistical methods, average value completion, etc.

[0720] "Means for standardizing imputed health data" refers to an operational means by which the server converts the imputed health data into a certain scale (e.g., a range of 0 to 1) in order to convert it into a format suitable for machine learning models.

[0721] "Means for dividing a dataset into features and labels" refers to the means by which the server classifies health data into features required for prediction (age, weight, etc.) and labels for the prediction target (presence or absence of health risk).

[0722] The "means for dividing into training data and test data" refers to the means by which the server divides the health data into data for training the model (learning data) and data for evaluating the model (test data).

[0723] MODE FOR CARRYING OUT THE INVENTION

[0724] The present invention provides a system for acquiring a user's health data and predicting health risks using a machine learning model. This system operates in cooperation with a server, a terminal, and a user.

[0725] The server retrieves the user's health data sent from the device. This health data includes information such as age, weight, height, blood pressure, heart rate, and medical history. The retrieved health data is preprocessed using "scikit-learn." Specifically, preprocessing involves filling in missing values ​​with the data's average value, etc., and then standardizing the data. Standardization is the process of converting the data to a certain scale to make it suitable for machine learning models.

[0726] The preprocessed data is then used to train a multi-layer neural network model built using TensorFlow. During the training process, the dataset is divided into features (e.g., age and weight) and labels (e.g., whether or not a person has a health risk), and then further divided into training data and test data. The server uses the training data to train the model, enabling it to predict health risks with high accuracy.

[0727] The device is responsible for transmitting health data entered by the user to the server. When the user enters health data through the device, the device sends the data to the server. The server uses the received data to predict health risks and returns the results to the device. The device notifies the user of the prediction results. This notification allows the user to understand their own health risks and proactively manage their health as necessary.

[0728] For example, a user has the following data:

[0729] Age: 25

[0730] Weight: 70kg

[0731] Height: 1.75m

[0732] Blood pressure: 120mmHg

[0733] When the user enters this data from the device, the device sends the data to the server. The server preprocesses the data and inputs it into a trained machine learning model to predict health risks. The prediction results are returned to the device, which then notifies the user. For example, if the health risk score is 0.7, the device will display the message "There is a health risk. We recommend that you seek medical advice." If the score is 0.2, the device will display the message "Your health risk is low."

[0734] This system allows users to understand their own health condition in real time and take appropriate measures early on, making early detection and prevention possible, something that was difficult to achieve with conventional medical systems.

[0735] An example of a prompt to input to a generative AI model would be:

[0736] "Predict the health risk of a user with the following health data: age: 25, weight: 70kg, height: 1.75m, blood pressure: 120mmHg."

[0737] This concrete example clearly demonstrates how users and related devices work together to implement the system, ensuring the smooth operation of a series of processes, including health data collection, pre-processing, model training, prediction, and result notification, throughout the system.

[0738] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0739] Step 1:

[0740] The user enters health data.

[0741] Specifically, the user uses a dedicated application or an input form on the device to enter health information such as their age, weight, height, blood pressure, heart rate, and medical history.

[0742] Inputs include health data such as age, weight, height, blood pressure, heart rate, and medical history.

[0743] As an output, the entered health data is displayed in the input field on the terminal.

[0744] Step 2:

[0745] The device sends the health data to the server.

[0746] Specifically, the device converts the data entered by the user into JSON format and makes an API request to the server.

[0747] The input is health data entered by the user into the terminal.

[0748] As output, health data in JSON format is sent to the server.

[0749] Step 3:

[0750] The server acquires and pre-processes the health data.

[0751] Specifically, the server complements missing values ​​in the acquired health data and standardizes the data.

[0752] The input is health data in JSON format sent from the device.

[0753] As an output, imputed and standardized health data is generated.

[0754] Step 4:

[0755] The server trains the machine learning model.

[0756] Specifically, the server uses the standardized data to train a model using a machine learning library (e.g., TensorFlow). The data is split into features and labels, and then further split into training data and test data.

[0757] The input is preprocessed health data.

[0758] The output is a trained machine learning model.

[0759] Step 5:

[0760] The device again sends the user's health data to the server and makes a prediction.

[0761] Specifically, the device again acquires the user's health data and sends it to the server, which uses this data to predict health risks.

[0762] The input is the user's health data.

[0763] As an output, the server returns the predicted health risk results.

[0764] Step 6:

[0765] The server predicts health risks and returns the results to the device.

[0766] Specifically, the server uses the trained model to analyze the health data, calculate a risk score, and send the result back to the device.

[0767] The input is the user's health data, again submitted.

[0768] As an output, a predicted health risk score is generated and sent back to the terminal.

[0769] Step 7:

[0770] The terminal notifies the user of the result.

[0771] Specifically, the device receives the prediction results from the server and displays them to the user, who is then notified of their health risk score and advice based on it.

[0772] The input is the predicted health risk score sent from the server.

[0773] As an output, the prediction results and advice are displayed on the terminal screen and notified to the user.

[0774] (Application example 1)

[0775] 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."

[0776] Conventional food delivery services do not suggest menus that take into account the user's health condition. This can lead to users with health risks choosing inappropriate meals, further worsening their health. Another problem is that it is difficult for users to understand their own health information and manage it themselves.

[0777] 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.

[0778] In this invention, the server includes means for acquiring health data, means for preprocessing the acquired health data, means for training a machine learning model using the preprocessed health data, means for predicting a user's health risk using the trained machine learning model, means for proposing a meal menu based on the health risk, and means for notifying the user of the prediction result. This makes it possible to propose a meal menu suitable for a user with health risks, allowing the user to select an appropriate meal while understanding their own health condition.

[0779] "Health Data" is physical and medical information about a user, such as age, weight, height, blood pressure, heart rate, and medical history.

[0780] "Preprocessing" refers to the process of complementing and standardizing missing values ​​in the acquired health data.

[0781] A "machine learning model" is an algorithm designed as a multi-layer neural network for predicting health risks based on health data.

[0782] "Training" is the process by which a machine learning model learns patterns based on health data and improves its prediction accuracy.

[0783] "Health risk" is an indicator that indicates the possibility that a user has a particular health problem and the degree of that risk.

[0784] "Meal menu suggestion" is the act of recommending appropriate meal choices based on the user's health risks.

[0785] "Notification" is the process of informing the user of prediction results and suggested meal menus.

[0786] "Missing value imputation" is the process of filling in missing information in a dataset using the average value of the data or other methods.

[0787] "Standardization" is the act of transforming data so that it is processed on a consistent scale.

[0788] A "feature" is an element used as input for a machine learning model, such as age or weight in a training dataset.

[0789] A "label" is correct data corresponding to input data, such as the presence or absence of a health risk predicted by a machine learning model.

[0790] "Training data" is a dataset used to train a machine learning model.

[0791] "Test data" is a dataset used to evaluate a trained machine learning model.

[0792] A "server" is a computer that acquires health data, preprocesses it, trains machine learning models, and performs prediction processing.

[0793] A "terminal" is a device through which a user inputs health data and receives prediction results and meal menu suggestions.

[0794] A system for realizing this invention involves a server and a terminal working together to acquire health data, preprocess it, train a machine learning model, make predictions, and provide notifications.

[0795] Server Processing

[0796] First, the server obtains health data from the user. This health data includes age, weight, height, blood pressure, heart rate, medical history, etc. The server then preprocesses this data. Specifically, it imputes missing values ​​using the data's average value, and then standardizes the data. Standardization is the process of converting data to a certain scale.

[0797] After the data is preprocessed, the server builds a machine learning model. This model is a multi-layered neural network, with each layer having a different number of nodes. The model is then trained using health data. During the training process, the dataset is divided into features (such as age and weight) and labels (whether or not a person has a health risk), and then further divided into training data and test data. The model uses this training data to learn patterns and improve its prediction accuracy.

[0798] Terminal handling

[0799] The device is used to input and transmit the user's health data. The user inputs their health data into the device, and the device transmits the data to the server. Once the server performs a prediction, the result is returned to the device.

[0800] Before notifying the user, a process is performed to suggest an appropriate meal menu based on the predicted health risk. For example, a balanced, low-calorie menu is suggested to a user with a high health risk, while a regular menu is suggested to a user with a low health risk.

