Artificial intelligence assisted health state monitoring system and method

Abstract

The invention discloses an artificial intelligence assisted health state monitoring system and method, and the method comprises the steps: collecting the facial image features of a user when the user looks into a mirror, and storing the facial features; according to the collected facial image feature picture, matching with a plurality of stored user images, and identifying the identity of the current user; calling all historical facial feature pictures stored by the current user, and gradually comparing the historical facial feature pictures with the last facial feature picture; when the facial feature pictures are gradually compared with the facial feature pictures of the last time, a far-end artificial intelligence proxy server is connected, and comprehensive online analysis and processing are carried out on the health state of the user; and in combination with online analysis, generating a comparison health detection report of the current face collection and the last face collection.

Description

Technical Field

[0001] The present invention relates to an artificial intelligence-assisted health status monitoring system and method. Background Art

[0002] With the rapid development of technologies such as mobile communications, computers, and the Internet, people can more conveniently use some wearable electronic devices, smart mobile terminals, and networks to monitor changes in their physical health status at all times, thereby achieving the purpose of disease prevention. However, the existing health status monitoring system is not perfect, which makes such detection insufficient in terms of intelligence and precision, limiting people's use of this type of service. Therefore, it is necessary to propose an artificial intelligence-assisted health status monitoring device and method, which can bring convenience to people while improving the accuracy and timeliness of monitoring data and analysis reports, and play a useful reference role in disease prevention. Summary of the invention

[0003] The present invention provides an artificial intelligence-assisted health status monitoring system and method in order to solve the problems existing in the above-mentioned prior art.

[0004] The technical solutions adopted in the present invention are: An artificial intelligence-assisted health status monitoring method, comprising: When the user looks in the mirror, the user's facial image features are collected and stored; According to the collected facial image feature picture, it is matched with several stored user images to identify the current user identity; Retrieve all historical facial feature images stored by the current user and compare them with the previous facial feature images step by step; While gradually comparing the facial feature images with the previous ones, it connects to the remote artificial intelligence proxy server to conduct comprehensive online analysis and processing of the user's health status; Combined with online analysis, a comparative health detection report between this facial collection and the previous facial collection is generated.

[0005] Furthermore, the collected facial image features include: skin color, eyes and tongue.

[0006] Furthermore, based on the skin color information, the user's facial changes are determined and compared with the previous facial changes, and a remote artificial intelligence proxy server is connected to analyze the differences in the comparison; Based on the information of the white of the eye, the white of the eye is compared with the last photo of the white of the eye, and the remote artificial intelligence proxy server is connected to analyze the difference in the comparison; Based on the information of the tongue, the color and shape changes of the tongue are determined and compared with the previous tongue photo. At the same time, the remote artificial intelligence proxy server is connected to analyze the differences in the comparison.

[0007] Furthermore, the mirror is provided with a touch screen for user interaction, and the user inputs facial information for identity verification before use.

[0008] The present invention also discloses an artificial intelligence-assisted health status monitoring system, comprising A mirror, wherein the mirror is provided with a data acquisition module for acquiring facial image features of a user; A data input module, which is used for inputting user requests, and the input content includes user instructions, training data sets, and supplementary data obtained from interviews; The data processing module is used to pre-process the data obtained by the data input module and the data acquisition module, and generate corresponding tasks. It can also decompose the tasks to form prompt words for related subtasks, and connect to the artificial intelligence proxy server to guide it to perform online analysis of the user's health status data based on the prompt words to obtain a human health status analysis report; A data display module, used to display the data input information and human health status analysis report information; A data communication module is used to connect to the artificial intelligence proxy server to transmit and receive prompt words and human health status analysis report information; The data storage module is used to store facial image features.

[0009] Furthermore, a touch screen display is embedded in the mirror.

[0010] Furthermore, the data acquisition module is a camera.

[0011] Furthermore, the data acquisition module also includes a microphone and a heart rate sensor.

[0012] The present invention has the following beneficial effects: The present invention collects user facial information by embedding a sensor that can detect human health status in the smart mirror, connects to a remote artificial intelligence proxy server through the embedded wireless communication module, and uses ChatGPT technology to realize intelligent online analysis of user health status data, thereby improving the accuracy and timeliness of the analysis report, and realizing real-time monitoring of the user's own health status information through the simple method of looking in the mirror every day. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a principle block diagram of the present invention.

