CN-121983285-A - AIGC-based health diagnosis and accompanying system and method for old people

CN121983285ACN 121983285 ACN121983285 ACN 121983285ACN-121983285-A

Abstract

The invention discloses a AIGC-based health diagnosis and accompanying system and method for old people, and relates to the technical field of intelligent care, wherein the intelligent care system comprises the steps of collecting multi-source heterogeneous data of the old people in real time through an intelligent sensor, preprocessing the multi-source heterogeneous data to generate a structured multi-mode time sequence data set, fusing the structured multi-mode time sequence data set, pre-stored life experience data and health files, constructing and dynamically updating a personal full-dimensional knowledge base, retrieving associated memory materials from the personal full-dimensional knowledge base according to emotion requirement identification results, generating multi-mode accompanying content through AIGC, constructing a structured interaction script based on health risk assessment results and the multi-mode accompanying content, and executing health diagnosis assistance and accompanying content in a cross-equipment scheduling mode. The invention achieves the active senile accompanying with physical and mental integration and memory continuity.

Inventors

LIANG BINGJIN
TU RUIWEI
ZHANG QIAN

Assignees

广东食品药品职业学院

Dates

Publication Date: 20260505
Application Date: 20260123

Claims (10)

1. A AIGC-based health diagnosis and accompanying method for the elderly is characterized by comprising the following steps of, The multi-source heterogeneous data of the old are collected in real time through an intelligent sensor and preprocessed, and a structured multi-mode time sequence data set is generated; Fusing the structured multi-mode time sequence data set, the pre-stored life experience data and the health file, and constructing and dynamically updating a personal full-dimensional knowledge base; based on the structured multi-mode time sequence data set and the personal full-dimensional knowledge base, analyzing the health risk state and emotion demand state, generating a health risk assessment result, and identifying emotion demand; According to emotion demand recognition results, retrieving associated memory materials from a personal full-dimensional knowledge base, and generating multi-mode accompanying content through AIGC; Based on the health risk assessment result and the multi-mode accompanying content, a structured interactive script is constructed, and health diagnosis assistance and accompanying content are cooperatively executed in a cross-device scheduling mode.
2. The method for diagnosis and accompanying health of elderly people based on AIGC according to claim 1, wherein the multi-source heterogeneous data includes physiological data, behavioral data, environmental data, and emotional-rich interaction data; The preprocessing includes data cleaning, denoising, calibration, and timing alignment.
3. The method for diagnosis and accompanying health of elderly people based on AIGC, as set forth in claim 2, wherein the steps of fusing the structured multi-modal time series dataset, pre-stored personal experience data and health profile are as follows, Performing time alignment and semantic matching on the structured multi-mode time sequence data set and pre-stored life experience data to obtain a life experience fusion result; and comparing and analyzing the structured multi-modal time sequence data set with the pre-stored health files, and calculating the difference value between the current value of the physiological parameter and the personal health baseline index to obtain a health file fusion result.
4. The senior health diagnosis and accompanying method based on AIGC according to claim 3, wherein the steps of constructing and dynamically updating the personal full-dimensional knowledge base are as follows, Constructing a experience map and memory association network according to the fusion result of the life experience and the health file fusion result, and generating an initial personal full-dimensional knowledge base; And dynamically updating the initial personal full-dimensional knowledge base based on the structured multi-mode time sequence data set to generate a dynamically updated personal full-dimensional knowledge base.
5. The method for the diagnosis and accompanying health of elderly people based on AIGC according to claim 4, wherein the step of generating a health risk assessment result is as follows, Calculating the deviation degree of the current health state and the individual health baseline index based on the structured multi-mode time sequence data set and the pre-stored health file, and obtaining the health state deviation data; And comprehensively judging the degree of abnormality of physiological parameters, behavior pattern change and medication compliance in the health state deviation data through a green-yellow-red three-level early warning standard to generate a health risk assessment result.
6. The method for diagnosing and accompanying health of elderly people based on AIGC, wherein the emotion requirement recognition is performed as follows, Carrying out keyword matching and emotion dictionary analysis on the rich emotion interaction data and the behavior data, extracting emotion keywords, and identifying the current emotion theme of the user; And correlating the current emotion theme of the user with important life node events in the experience map, judging that the user is in the emotion state of being in the idea, solitary and needing social support, and generating an emotion demand recognition result.
7. The senior health diagnosis and accompanying method based on AIGC according to claim 6, wherein the generating multi-modal accompanying content comprises the steps of, Carrying out semantic matching on emotion labels in emotion demand recognition results and metadata labels of multi-mode memory materials in a personal full-dimensional knowledge base, and retrieving candidate multi-mode memory materials associated with nodes subjected to map events from a memory association network; Calculating semantic similarity between the emotion tag and metadata tags of candidate multi-mode memory materials, sorting according to similarity scores, and taking the multi-mode memory material with the highest score as a selected multi-mode memory material; and generating multimode accompanying content by AIGC taking the selected multimode memory material as a context prompt.
8. The method for diagnosis and accompanying health of elderly people based on AIGC according to claim 7, wherein the steps of constructing a structured interactive script based on the health risk assessment result and the multi-modal accompanying content are as follows, Based on the early warning level in the health risk assessment result, generating a corresponding health prompt, setting the upper limit of the playing duration of the accompanying content, arranging the health prompt to play before the start of the multi-mode accompanying content, and generating a health intervention programming instruction; and integrating the health prompt and the multi-mode accompanying content in time sequence according to the playing precedence relationship and the upper limit of the duration to obtain the structured interaction scenario.
9. The method for health diagnosis and accompanying for senior citizens based on AIGC according to claim 8, wherein the steps of cooperatively performing health diagnosis assistance and accompanying contents by a cross-device scheduling method are as follows, Analyzing the equipment instruction in the structured interaction scenario, distributing different contents to corresponding target terminal equipment, and outputting a mapping distribution result of the contents and the equipment; Based on the mapping distribution result of the content and the equipment, each target terminal equipment starts playing according to the interactive time sequence in the structured interactive scenario in a cross-equipment scheduling mode, and the playing synchronization and the time sequence alignment are kept.
10. A AIGC-based health diagnosis and accompanying system for the elderly, based on the AIGC-based health diagnosis and accompanying method of any one of claims 1 to 9, is characterized by comprising, The data preprocessing module is used for acquiring multi-source heterogeneous data of the old through the intelligent sensor in real time, preprocessing the multi-source heterogeneous data and generating a structured multi-mode time sequence data set; The knowledge fusion module is used for fusing the structured multi-mode time sequence data set, the pre-stored life experience data and the health file, and constructing and dynamically updating the personal full-dimensional knowledge base; the state identification module is used for analyzing the health risk state and emotion demand state based on the structured multi-mode time sequence data set and the personal full-dimensional knowledge base, generating a health risk assessment result and identifying emotion demand; The content generation module is used for retrieving the associated memory materials from the personal full-dimensional knowledge base according to the emotion demand identification result and generating multi-mode accompanying content through AIGC; and the collaborative execution module is used for constructing a structured interaction script based on the health risk assessment result and the multi-mode accompanying content and executing health diagnosis assistance and accompanying content in a collaborative manner in a cross-equipment scheduling mode.