[0801] Hardware and software used

[0802] Hardware: High-performance servers, users' smartphones

[0803] Software: Python, TensorFlow (machine learning library), NumPy, Scikit-learn (for data preprocessing)

[0804] Specific examples

[0805] For example, suppose a user enters the following health data: age: 29, weight: 68 kg, height: 1.80 m, blood pressure: 125 mmHg, heart rate: 70 bpm. This data is sent to the server via the device. The server preprocesses and standardizes the data, then inputs it into a trained machine learning model. The model predicts health risks based on this information and returns the prediction results to the device. The device then suggests a meal menu to the user based on the health risks, and finally notifies the user of the prediction results.

[0806] Prompt Sentence Examples

[0807] Based on the user's health data, suggest the following food menu:

[0808] Age: 29

[0809] Weight: 68kg

[0810] Height: 1.80m

[0811] Blood pressure: 125mmHg

[0812] Heart rate: 70 bpm

[0813] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0814] Step 1:

[0815] The user inputs health data into the device, including age, weight, height, blood pressure, heart rate, etc. This input data is temporarily stored inside the device.

[0816] Step 2:

[0817] The device sends the user's health data to the server. The input data is packaged in a data format such as JSON and sent to the server.

[0818] Step 3:

[0819] The server receives the received health data and imputes missing values. For example, if there are missing values ​​in the dataset, they are imputed with the average value of the data. At this point, data with the missing values ​​of the input data imputed is obtained.

[0820] Step 4:

[0821] The server standardizes the imputed health data. Specifically, it converts the data to a certain scale (e.g., mean 0, standard deviation 1), thereby obtaining standardized data.

[0822] Step 5:

[0823] The server uses the preprocessed health data to train a machine learning model. This process involves splitting the dataset into features (e.g., age, weight, height, etc.) and labels (e.g., whether or not a person has a health risk), and then splitting it into training data and test data. The model uses the training data to learn patterns.

[0824] Step 6:

[0825] The server inputs the user's health data and uses a trained machine learning model to predict health risks. Standardized user health data is input into the model, and a predicted result (e.g., a health risk score) is output.

[0826] Step 7:

[0827] The server then suggests appropriate meal plans based on the predicted health risk score: for a high risk score, a balanced, low-calorie menu is suggested; for a low risk score, a regular menu is suggested.

[0828] Step 8:

[0829] The prediction results and meal menu suggestions obtained from the server are sent back to the device, which receives the information and notifies the user. Ultimately, the user can check their own health risks and appropriate meal menus through the device.

[0830] Examples:

[0831] For example, a user enters the following health data: age: 29, weight: 68 kg, height: 1.80 m, blood pressure: 125 mmHg, heart rate: 70 bpm. This data is sent to the server in JSON format in step 2. The server imputes missing values ​​in step 3 and standardizes them in step 4. The preprocessed data is used to train a model in step 5, and health risk is predicted in step 6. For example, if the health risk score is 0.7, a balanced, low-calorie menu is suggested in step 7, and the user is notified in step 8.

[0832] Example prompt sentence:

[0833] Based on the user's health data, suggest the following food menu:

[0834] Age: 29

[0835] Weight: 68kg

[0836] Height: 1.80m

[0837] Blood pressure: 125mmHg

[0838] Heart rate: 70 bpm

[0839] 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.

[0840] The present invention combines a system that acquires health data and predicts a user's health risks using machine learning models with an emotion engine that recognizes the user's emotions. This system improves the accuracy of health risk predictions and provides customized feedback according to the user's emotional state.

[0841] Health data acquisition and preprocessing

[0842] The server acquires health data provided by users, including individual physical and medical information such as age, weight, height, blood pressure, heart rate, medical history, etc. After acquiring the data, the server performs preprocessing, including imputing and standardizing missing values.

[0843] Building and training machine learning models

[0844] The server separates the preprocessed data into features (e.g., age, weight, height, etc.) and labels (e.g., whether or not there is a health risk). The server then splits the dataset into training data and test data and builds a machine learning model. This model is designed as a multi-layer neural network and is trained to learn specific patterns.

[0845] Introducing the Emotion Engine

[0846] The emotion engine recognizes the user's emotional state. This engine collects and analyzes the user's emotional data using techniques such as voice analysis, facial expression analysis, or text analysis. The emotion engine then sends the results to the server.

[0847] Combining health risk prediction and emotions

[0848] The server combines the emotion data obtained from the emotion engine with the preprocessed health data to predict the user's health risk using a machine learning model. Through this process, the emotional state is reflected as additional information in the health risk assessment.

[0849] Notification and feedback of prediction results

[0850] The server sends the prediction results to the device, which then notifies the user. The generated message takes into account the user's emotional state, providing more personalized feedback. For example, even if the risk score is high, if the user is feeling depressed, an encouraging message such as "You need to be careful about your current health condition, but don't panic and consult a specialist" will be displayed.

[0851] Specific examples

[0852] For example, suppose a user has data on age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg, and the emotion engine detects stress from the user's tone of voice. This data is sent from the device to the server. The server preprocesses this information and inputs it into a trained machine learning model. The model predicts health risks and generates feedback based on the results and emotion data. For example, if the model predicts "moderate health risk" and the emotion engine detects "high stress," the device will notify the user, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[0853] This system allows for more detailed and individualized management of the user's health status, enabling early and appropriate treatment. It also takes into account the user's emotional state, providing customized support for the user, leading to more effective health management.

[0854] The processing flow will be explained below.

[0855] Step 1:

[0856] The server acquires health data, reading physical and medical information provided by users and medical institutions from files such as "health_data.csv."

[0857] Step 2:

[0858] The server preprocesses the acquired health data. Missing values ​​are filled with the mean value of the data, and then the data is standardized. Standardization is the process of subtracting the mean value of the data and dividing by the standard deviation.

[0859] Step 3:

[0860] The server separates the preprocessed data into features (age, weight, height, blood pressure, etc.) and labels (presence or absence of health risks), preparing the data in a format suitable for model training.

[0861] Step 4:

[0862] The server splits the dataset into training data and test data, which are used to train the machine learning model and evaluate the model's performance.

[0863] Step 5:

[0864] The server builds a machine learning model, designed as a multi-layer neural network with different numbers of nodes in each layer.

[0865] Step 6:

[0866] The server trains the machine learning model using the training data. Using the features and labels, the model learns the patterns needed to predict health risks.

[0867] Step 7:

[0868] The device inputs the user's health data and sends it to the server. For example, the user inputs health information such as age, weight, height, and blood pressure into the device.

[0869] Step 8:

[0870] The emotion engine recognizes the user's emotional state, collects emotion data using voice analysis, facial expression analysis, or text analysis, and sends the analysis results to the server.

[0871] Step 9:

[0872] The server combines the emotion data obtained from the emotion engine with the pre-processed health data and predicts the user's health risk using a trained machine learning model.

[0873] Step 10:

[0874] The server sends the prediction results to the device, and a customized message is generated based on the prediction results, taking into account the user's emotional state.

[0875] Step 11:

[0876] The device will notify the user of the prediction results, for example, by displaying a message such as, "Your current health condition requires attention, but please do not rush into anything and consult a specialist."

[0877] Step 12:

[0878] Users can check the prediction results on their device and take medical advice or lifestyle changes as needed, thus receiving personalized health advice that takes into account their emotional state.

[0879] Example 2

[0880] 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."

[0881] Many conventional health management systems collect user health data and predict health risks using machine learning models. However, they lack the ability to predict health risks taking into account the user's emotional state and provide feedback based on the results. This makes it difficult to improve prediction accuracy and provide customized responses to users.

[0882] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0883] In this invention, the server includes means for acquiring health data, means for preprocessing the acquired health data, means for training a machine learning model using the preprocessed health data, means for predicting a user's health risk using the trained machine learning model, an emotion engine for recognizing the user's emotional state, means for predicting health risk by combining the emotion data obtained from the emotion engine with the preprocessed health data, and means for notifying the user of the prediction result and providing customized feedback according to the user's emotional state. This enables more accurate health risk prediction and personalized feedback that takes the user's emotional state into consideration.

[0884] "Health Data" refers collectively to a user's individual physical and medical information, such as age, weight, height, blood pressure, heart rate, and medical history.

[0885] "Preprocessing" refers to the process of filling in missing values ​​and standardizing the acquired health data.

[0886] A "machine learning model" is an algorithmic structure that learns patterns based on data and makes predictions and classifications.

[0887] An "emotion engine" is a system that recognizes and analyzes user emotional data using techniques such as voice analysis, facial expression analysis, and text analysis.

[0888] "Emotion data" is information about the user's emotional state, specifically data about the user's emotions such as stress, joy, sadness, etc.