[0014] Figure 2This is a flow chart of a method for monitoring user health status assisted by artificial intelligence according to an embodiment of the present invention; Figure 3 The present invention is a flowchart of iterative generation of prompt words (Prompt) based on a feedback mechanism assisted by ChatGPT in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The present invention will be further described below in conjunction with the accompanying drawings.

[0016] like Figure 1 As shown, the present invention provides an artificial intelligence-assisted health status monitoring method, and its research content mainly includes two parts: (1) artificial intelligence-assisted health status monitoring smart mirror; (2) artificial intelligence-assisted health status monitoring method. The specific description is as follows: (1) By embedding sensors that can detect human health status in the mirror, data information on the user's face, tongue coating and eyeballs can be collected, and the user's health status analysis report can be obtained by connecting to a remote artificial intelligence proxy server through the embedded wireless communication module. Therefore, real-time monitoring of the user's own health status information can be achieved by simply looking in the mirror every day.

[0017] (2) Preprocess the acquired user health status data, generate corresponding tasks based on the user input information, and convert it into corresponding prompts. At the same time, use ChatGPT technology to realize intelligent analysis and processing of user health status data based on the prompts, so as to improve the accuracy and timeliness of human health status analysis reports.

[0018] Figure 1 It is a principle block diagram of the present invention, and the present invention includes: The data input module 110 is used for inputting user requests. The data input comes from the interface input in the data display module 140, which can be realized by voice input or touch screen input. The input data content includes user instructions, training data sets, and supplementary data obtained from TCM consultation clinics.

[0019] User instructions include data collection commands, user query commands, data set training, etc. Among them, the training data set specifically refers to the cloud medical database related to human health status, which is a local self-built health diagnosis database or a major hospital database in the cloud and a medical expert diagnosis database.

[0020] The data acquisition module 120 is a camera used to collect facial image features. Specifically: Skin condition: Healthy skin is usually smooth, shiny, and free of obvious acne, dryness, or redness. Abnormal skin, such as pale complexion, jaundice (yellowing of the skin), or a large number of acne, may be a sign of health problems. The overall skin tone should be even, with a natural rosy color on the cheeks. Dull skin or obvious spots may be related to lifestyle habits or health problems.

[0021] Eyes: The white of the eye should be bright and not yellow or bloodshot. The condition of the eyes can reflect the health of the liver, kidneys and other internal organs.

[0022] Tongue: Use a mirror to check the color and shape of your tongue. A healthy tongue should be a healthy pink color with a smooth surface. A thick, white or red tongue coating may be a sign of a physical problem.

[0023] The data acquisition module also includes a microphone and a heart rate sensor. The microphone is used to capture the changing characteristics of sound waves and convert them into audio information. The heart rate sensor is placed on the mirror surface and can be touched by the mirror surface so as to collect pulse and heart rate information through fingers.

[0024] The data processing module 130 is used to pre-process the data obtained in the data input module 110 and the data acquisition module 120, and generate corresponding tasks. The tasks include training a human health status analysis model based on a local data set or an online data set, and analyzing the current user health status online.

[0025] The data processing module 130 should be able to decompose the task, form prompt words for related sub-tasks, and be connected to the remote artificial intelligence proxy server through the data communication module 150, so that it can use ChatGPT technology to further analyze and process the data based on the guidance of the prompt words, and return the processing results, and output them through the display module 140.

[0026] The task is to analyze the user's current health status data, and the analysis result is the corresponding human health status analysis report. Among them, the human health status analysis report is a health diagnosis reference suggestion given based on the user's current health status data.

[0027] The data display module 140 is used to display the data input information and the human health status analysis report information. The data display module is placed on a mirror surface, which is an LCD display screen or a CRT display screen.

[0028] The data communication module 150 is used to send user instructions or prompt word information to the artificial intelligence proxy server, and receive the human health status analysis report information from the artificial intelligence proxy server. The data communication module includes at least one of a WiFi module or a mobile communication module for connecting to the Internet.