[0889] "Feedback" refers to messages of instruction or advice provided to the user based on the prediction results.

[0890] "Standardization" refers to the process of converting data to a certain scale to improve the efficiency of model training.

[0891] "Notification" refers to the act of informing the user of the prediction results, particularly via a terminal.

[0892] The present invention combines a system that acquires health data and predicts a user's health risks using machine learning models with an emotion engine that recognizes the user's emotions. This system improves the accuracy of health risk predictions and provides customized feedback according to the user's emotional state.

[0893] Health data acquisition and preprocessing

[0894] The server receives health data provided by users. This data includes physical and medical information such as age, weight, height, blood pressure, heart rate, and medical history. Users enter this information using a terminal, which then sends the data to the server. The server then performs preprocessing, including imputing missing values ​​and standardizing the data. Python's Pandas library is used for imputing missing values, and Scikit-learn's StandardScaler is used for standardizing.

[0895] Building and training machine learning models

[0896] The server builds a machine learning model using the preprocessed health data. The data is first divided into features and labels, and then split into training data and test data. A multi-layer neural network model is designed using TensorFlow and the model is trained using the training data.

[0897] Introducing the Emotion Engine

[0898] The emotion engine is used to recognize the user's emotional state. This engine collects and analyzes user emotion data using techniques such as speech analysis, facial expression analysis, and text analysis. For example, it uses the Google Cloud Speech-to-Text API to convert speech to text and performs sentiment analysis of the text using the Natural Language Toolkit (NLTK).

[0899] Combining health risk prediction and emotions

[0900] The server combines the emotion data obtained from the emotion engine with the preprocessed health data and uses a machine learning model to predict the user's health risk. By incorporating the emotion data into the health risk assessment, the accuracy of the prediction is improved.

[0901] Notification and feedback of prediction results

[0902] The server sends the prediction results to the device, which then generates a personalized message based on the predicted health risk and the user's emotional state and notifies the user. For example, even if the risk score is high, if the user is in a state of high stress, a message such as "You need to be careful about your current health condition, but don't rush and consult a specialist" will be displayed.

[0903] Specific examples

[0904] For example, suppose a user has data on age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg, and the emotion engine detects stress from the user's tone of voice. This data is sent from the device to the server. The server preprocesses this information and inputs it into a trained machine learning model. The model predicts health risks and generates feedback based on the results and emotion data. For example, if the model predicts "moderate health risk" and the emotion engine detects "high stress," the device will notify the user, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[0905] Example prompts for generative AI models

[0906] "Please predict the health risk of a user with the following data: age 25, weight 70kg, height 1.75m, and blood pressure 120mmHg. Also, high stress has been detected from the user's voice tone. Please generate a feedback message based on this."

[0907] This system allows users to manage their health condition in more detail and individually than ever before, enabling early and appropriate treatment. It also takes into account the user's emotional state, providing customized support for the user and realizing more effective health management.

[0908] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0909] Step 1:

[0910] The server collects health data provided by the user. The user uses a device to input information such as age, weight, height, blood pressure, heart rate, and medical history, and the device then sends this data to the server. The input health data is then aggregated on the server.

[0911] Step 2:

[0912] The server preprocesses the acquired health data. Specifically, it uses Python's Pandas library to impute missing values ​​in the data and standardizes the data using Scikit-learn's StandardScaler. The input is the acquired health data, and the output is the imputed and standardized health data.

[0913] Step 3:

[0914] The server uses the preprocessed data to build and train a machine learning model. It separates the data into features and labels, and splits it into training data and test data. It designs a multilayer neural network model using TensorFlow and trains the model using the training data. The input is the preprocessed health data, and the output is a trained machine learning model.

[0915] Step 4:

[0916] The emotion engine recognizes the user's emotional state. The device collects the user's voice data and converts it into text using voice analysis (for example, Google Cloud Speech-to-Text API). The text data is sent to a server, which performs emotion analysis using a library such as NLTK. The input is voice data, and the output is emotion data (for example, "high stress").

[0917] Step 5:

[0918] The server combines the emotion data with preprocessed health data and inputs it into a machine learning model to predict health risks. Including information about emotional states improves the accuracy of predictions. The inputs are emotion data and preprocessed health data, and the output is health risk prediction results.

[0919] Step 6:

[0920] The server sends the prediction results to the device. The device then generates a personalized message based on the predicted health risk and the user's emotional state, and notifies the user. For example, a message such as "There is a moderate health risk. Pay attention to stress management and try relaxation techniques" is displayed. The input is the health risk prediction result and emotional data, and the output is a notification message to the user.

[0921] This series of processes allows for more detailed management of the user's health status and provides feedback tailored to individual situations, enabling early and appropriate action to be taken.

[0922] (Application example 2)

[0923] 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."

[0924] Conventional health management systems predict health risks based solely on a user's health data, without taking the user's emotional state into account, resulting in poor prediction accuracy. Furthermore, because customized feedback based on the user's emotional state is not provided, it is difficult to provide appropriate advice. Therefore, there is a need for a system that can more accurately predict a user's health risks and provide feedback that takes their emotional state into account.

[0925] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring health data, means for pre-processing the acquired health data, means for training a machine learning model using the pre-processed health data, means for acquiring emotion data, means for analyzing the acquired emotion data, means for training a machine learning model by combining the pre-processed health data and the emotion data, means for predicting a user's health risk using the trained machine learning model, and means for notifying the user of the prediction result together with feedback customized based on the user's emotional state. This enables highly accurate health risk prediction that takes the user's emotional state into account and the provision of appropriate feedback that is individually customized.

[0926] "Health Data" refers to physical and medical information about a User, such as age, weight, height, blood pressure, heart rate, and medical history.

[0927] "Preprocessing" refers to the process of filling in missing values ​​in the acquired data and standardizing it to improve data quality.

[0928] A "machine learning model" is an algorithm that learns patterns and rules from massive data sets and makes predictions and classifications for new data.

[0929] "Emotional Data" refers to information related to a user's emotional state collected through speech analysis, facial expression analysis, and text analysis.

[0930] "Emotion engine" refers to a system that recognizes and analyzes a user's emotional state using voice analysis, facial expression analysis, or text analysis.

[0931] "Feedback" refers to advice or information provided to the user based on the system's predicted health risks.

[0932] "Customized feedback" refers to individually tailored advice or information that is tailored based on the user's emotional state.

[0933] "Features" refer to specific attributes or aspects of data that are input into a machine learning model, and include information that serves as the basis for classification and prediction.

[0934] "Labels" refer to the correct outcomes or categories of data used to train machine learning models.

[0935] "Missing value imputation" refers to the process of filling in missing data points with appropriate values.

[0936] "Standardization" refers to the process of standardizing the scale and distribution of data.

[0937] "Notification" refers to a message or information sent from the system to a user.

[0938] This invention is a system that acquires health data and predicts a user's health risks using a machine learning model. By combining it with an emotion engine that recognizes the user's emotional state, it provides even more accurate health risk predictions and personalized feedback.

[0939] System configuration

[0940] The system consists of the following main components:

[0941] 1. Means of collecting health data

[0942] The server obtains the user's physical and medical information, such as age, weight, height, blood pressure, heart rate, and medical history, from their smartphone or wearable device.

[0943] 2. Data preprocessing methods

[0944] The server preprocesses the acquired health data to improve the data quality, specifically by imputing missing values ​​and standardizing the data, using the Python Pandas library.

[0945] 3. Machine learning model training methods

[0946] The preprocessed data is used to train a machine learning model, here using a multi-layer neural network (MLPClassifier) ​​and the Scikit-learn library.

[0947] 4. Means of acquiring emotional data

[0948] The emotion engine is used to obtain emotion data by analyzing the user's voice, facial expression, and text. For example, the open source library LibROSA is used for voice data analysis.

[0949] 5. Sentiment Data Analysis Methods

[0950] We use natural language processing (NLP) techniques to analyze the acquired emotional data and recognize the user's emotional state. In particular, we use the SpaCy library to analyze text data.

[0951] 6. Machine learning model (emotion data combination) training method

[0952] The machine learning model is retrained by combining health data and emotion data as new features, which enables even more accurate prediction of health risks.

[0953] 7. Health risk prediction tools

[0954] A trained machine learning model is used to predict the user's health risks, and the predictions are made in real time, with the results sent from the server to the device.

[0955] 8. Feedback Notification Methods

[0956] The server notifies the user of the prediction results along with customized feedback based on the user's emotional state. For example, if the user is in a high stress state, the server provides feedback such as "Try some relaxation techniques."

[0957] Usage example

[0958] For example, a 30-year-old employee connects to the system using a smartphone and enters their health data. At the same time, their voice is also collected through the smartphone's microphone. The server preprocesses this data and analyzes their emotional data. The machine learning model combines this information to predict the employee's health risk and provide feedback. Specifically, the message displayed might read, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[0959] Example prompts for generative AI models

[0960] You can provide input to a generative AI model using prompt statements like the following:

[0961] This article introduces the functionality of a system that predicts health risks based on an employee's health data and emotional state. For example, consider an employee who is 30 years old, weighs 75 kg, is 1.80 m tall, has a blood pressure of 130 mmHg, and a heart rate of 80, and is in a high-stress situation. By inputting this information into the system, it can predict health risks and provide appropriate feedback. It also asks for specific prompts.

[0962] This system can predict users' health risks with greater accuracy than before and provide customized feedback that takes into account their emotional state, enabling more effective health management.

[0963] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0964] Step 1:

[0965] The server acquires health data from the user. The input includes information such as the user's age, weight, height, blood pressure, heart rate, and medical history. This data is collected from smartphones and wearable devices. The output is the acquired raw health data.

[0966] Step 2:

[0967] The server preprocesses the acquired health data. Here, missing values ​​in the data are imputed and standardized. Missing values ​​are generally imputed with the mean or median, and standardization is usually done by converting each feature to the same scale, usually with a mean of 0 and a standard deviation of 1. The input is the raw health data acquired in step 1, and the output is preprocessed health data.

[0968] Step 3:

[0969] The server trains a machine learning model using the preprocessed health data. It uses a multi-layer neural network (MLPClassifier) ​​and leverages the Scikit-learn library. The input includes the preprocessed data and its corresponding labels, and the output is a trained model.

[0970] Step 4:

[0971] The server uses an emotion engine to obtain the user's emotional data. Emotional data is collected through voice analysis, facial expression analysis, and text analysis. Specifically, data is collected from the smartphone's microphone and camera. The input includes the user's voice and image data, and the analyzed emotional data is obtained as the output.

[0972] Step 5:

[0973] The server analyzes the acquired emotional data. The emotion engine uses natural language processing (NLP) technology to recognize the user's emotional state from voice and text data. Specifically, it uses the SpaCy library. The input is emotional data, and the output is the emotional state resulting from the analysis.

[0974] Step 6:

[0975] The server combines the preprocessed health data and emotion data to create new features and retrain the machine learning model, enabling highly accurate health risk prediction that takes emotional states into account. The input includes the preprocessed health data and emotion data, and the output is the retrained model.

[0976] Step 7:

[0977] The server uses a trained machine learning model to predict the user's health risk. This prediction is performed in real time and uses the model obtained in step 6. The input includes current health data and emotion data, and the output is a predicted health risk result.

[0978] Step 8:

[0979] The server notifies the user of the prediction result along with customized feedback based on the user's emotional state. For example, if the risk is high and stress is high, a message such as "Try relaxation techniques" is displayed. The input includes the predicted health risk result and the user's emotional state, and the output is a customized feedback message.

[0980] 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.

[0981] 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.

[0982] 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.

[0983] [Fourth embodiment]

[0984] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0985] 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.

[0986] 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).

[0987] 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.

[0988] 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.

[0989] 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).

[0990] 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.

[0991] 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.

[0992] 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.

[0993] 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.

[0994] 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.

[0995] 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.

[0996] 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."

[0997] The present invention provides a system for acquiring health data and predicting a user's health risk using a machine learning model. To implement this system, a server, a terminal, and a user work together.

[0998] Health data acquisition and preprocessing

[0999] The server acquires health data provided by the user. The acquired health data includes individual physical and medical information such as age, weight, height, blood pressure, heart rate, and medical history. The server then preprocesses this health data. Specifically, it imputes missing values ​​using the data's average value, and then standardizes the data. Standardization is the process of converting data to a certain scale.

[1000] Building and training machine learning models

[1001] After the health data is preprocessed, the server builds a machine learning model. The model is designed as a multi-layered neural network, with each layer having a different number of nodes. The built model is then trained using the health data. During this training process, the dataset is divided into features (such as age and weight) and labels (whether or not a person has a health risk), and then further divided into training data and test data. The model uses the training data to learn patterns and improve its prediction accuracy.

[1002] Predictions and Notifications

[1003] Using the trained model, the device inputs the user's health data and predicts health risks. Once the user's health data is input into the device, the device sends the data to a server, which then performs the prediction. The prediction results are returned to the device and ultimately notified to the user. This allows the user to understand their health risks and take proactive health management measures as needed.

[1004] Specific examples

[1005] For example, suppose a user has the following data: age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg. This user data is sent to the server via the device. The server preprocesses and standardizes this data, then inputs it into a trained machine learning model. The model predicts health risks based on this information and returns the prediction results to the device. The device then notifies the user of the prediction results. For example, if the health risk score is 0.7, the user is notified, "There is a health risk. We recommend that you seek medical advice." Conversely, if the score is 0.2, the user is notified, "Your health risk is low."

[1006] This system allows users to proactively manage their own health status and take appropriate measures in advance, enabling early detection and prevention that are difficult to achieve with conventional medical systems.

[1007] The processing flow will be explained below.

[1008] Step 1:

[1009] The server acquires health data. It reads physical and medical information provided by the user or medical institution from a file such as "health_data.csv."

[1010] Step 2:

[1011] The server preprocesses the acquired health data by imputing missing values ​​with the mean value of the data and then standardizing the data by subtracting the mean value and dividing by the standard deviation.

[1012] Step 3:

[1013] The server separates the preprocessed data into features and labels. Features include age, weight, height, blood pressure, etc., and labels include health risks (e.g., whether or not a disease is present).

[1014] Step 4:

[1015] The server splits the dataset into training data and test data, which are used to train the machine learning model and the test data, which are used to evaluate the model.

[1016] Step 5:

[1017] The server builds a machine learning model, which is designed as a multi-layer neural network with different numbers of nodes in each layer.

[1018] Step 6:

[1019] The server trains a machine learning model using the training data. Using the features and labels contained in the dataset, the model learns the patterns needed to predict health risks.

[1020] Step 7:

[1021] The device inputs the user's health data and sends it to the server. The user inputs health information such as age, weight, height, and blood pressure into the device.

[1022] Step 8:

[1023] The server receives the user's health data and uses a trained machine learning model to predict health risks, which are then calculated as a risk score.

[1024] Step 9:

[1025] The server sends the prediction results to the device, and generates a message based on the risk score to notify the user.

[1026] Step 10:

[1027] The user checks the prediction results on the device, receiving messages such as "There is a health risk. We recommend you consult a doctor" or "The health risk is low," and understanding their own health condition.

[1028] Example 1

[1029] 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."

[1030] With conventional health management systems, it was difficult for users to accurately assess their own health risks and take early action. Furthermore, complex processes such as data imputation, standardization, and training of machine learning models often required manual work, which was time-consuming and labor-intensive. In addition, delays in notification of prediction results meant that users were unable to respond promptly.

[1031] 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.

[1032] In this invention, the server includes a means for a user to input health data, a means for a terminal to transmit the health data to the server, a means for the server to acquire and preprocess the health data, a means for the server to train a machine learning model, a means for predicting the user's health risk using the trained machine learning model, and a means for notifying the user of the prediction result. This allows users to simply input their own health data, and the server automatically preprocesses and analyzes it, thereby enabling quick prediction and notification of health risks.

[1033] "Means for users to input health data" refers to the operating means by which users input health information such as their age, weight, height, blood pressure, heart rate, and medical history using a dedicated application or an input form on a device.

[1034] The "means by which the device transmits health data to the server" refers to the means by which the device transmits the input health data to the server using a data transmission method such as an API request.

[1035] The "means for the server to acquire and preprocess health data" refers to the means by which the server receives health data sent from the terminal and automatically performs preprocessing such as missing value completion and standardization of the data.

[1036] "Means for the server to train a machine learning model" refers to means for the server to use the preprocessed data to train a machine learning model, such as a multi-layer neural network, and learn health risk patterns from the data.

[1037] "Means for predicting a user's health risk using a trained machine learning model" means a means in which a server uses a trained model to automatically predict a user's health risk from input health data.

[1038] The "means for notifying the user of the prediction results" refers to a means by which the server or terminal visually or audibly notifies the user of the health risk prediction results obtained by the machine learning model.