[0029] The data storage module 160 is used to store relevant data information corresponding to TCM inspection, auscultation, questioning and palpation, as well as converted prompt words, and corresponding human health status analysis report information.

[0030] The relationship between the components in the smart mirror is as follows: The data processing module 130 is a core module, and is connected to the data input module 110, the data acquisition module 120, the data display module 140, the data communication module 150 and the data display module 160, while the data input module 110 is connected to the data acquisition module 120, the data processing module 130 and the data display module 140. The data input module 110 converts the voice input information in the data acquisition module 120 and outputs it through the data display module 140.

[0031] Figure 2 This is a flow chart of a method for monitoring user health status assisted by artificial intelligence according to an embodiment of the present invention. Based on the method of "looking, smelling, asking, and palpating" in traditional Chinese medicine diagnosis, various types of health status data of users are collected, and artificial intelligence-assisted comprehensive diagnosis and analysis are performed. The key lies in the accuracy of human health status analysis model training and the accuracy of task prompt word generation, and includes the following steps S210-S250: S210, the user inputs information to train a human health status analysis model.

[0032] In the initial state of the system, the ChatGPT technology is used to train the human health status analysis model online based on the training data set provided according to the user input instructions. Among them, the training data set is a local self-built health diagnosis database or an online database of major hospitals and medical expert diagnosis databases. Based on the trained human health status analysis model, the user is prompted to enter the current physical health status information according to the questionnaire as supplementary data for TCM consultation.

[0033] S220, intelligently obtain the user's health status data. Through intelligent voice input and control of sensors in the data acquisition module, facial image feature information required for "inspection" is collected in sequence.

[0034] S230, data preprocessing, generating effective prompt words. The collected user health status data is preprocessed, and its feature data is extracted using image processing and other technologies. At the same time, it is decomposed into multiple subtasks and effective prompt words are generated to guide the artificial intelligence proxy server to perform related subtasks. Among them, effective prompt words can be iteratively generated based on the feedback mechanism, such as Figure 3 The process is shown.

[0035] S240, according to the collected facial image feature picture, match it with several stored user images to identify the current user identity and ensure privacy. After confirming the user identity information, retrieve all the historical facial feature pictures stored by the current user, and gradually compare them with the previous facial feature pictures. At the same time, connect to the remote artificial intelligence proxy server, and then conduct comprehensive online analysis and processing of the user's health status based on the comparison differences and task prompts based on ChatGPT technology.

[0036] The medical diagnosis expert or health expert identity of the artificial intelligence proxy server can be set through prompt words, and then the task prompt words can be entered, supplemented by a test data set, to guide the artificial intelligence proxy server to use ChatGPT technology to conduct comprehensive online analysis and processing of the user's health status and give the analysis results.

[0037] S250, obtaining a human health status analysis report and displaying and outputting it. The artificial intelligence proxy server returns the human health status analysis report and outputs it through the data display module of the smart mirror.

[0038] The prompt words are in text form, and the language type is not limited to Chinese and English.

[0039] The test dataset is multimodal data, and its type is not limited to text files. It includes text, pictures, voice, video and other files. It mainly contains data on the user's current health status and can be provided as auxiliary parameters based on the technical development version of ChatGPT.

[0040] Figure 3 The ChatGPT-assisted prompt word (Prompt) iterative generation flow chart based on the feedback mechanism of the embodiment of the present invention is aimed at continuously refining and optimizing the generated task prompt words, including the following steps S310-S350: S310, data preprocessing, generating subtasks and data sets. The collected user health status data is preprocessed, and demand analysis is performed to generate multiple subtasks and data sets.

[0041] S320, generating task prompt words. Generate prompt words for each subtask and make judgments. If an error message is generated, execute step S350 and hand it over to the interpreter for processing. Otherwise, execute the following steps.

[0042] S330, generating a solution with certain rules, connecting to a remote artificial intelligence proxy server, and executing the selected solution according to the task prompt based on ChatGPT technology. The rules adopted by the solution include setting the priority weights of each subtask, and reasonable judgment of the association relationship between subtasks.

[0043] S340, obtaining a human health status analysis report. The artificial intelligence proxy server executes the solution to generate a human health status analysis report and returns it to the smart mirror receiving end. If the analysis report is of poor quality or has obvious analysis errors, step S350 is executed, otherwise the process ends.