[1039] "Means for completing missing values ​​in health data" refers to means for completing missing parts of health data acquired by the server using statistical methods, average value completion, etc.

[1040] "Means for standardizing imputed health data" refers to an operational means by which the server converts the imputed health data into a certain scale (e.g., a range of 0 to 1) in order to convert it into a format suitable for machine learning models.

[1041] "Means for dividing a dataset into features and labels" refers to the means by which the server classifies health data into features required for prediction (age, weight, etc.) and labels for the prediction target (presence or absence of health risk).

[1042] The "means for dividing into training data and test data" refers to the means by which the server divides the health data into data for training the model (learning data) and data for evaluating the model (test data).

[1043] MODE FOR CARRYING OUT THE INVENTION

[1044] The present invention provides a system for acquiring a user's health data and predicting health risks using a machine learning model. This system operates in cooperation with a server, a terminal, and a user.

[1045] The server retrieves the user's health data sent from the device. This health data includes information such as age, weight, height, blood pressure, heart rate, and medical history. The retrieved health data is preprocessed using "scikit-learn." Specifically, preprocessing involves filling in missing values ​​with the data's average value, etc., and then standardizing the data. Standardization is the process of converting the data to a certain scale to make it suitable for machine learning models.

[1046] The preprocessed data is then used to train a multi-layer neural network model built using TensorFlow. During the training process, the dataset is divided into features (e.g., age and weight) and labels (e.g., whether or not a person has a health risk), and then further divided into training data and test data. The server uses the training data to train the model, enabling it to predict health risks with high accuracy.

[1047] The device is responsible for transmitting health data entered by the user to the server. When the user enters health data through the device, the device sends the data to the server. The server uses the received data to predict health risks and returns the results to the device. The device notifies the user of the prediction results. This notification allows the user to understand their own health risks and proactively manage their health as necessary.

[1048] For example, a user has the following data:

[1049] Age: 25

[1050] Weight: 70kg

[1051] Height: 1.75m

[1052] Blood pressure: 120mmHg

[1053] When the user enters this data from the device, the device sends the data to the server. The server preprocesses the data and inputs it into a trained machine learning model to predict health risks. The prediction results are returned to the device, which then notifies the user. For example, if the health risk score is 0.7, the device will display the message "There is a health risk. We recommend that you seek medical advice." If the score is 0.2, the device will display the message "Your health risk is low."

[1054] This system allows users to understand their own health condition in real time and take appropriate measures early on, making early detection and prevention possible, something that was difficult to achieve with conventional medical systems.

[1055] An example of a prompt to input to a generative AI model would be:

[1056] "Predict the health risk of a user with the following health data: age: 25, weight: 70kg, height: 1.75m, blood pressure: 120mmHg."

[1057] This concrete example clearly demonstrates how users and related devices work together to implement the system, ensuring the smooth operation of a series of processes, including health data collection, pre-processing, model training, prediction, and result notification, throughout the system.

[1058] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1059] Step 1:

[1060] The user enters health data.

[1061] Specifically, the user uses a dedicated application or an input form on the device to enter health information such as their age, weight, height, blood pressure, heart rate, and medical history.

[1062] Inputs include health data such as age, weight, height, blood pressure, heart rate, and medical history.

[1063] As an output, the entered health data is displayed in the input field on the terminal.

[1064] Step 2:

[1065] The device sends the health data to the server.

[1066] Specifically, the device converts the data entered by the user into JSON format and makes an API request to the server.

[1067] The input is health data entered by the user into the terminal.

[1068] As output, health data in JSON format is sent to the server.

[1069] Step 3:

[1070] The server acquires and pre-processes the health data.

[1071] Specifically, the server complements missing values ​​in the acquired health data and standardizes the data.

[1072] The input is health data in JSON format sent from the device.

[1073] As an output, imputed and standardized health data is generated.

[1074] Step 4:

[1075] The server trains the machine learning model.

[1076] Specifically, the server uses the standardized data to train a model using a machine learning library (e.g., TensorFlow). The data is split into features and labels, and then further split into training data and test data.

[1077] The input is preprocessed health data.

[1078] The output is a trained machine learning model.

[1079] Step 5:

[1080] The device again sends the user's health data to the server and makes a prediction.

[1081] Specifically, the device again acquires the user's health data and sends it to the server, which uses this data to predict health risks.

[1082] The input is the user's health data.

[1083] As an output, the server returns the predicted health risk results.

[1084] Step 6:

[1085] The server predicts health risks and returns the results to the device.

[1086] Specifically, the server uses the trained model to analyze the health data, calculate a risk score, and send the result back to the device.

[1087] The input is the user's health data, again submitted.

[1088] As an output, a predicted health risk score is generated and sent back to the terminal.

[1089] Step 7:

[1090] The terminal notifies the user of the result.

[1091] Specifically, the device receives the prediction results from the server and displays them to the user, who is then notified of their health risk score and advice based on it.

[1092] The input is the predicted health risk score sent from the server.

[1093] As an output, the prediction results and advice are displayed on the terminal screen and notified to the user.

[1094] (Application example 1)

[1095] 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."

[1096] Conventional food delivery services do not suggest menus that take into account the user's health condition. This can lead to users with health risks choosing inappropriate meals, further worsening their health. Another problem is that it is difficult for users to understand their own health information and manage it themselves.

[1097] 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.

[1098] In this invention, the server includes means for acquiring health data, means for preprocessing the acquired health data, means for training a machine learning model using the preprocessed health data, means for predicting a user's health risk using the trained machine learning model, means for proposing a meal menu based on the health risk, and means for notifying the user of the prediction result. This makes it possible to propose a meal menu suitable for a user with health risks, allowing the user to select an appropriate meal while understanding their own health condition.

[1099] "Health Data" is physical and medical information about a user, such as age, weight, height, blood pressure, heart rate, and medical history.

[1100] "Preprocessing" refers to the process of complementing and standardizing missing values ​​in the acquired health data.

[1101] A "machine learning model" is an algorithm designed as a multi-layer neural network for predicting health risks based on health data.

[1102] "Training" is the process by which a machine learning model learns patterns based on health data and improves its prediction accuracy.

[1103] "Health risk" is an indicator that indicates the possibility that a user has a particular health problem and the degree of that risk.

[1104] "Meal menu suggestion" is the act of recommending appropriate meal choices based on the user's health risks.

[1105] "Notification" is the process of informing the user of prediction results and suggested meal menus.

[1106] "Missing value imputation" is the process of filling in missing information in a dataset using the average value of the data or other methods.

[1107] "Standardization" is the act of transforming data so that it is processed on a consistent scale.

[1108] A "feature" is an element used as input for a machine learning model, such as age or weight in a training dataset.

[1109] A "label" is correct data corresponding to input data, such as the presence or absence of a health risk predicted by a machine learning model.

[1110] "Training data" is a dataset used to train a machine learning model.

[1111] "Test data" is a dataset used to evaluate a trained machine learning model.

[1112] A "server" is a computer that acquires health data, preprocesses it, trains machine learning models, and performs prediction processing.

[1113] A "terminal" is a device through which a user inputs health data and receives prediction results and meal menu suggestions.

[1114] A system for realizing this invention involves a server and a terminal working together to acquire health data, preprocess it, train a machine learning model, make predictions, and provide notifications.

[1115] Server Processing

[1116] First, the server obtains health data from the user. This health data includes age, weight, height, blood pressure, heart rate, medical history, etc. The server then preprocesses this data. Specifically, it imputes missing values ​​using the data's average value, and then standardizes the data. Standardization is the process of converting data to a certain scale.

[1117] After the data is preprocessed, the server builds a machine learning model. This model is a multi-layered neural network, with each layer having a different number of nodes. The model is then trained using health data. During the training process, the dataset is divided into features (such as age and weight) and labels (whether or not a person has a health risk), and then further divided into training data and test data. The model uses this training data to learn patterns and improve its prediction accuracy.

[1118] Terminal handling

[1119] The device is used to input and transmit the user's health data. The user inputs their health data into the device, and the device transmits the data to the server. Once the server performs a prediction, the result is returned to the device.

[1120] Before notifying the user, a process is performed to suggest an appropriate meal menu based on the predicted health risk. For example, a balanced, low-calorie menu is suggested to a user with a high health risk, while a regular menu is suggested to a user with a low health risk.