[0044] S350, the interpreter analyzes the wrong and poor quality results. If the task prompt word is generated incorrectly, the user is prompted to enter more information, and step S310 is repeated and the subtasks are re-decomposed. If the analysis report is of poor quality or contains obvious errors, other solutions need to be provided, and step S310 is repeated to generate a new solution.

[0045] Among them, the subtasks and data sets include: performing health status analysis based on facial data features, and the data set is the corresponding facial image feature data.

[0046] The subtasks and data sets also include: performing health status analysis based on tongue coating data features, and the data set is the corresponding tongue coating image feature data.

[0047] The subtasks and data sets also include: performing health status analysis based on pulse data characteristics, and the data set is corresponding pulse, heart rate, blood pressure and other characteristic data.

[0048] The subtasks and data sets also include: performing health status analysis based on supplementary data information, and the data set is corresponding physical health status questionnaire data.

[0049] The present invention is based on a medical-related training data set and adopts ChatGPT technology to train a human health status analysis model online; various types of health status data of users are collected based on the method of "looking, smelling, asking, and palpating" in traditional Chinese medicine diagnosis; the collected user health status data is pre-processed, its feature data is extracted, decomposed into multiple subtasks and effective prompt words are generated; the remote artificial intelligence proxy server is connected, and the user's health status is comprehensively analyzed and processed online according to the task prompt words based on ChatGPT technology to obtain a human health status analysis report.

[0050] The above description is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be regarded as within the protection scope of the present invention.

Claims

1. An artificial intelligence-assisted health status monitoring method, characterized in that: include When the user looks in the mirror, the user's facial image features are collected and stored; According to the collected facial image feature picture, it is matched with several stored user images to identify the current user identity; Retrieve all historical facial feature images stored by the current user and compare them with the previous facial feature images step by step; While gradually comparing the facial feature images with the previous ones, it connects to the remote artificial intelligence proxy server to conduct comprehensive online analysis and processing of the user's health status; Combined with online analysis, a comparative health detection report between this facial collection and the previous facial collection is generated.

2. The artificial intelligence-assisted health status monitoring method according to claim 1, characterized in that: The facial image features collected include: skin complexion, eyes and tongue.

3. The artificial intelligence-assisted health status monitoring method according to claim 2, characterized in that: Determine the user's facial changes based on the skin color information and compare them with the previous facial changes. At the same time, connect to the remote artificial intelligence proxy server to analyze the differences in the comparison. Based on the information of the white of the eye, the white of the eye is compared with the last photo of the white of the eye, and the remote artificial intelligence proxy server is connected to analyze the difference in the comparison; Based on the information of the tongue, the color and shape changes of the tongue are determined and compared with the previous tongue photo. At the same time, the remote artificial intelligence proxy server is connected to analyze the differences in the comparison.

4. The artificial intelligence-assisted health status monitoring method according to claim 1, characterized in that: The mirror is provided with a touch screen for user interaction. Before use, the user inputs facial information for identity verification.

5. An artificial intelligence-assisted health status monitoring system, characterized in that: include A mirror, wherein the mirror is provided with a data acquisition module for acquiring facial image features of a user; A data input module, which is used for inputting user requests, and the input content includes user instructions, training data sets, and supplementary data obtained from interviews; The data processing module is used to pre-process the data obtained by the data input module and the data acquisition module, and generate corresponding tasks. It can also decompose the tasks to form prompt words for related subtasks, and connect to the artificial intelligence proxy server to guide it to perform online analysis of the user's health status data based on the prompt words to obtain a human health status analysis report; A data display module, used to display the data input information and human health status analysis report information; A data communication module is used to connect to the artificial intelligence proxy server to transmit and receive prompt words and human health status analysis report information; The data storage module is used to store facial image features.

6. The artificial intelligence-assisted health status monitoring system according to claim 5, characterized in that: A touch screen display is embedded in the mirror.

7. The artificial intelligence-assisted health status monitoring system according to claim 5, characterized in that: The data acquisition module is a camera.

8. The artificial intelligence-assisted health status monitoring system according to claim 5, characterized in that: The data acquisition module also includes a microphone and a heart rate sensor.

Images