[1121] Hardware and software used

[1122] Hardware: High-performance servers, users' smartphones

[1123] Software: Python, TensorFlow (machine learning library), NumPy, Scikit-learn (for data preprocessing)

[1124] Specific examples

[1125] For example, suppose a user enters the following health data: age: 29, weight: 68 kg, height: 1.80 m, blood pressure: 125 mmHg, heart rate: 70 bpm. This data is sent to the server via the device. The server preprocesses and standardizes the data, then inputs it into a trained machine learning model. The model predicts health risks based on this information and returns the prediction results to the device. The device then suggests a meal menu to the user based on the health risks, and finally notifies the user of the prediction results.

[1126] Prompt Sentence Examples

[1127] Based on the user's health data, suggest the following food menu:

[1128] Age: 29

[1129] Weight: 68kg

[1130] Height: 1.80m

[1131] Blood pressure: 125mmHg

[1132] Heart rate: 70 bpm

[1133] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1134] Step 1:

[1135] The user inputs health data into the device, including age, weight, height, blood pressure, heart rate, etc. This input data is temporarily stored inside the device.

[1136] Step 2:

[1137] The device sends the user's health data to the server. The input data is packaged in a data format such as JSON and sent to the server.

[1138] Step 3:

[1139] The server receives the received health data and imputes missing values. For example, if there are missing values ​​in the dataset, they are imputed with the average value of the data. At this point, data with the missing values ​​of the input data imputed is obtained.

[1140] Step 4:

[1141] The server standardizes the imputed health data. Specifically, it converts the data to a certain scale (e.g., mean 0, standard deviation 1), thereby obtaining standardized data.

[1142] Step 5:

[1143] The server uses the preprocessed health data to train a machine learning model. This process involves splitting the dataset into features (e.g., age, weight, height, etc.) and labels (e.g., whether or not a person has a health risk), and then splitting it into training data and test data. The model uses the training data to learn patterns.

[1144] Step 6:

[1145] The server inputs the user's health data and uses a trained machine learning model to predict health risks. Standardized user health data is input into the model, and a predicted result (e.g., a health risk score) is output.

[1146] Step 7:

[1147] The server then suggests appropriate meal plans based on the predicted health risk score: for a high risk score, a balanced, low-calorie menu is suggested; for a low risk score, a regular menu is suggested.

[1148] Step 8:

[1149] The prediction results and meal menu suggestions obtained from the server are sent back to the device, which receives the information and notifies the user. Ultimately, the user can check their own health risks and appropriate meal menus through the device.

[1150] Examples:

[1151] For example, a user enters the following health data: age: 29, weight: 68 kg, height: 1.80 m, blood pressure: 125 mmHg, heart rate: 70 bpm. This data is sent to the server in JSON format in step 2. The server imputes missing values ​​in step 3 and standardizes them in step 4. The preprocessed data is used to train a model in step 5, and health risk is predicted in step 6. For example, if the health risk score is 0.7, a balanced, low-calorie menu is suggested in step 7, and the user is notified in step 8.

[1152] Example prompt sentence:

[1153] Based on the user's health data, suggest the following food menu:

[1154] Age: 29

[1155] Weight: 68kg

[1156] Height: 1.80m

[1157] Blood pressure: 125mmHg

[1158] Heart rate: 70 bpm

[1159] 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.

[1160] The present invention combines a system that acquires health data and predicts a user's health risks using machine learning models with an emotion engine that recognizes the user's emotions. This system improves the accuracy of health risk predictions and provides customized feedback according to the user's emotional state.

[1161] Health data acquisition and preprocessing

[1162] The server acquires health data provided by users, including individual physical and medical information such as age, weight, height, blood pressure, heart rate, medical history, etc. After acquiring the data, the server performs preprocessing, including imputing and standardizing missing values.

[1163] Building and training machine learning models

[1164] The server separates the preprocessed data into features (e.g., age, weight, height, etc.) and labels (e.g., whether or not there is a health risk). The server then splits the dataset into training data and test data and builds a machine learning model. This model is designed as a multi-layer neural network and is trained to learn specific patterns.

[1165] Introducing the Emotion Engine

[1166] The emotion engine recognizes the user's emotional state. This engine collects and analyzes the user's emotional data using techniques such as voice analysis, facial expression analysis, or text analysis. The emotion engine then sends the results to the server.

[1167] Combining health risk prediction and emotions

[1168] The server combines the emotion data obtained from the emotion engine with the preprocessed health data to predict the user's health risk using a machine learning model. Through this process, the emotional state is reflected as additional information in the health risk assessment.

[1169] Notification and feedback of prediction results

[1170] The server sends the prediction results to the device, which then notifies the user. The generated message takes into account the user's emotional state, providing more personalized feedback. For example, even if the risk score is high, if the user is feeling depressed, an encouraging message such as "You need to be careful about your current health condition, but don't panic and consult a specialist" will be displayed.

[1171] Specific examples

[1172] For example, suppose a user has data on age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg, and the emotion engine detects stress from the user's tone of voice. This data is sent from the device to the server. The server preprocesses this information and inputs it into a trained machine learning model. The model predicts health risks and generates feedback based on the results and emotion data. For example, if the model predicts "moderate health risk" and the emotion engine detects "high stress," the device will notify the user, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[1173] This system allows for more detailed and individualized management of the user's health status, enabling early and appropriate treatment. It also takes into account the user's emotional state, providing customized support for the user, leading to more effective health management.

[1174] The processing flow will be explained below.

[1175] Step 1:

[1176] The server acquires health data, reading physical and medical information provided by users and medical institutions from files such as "health_data.csv."

[1177] Step 2:

[1178] The server preprocesses the acquired health data. Missing values ​​are filled with the mean value of the data, and then the data is standardized. Standardization is the process of subtracting the mean value of the data and dividing by the standard deviation.

[1179] Step 3:

[1180] The server separates the preprocessed data into features (age, weight, height, blood pressure, etc.) and labels (presence or absence of health risks), preparing the data in a format suitable for model training.

[1181] Step 4:

[1182] The server splits the dataset into training data and test data, which are used to train the machine learning model and evaluate the model's performance.

[1183] Step 5:

[1184] The server builds a machine learning model, designed as a multi-layer neural network with different numbers of nodes in each layer.

[1185] Step 6:

[1186] The server trains the machine learning model using the training data. Using the features and labels, the model learns the patterns needed to predict health risks.

[1187] Step 7:

[1188] The device inputs the user's health data and sends it to the server. For example, the user inputs health information such as age, weight, height, and blood pressure into the device.

[1189] Step 8:

[1190] The emotion engine recognizes the user's emotional state, collects emotion data using voice analysis, facial expression analysis, or text analysis, and sends the analysis results to the server.

[1191] Step 9:

[1192] The server combines the emotion data obtained from the emotion engine with the pre-processed health data and predicts the user's health risk using a trained machine learning model.

[1193] Step 10:

[1194] The server sends the prediction results to the device, and a customized message is generated based on the prediction results, taking into account the user's emotional state.

[1195] Step 11:

[1196] The device will notify the user of the prediction results, for example, by displaying a message such as, "Your current health condition requires attention, but please do not rush into anything and consult a specialist."

[1197] Step 12:

[1198] Users can check the prediction results on their device and take medical advice or lifestyle changes as needed, thus receiving personalized health advice that takes into account their emotional state.

[1199] Example 2

[1200] 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."

[1201] Many conventional health management systems collect user health data and predict health risks using machine learning models. However, they lack the ability to predict health risks taking into account the user's emotional state and provide feedback based on the results. This makes it difficult to improve prediction accuracy and provide customized responses to users.

[1202] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1203] In this invention, the server includes means for acquiring health data, means for preprocessing the acquired health data, means for training a machine learning model using the preprocessed health data, means for predicting a user's health risk using the trained machine learning model, an emotion engine for recognizing the user's emotional state, means for predicting health risk by combining the emotion data obtained from the emotion engine with the preprocessed health data, and means for notifying the user of the prediction result and providing customized feedback according to the user's emotional state. This enables more accurate health risk prediction and personalized feedback that takes the user's emotional state into consideration.

[1204] "Health Data" refers collectively to a user's individual physical and medical information, such as age, weight, height, blood pressure, heart rate, and medical history.

[1205] "Preprocessing" refers to the process of filling in missing values ​​and standardizing the acquired health data.

[1206] A "machine learning model" is an algorithmic structure that learns patterns based on data and makes predictions and classifications.

[1207] An "emotion engine" is a system that recognizes and analyzes user emotional data using techniques such as voice analysis, facial expression analysis, and text analysis.

[1208] "Emotion data" is information about the user's emotional state, specifically data about the user's emotions such as stress, joy, sadness, etc.

[1209] "Feedback" refers to messages of instruction or advice provided to the user based on the prediction results.

[1210] "Standardization" refers to the process of converting data to a certain scale to improve the efficiency of model training.

[1211] "Notification" refers to the act of informing the user of the prediction results, particularly via a terminal.

[1212] The present invention combines a system that acquires health data and predicts a user's health risks using machine learning models with an emotion engine that recognizes the user's emotions. This system improves the accuracy of health risk predictions and provides customized feedback according to the user's emotional state.

[1213] Health data acquisition and preprocessing

[1214] The server receives health data provided by users. This data includes physical and medical information such as age, weight, height, blood pressure, heart rate, and medical history. Users enter this information using a terminal, which then sends the data to the server. The server then performs preprocessing, including imputing missing values ​​and standardizing the data. Python's Pandas library is used for imputing missing values, and Scikit-learn's StandardScaler is used for standardizing.

[1215] Building and training machine learning models

[1216] The server builds a machine learning model using the preprocessed health data. The data is first divided into features and labels, and then split into training data and test data. A multi-layer neural network model is designed using TensorFlow and the model is trained using the training data.

[1217] Introducing the Emotion Engine

[1218] The emotion engine is used to recognize the user's emotional state. This engine collects and analyzes user emotion data using techniques such as speech analysis, facial expression analysis, and text analysis. For example, it uses the Google Cloud Speech-to-Text API to convert speech to text and performs sentiment analysis of the text using the Natural Language Toolkit (NLTK).

[1219] Combining health risk prediction and emotions

[1220] The server combines the emotion data obtained from the emotion engine with the preprocessed health data and uses a machine learning model to predict the user's health risk. By incorporating the emotion data into the health risk assessment, the accuracy of the prediction is improved.

[1221] Notification and feedback of prediction results

[1222] The server sends the prediction results to the device, which then generates a personalized message based on the predicted health risk and the user's emotional state and notifies the user. For example, even if the risk score is high, if the user is in a state of high stress, a message such as "You need to be careful about your current health condition, but don't rush and consult a specialist" will be displayed.

[1223] Specific examples

[1224] For example, suppose a user has data on age 25, weight 70 kg, height 1.75 m, and blood pressure 120 mmHg, and the emotion engine detects stress from the user's tone of voice. This data is sent from the device to the server. The server preprocesses this information and inputs it into a trained machine learning model. The model predicts health risks and generates feedback based on the results and emotion data. For example, if the model predicts "moderate health risk" and the emotion engine detects "high stress," the device will notify the user, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[1225] Example prompts for generative AI models

[1226] "Please predict the health risk of a user with the following data: age 25, weight 70kg, height 1.75m, and blood pressure 120mmHg. Also, high stress has been detected from the user's voice tone. Please generate a feedback message based on this."

[1227] This system allows users to manage their health condition in more detail and individually than ever before, enabling early and appropriate treatment. It also takes into account the user's emotional state, providing customized support for the user and realizing more effective health management.

[1228] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1229] Step 1:

[1230] The server collects health data provided by the user. The user uses a device to input information such as age, weight, height, blood pressure, heart rate, and medical history, and the device then sends this data to the server. The input health data is then aggregated on the server.

[1231] Step 2:

[1232] The server preprocesses the acquired health data. Specifically, it uses Python's Pandas library to impute missing values ​​in the data and standardizes the data using Scikit-learn's StandardScaler. The input is the acquired health data, and the output is the imputed and standardized health data.

[1233] Step 3:

[1234] The server uses the preprocessed data to build and train a machine learning model. It separates the data into features and labels, and splits it into training data and test data. It designs a multilayer neural network model using TensorFlow and trains the model using the training data. The input is the preprocessed health data, and the output is a trained machine learning model.

[1235] Step 4:

[1236] The emotion engine recognizes the user's emotional state. The device collects the user's voice data and converts it into text using voice analysis (for example, Google Cloud Speech-to-Text API). The text data is sent to a server, which performs emotion analysis using a library such as NLTK. The input is voice data, and the output is emotion data (for example, "high stress").

[1237] Step 5:

[1238] The server combines the emotion data with preprocessed health data and inputs it into a machine learning model to predict health risks. Including information about emotional states improves the accuracy of predictions. The inputs are emotion data and preprocessed health data, and the output is health risk prediction results.

[1239] Step 6:

[1240] The server sends the prediction results to the device. The device then generates a personalized message based on the predicted health risk and the user's emotional state, and notifies the user. For example, a message such as "There is a moderate health risk. Pay attention to stress management and try relaxation techniques" is displayed. The input is the health risk prediction result and emotional data, and the output is a notification message to the user.

[1241] This series of processes allows for more detailed management of the user's health status and provides feedback tailored to individual situations, enabling early and appropriate action to be taken.

[1242] (Application example 2)

[1243] 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."

[1244] Conventional health management systems predict health risks based solely on a user's health data, without taking the user's emotional state into account, resulting in poor prediction accuracy. Furthermore, because customized feedback based on the user's emotional state is not provided, it is difficult to provide appropriate advice. Therefore, there is a need for a system that can more accurately predict a user's health risks and provide feedback that takes their emotional state into account.

[1245] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring health data, means for pre-processing the acquired health data, means for training a machine learning model using the pre-processed health data, means for acquiring emotion data, means for analyzing the acquired emotion data, means for training a machine learning model by combining the pre-processed health data and the emotion data, means for predicting a user's health risk using the trained machine learning model, and means for notifying the user of the prediction result together with feedback customized based on the user's emotional state. This enables highly accurate health risk prediction that takes the user's emotional state into account and the provision of appropriate feedback that is individually customized.

[1246] "Health Data" refers to physical and medical information about a User, such as age, weight, height, blood pressure, heart rate, and medical history.

[1247] "Preprocessing" refers to the process of filling in missing values ​​in the acquired data and standardizing it to improve data quality.

[1248] A "machine learning model" is an algorithm that learns patterns and rules from massive data sets and makes predictions and classifications for new data.

[1249] "Emotional Data" refers to information related to a user's emotional state collected through speech analysis, facial expression analysis, and text analysis.

[1250] "Emotion engine" refers to a system that recognizes and analyzes a user's emotional state using voice analysis, facial expression analysis, or text analysis.

[1251] "Feedback" refers to advice or information provided to the user based on the system's predicted health risks.

[1252] "Customized feedback" refers to individually tailored advice or information that is tailored based on the user's emotional state.

[1253] "Features" refer to specific attributes or aspects of data that are input into a machine learning model, and include information that serves as the basis for classification and prediction.

[1254] "Labels" refer to the correct outcomes or categories of data used to train machine learning models.

[1255] "Missing value imputation" refers to the process of filling in missing data points with appropriate values.

[1256] "Standardization" refers to the process of standardizing the scale and distribution of data.

[1257] "Notification" refers to a message or information sent from the system to a user.

[1258] This invention is a system that acquires health data and predicts a user's health risks using a machine learning model. By combining it with an emotion engine that recognizes the user's emotional state, it provides even more accurate health risk predictions and personalized feedback.

[1259] System configuration

[1260] The system consists of the following main components:

[1261] 1. Means of collecting health data

[1262] The server obtains the user's physical and medical information, such as age, weight, height, blood pressure, heart rate, and medical history, from their smartphone or wearable device.

[1263] 2. Data preprocessing methods

[1264] The server preprocesses the acquired health data to improve the data quality, specifically by imputing missing values ​​and standardizing the data, using the Python Pandas library.

[1265] 3. Machine learning model training methods

[1266] The preprocessed data is used to train a machine learning model, here using a multi-layer neural network (MLPClassifier) ​​and the Scikit-learn library.

[1267] 4. Means of acquiring emotional data

[1268] The emotion engine is used to obtain emotion data by analyzing the user's voice, facial expression, and text. For example, the open source library LibROSA is used for voice data analysis.

[1269] 5. Sentiment Data Analysis Methods

[1270] We use natural language processing (NLP) techniques to analyze the acquired emotional data and recognize the user's emotional state. In particular, we use the SpaCy library to analyze text data.

[1271] 6. Machine learning model (emotion data combination) training method

[1272] The machine learning model is retrained by combining health data and emotion data as new features, which enables even more accurate prediction of health risks.

[1273] 7. Health risk prediction tools

[1274] A trained machine learning model is used to predict the user's health risks, and the predictions are made in real time, with the results sent from the server to the device.

[1275] 8. Feedback Notification Methods

[1276] The server notifies the user of the prediction results along with customized feedback based on the user's emotional state. For example, if the user is in a high stress state, the server provides feedback such as "Try some relaxation techniques."

[1277] Usage example

[1278] For example, a 30-year-old employee connects to the system using a smartphone and enters their health data. At the same time, their voice is also collected through the smartphone's microphone. The server preprocesses this data and analyzes their emotional data. The machine learning model combines this information to predict the employee's health risk and provide feedback. Specifically, the message displayed might read, "You have a moderate health risk. Pay attention to stress management and try relaxation techniques."

[1279] Example prompts for generative AI models

[1280] You can provide input to a generative AI model using prompt statements like the following:

[1281] This article introduces the functionality of a system that predicts health risks based on an employee's health data and emotional state. For example, consider an employee who is 30 years old, weighs 75 kg, is 1.80 m tall, has a blood pressure of 130 mmHg, and a heart rate of 80, and is in a high-stress situation. By inputting this information into the system, it can predict health risks and provide appropriate feedback. It also asks for specific prompts.

[1282] This system can predict users' health risks with greater accuracy than before and provide customized feedback that takes into account their emotional state, enabling more effective health management.

[1283] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1284] Step 1:

[1285] The server acquires health data from the user. The input includes information such as the user's age, weight, height, blood pressure, heart rate, and medical history. This data is collected from smartphones and wearable devices. The output is the acquired raw health data.

[1286] Step 2:

[1287] The server preprocesses the acquired health data. Here, missing values ​​in the data are imputed and standardized. Missing values ​​are generally imputed with the mean or median, and standardization is usually done by converting each feature to the same scale, usually with a mean of 0 and a standard deviation of 1. The input is the raw health data acquired in step 1, and the output is preprocessed health data.

[1288] Step 3:

[1289] The server trains a machine learning model using the preprocessed health data. It uses a multi-layer neural network (MLPClassifier) ​​and leverages the Scikit-learn library. The input includes the preprocessed data and its corresponding labels, and the output is a trained model.

[1290] Step 4:

[1291] The server uses an emotion engine to obtain the user's emotional data. Emotional data is collected through voice analysis, facial expression analysis, and text analysis. Specifically, data is collected from the smartphone's microphone and camera. The input includes the user's voice and image data, and the analyzed emotional data is obtained as the output.

[1292] Step 5:

[1293] The server analyzes the acquired emotional data. The emotion engine uses natural language processing (NLP) technology to recognize the user's emotional state from voice and text data. Specifically, it uses the SpaCy library. The input is emotional data, and the output is the emotional state resulting from the analysis.

[1294] Step 6:

[1295] The server combines the preprocessed health data and emotion data to create new features and retrain the machine learning model, enabling highly accurate health risk prediction that takes emotional states into account. The input includes the preprocessed health data and emotion data, and the output is the retrained model.

[1296] Step 7:

[1297] The server uses a trained machine learning model to predict the user's health risk. This prediction is performed in real time and uses the model obtained in step 6. The input includes current health data and emotion data, and the output is a predicted health risk result.

[1298] Step 8:

[1299] The server notifies the user of the prediction result along with customized feedback based on the user's emotional state. For example, if the risk is high and stress is high, a message such as "Try relaxation techniques" is displayed. The input includes the predicted health risk result and the user's emotional state, and the output is a customized feedback message.

[1300] 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.

[1301] 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.

[1302] 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.

[1303] 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.

[1304] 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.

[1305] 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.

[1306] 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).

[1307] 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.

[1308] 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."

[1309] 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.

[1310] 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).

[1311] 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.

[1312] 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.

[1313] 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.

[1314] 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.

[1315] 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.

[1316] 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.

[1317] 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.

[1318] 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.

[1319] 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.

[1320] 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.

[1321] The following is further disclosed regarding the above embodiment.

[1322] (Claim 1)

[1323] a means for acquiring health data;

[1324] means for preprocessing the acquired health data;

[1325] means for training a machine learning model using the preprocessed health data;

[1326] a means for predicting a user's health risk using a trained machine learning model;

[1327] a means for notifying a user of the prediction result;

[1328] A system including:

[1329] (Claim 2)

[1330] a means of imputing missing values ​​in health data;

[1331] a means of standardizing imputed health data;

[1332] 10. The system of claim 1, further comprising:

[1333] (Claim 3)

[1334] A means of splitting the dataset into features and labels;

[1335] a means for splitting the data into training and testing data;

[1336] 10. The system of claim 1, further comprising:

[1337] "Example 1"

[1338] (Claim 1)

[1339] a means for a user to input health data;

[1340] A means for the terminal to transmit health data to a server;

[1341] A means for the server to acquire and pre-process the health data;

[1342] a means for the server to train a machine learning model;

[1343] a means for predicting a user's health risk using a trained machine learning model;

[1344] a means for notifying a user of the prediction result;

[1345] A system including:

[1346] (Claim 2)

[1347] a means of imputing missing values ​​in health data;

[1348] a means of standardizing imputed health data;

[1349] 10. The system of claim 1, further comprising:

[1350] (Claim 3)

[1351] A means of splitting the dataset into features and labels;

[1352] a means for splitting the data into training and testing data;

[1353] 10. The system of claim 1, further comprising:

[1354] "Application Example 1"

[1355] (Claim 1)

[1356] a means for acquiring health data;

[1357] means for preprocessing the acquired health data;

[1358] means for training a machine learning model using the preprocessed health data;

[1359] a means for predicting a user's health risk using a trained machine learning model;

[1360] A method to suggest meal menus based on health risks,

[1361] a means for notifying a user of the prediction result;

[1362] A system including:

[1363] (Claim 2)

[1364] a means of imputing missing values ​​in health data;

[1365] a means of standardizing imputed health data;

[1366] 10. The system of claim 1, further comprising:

[1367] (Claim 3)

[1368] A means of splitting the dataset into features and labels;

[1369] a means for splitting the data into training and testing data;

[1370] 10. The system of claim 1, further comprising:

[1371] "Example 2: Combining Emotion Engines"

[1372] (Claim 1)

[1373] a means for acquiring health data;

[1374] means for preprocessing the acquired health data;

[1375] means for training a machine learning model using the preprocessed health data;

[1376] a means for predicting a user's health risk using a trained machine learning model;

[1377] an emotion engine that recognizes the user's emotional state;

[1378] a means for predicting health risks by combining the emotion data obtained from the emotion engine with the preprocessed health data; and

[1379] means for notifying the user of the prediction result and providing customized feedback according to the user's emotional state;

[1380] A system including:

[1381] (Claim 2)

[1382] a means of imputing missing values ​​in health data;

[1383] a means of standardizing imputed health data;

[1384] 10. The system of claim 1, further comprising:

[1385] (Claim 3)

[1386] A means of splitting the dataset into features and labels;

[1387] a means for splitting the data into training and testing data;

[1388] 10. The system of claim 1, further comprising:

[1389] "Application example 2 when combining emotion engines"

[1390] (Claim 1)

[1391] a means for acquiring health data;

[1392] means for preprocessing the acquired health data;

[1393] means for training a machine learning model using the preprocessed health data;

[1394] A means for acquiring emotion data;

[1395] A means for analyzing the acquired emotion data;

[1396] a means for combining the preprocessed health data and emotion data to train a machine learning model; and

[1397] a means for predicting a user's health risk using a trained machine learning model;

[1398] means for notifying the user of the prediction result along with customized feedback based on the user's emotional state;

[1399] A system including:

[1400] (Claim 2)

[1401] a means of imputing missing values ​​in health data;

[1402] a means of standardizing imputed health data;

[1403] means for pre-processing the emotion data acquired by the emotion engine;

[1404] 10. The system of claim 1, further comprising:

[1405] (Claim 3)

[1406] A means of splitting the dataset into features and labels;

[1407] a means for splitting the data into training and testing data;

[1408] A means for combining the preprocessed health data and emotion data and dividing them into features and labels;

[1409] 10. The system of claim 1, further comprising: [Explanation of symbols]

[1410] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for acquiring health data; means for preprocessing the acquired health data; means for training a machine learning model using the preprocessed health data; a means for predicting a user's health risk using a trained machine learning model; a means for notifying a user of the prediction result; A system including:

2. a means of imputing missing values ​​in health data; a means of standardizing imputed health data; The system of claim 1 further comprising:

3. A means of splitting the dataset into features and labels; a means for splitting the data into training and testing data; The system of claim 1 further comprising:

Images