Description

AIGC-based health diagnosis and accompanying system and method for old people Technical Field The invention relates to the technical field of intelligent endowment, in particular to a AIGC-based health diagnosis and accompanying system and method for the elderly. Background The health diagnosis and accompanying technology of the elderly based on AIGC takes a vital role in the current intelligent care field, aims to realize dynamic monitoring of physiological states of the elderly and active response of emotion demands through multi-mode sensing data fusion, individual knowledge modeling and generation type content synthesis, generally combines a wearable device to acquire physiological indexes, utilizes an environment sensor to acquire living parameters, captures interaction behaviors through a voice or video terminal, further generates health prompts or standardized accompanying content based on preset rules, and also partially tries to combine life experience information to enhance interaction affinity, integrally shows a technical trend of evolution to personalized and active care of the elderly, and gradually combines a mode identification technology to improve understanding capability of the user states. In the field of AIGC-based senile health diagnosis and accompanying, the traditional senile health diagnosis and accompanying method has the defects that firstly, the health diagnosis and emotion accompanying function are mutually split, physiological risk early warning is not linked with emotion state identification, so that intervention content lacks physical and psychological cooperativity, secondly, accompanying content generation depends on a general corpus or a static template, real-time emotion subjects cannot be deeply associated with life node events and memory materials with high emotion value in a personal full-dimensional knowledge base, and resonance accompanying with individual memory anchor points is difficult to realize. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides a AIGC-based old person health diagnosis and accompanying method for solving the problems of deep fusion of health diagnosis and emotion accompanying and cutting and lack of personal experience of accompanying content. In order to solve the technical problems, the invention provides the following technical scheme: The invention provides a method for diagnosing and accompanying health of the elderly based on AIGC, which comprises the steps of collecting multi-source heterogeneous data of the elderly in real time through an intelligent sensor, preprocessing the multi-source heterogeneous data to generate a structured multi-mode time sequence data set, fusing the structured multi-mode time sequence data set, pre-stored personal experience data and health files to construct and dynamically update a personal full-dimensional knowledge base, analyzing health risk states and emotion requirement states based on the structured multi-mode time sequence data set and the personal full-dimensional knowledge base to generate health risk assessment results and emotion requirement identification, retrieving associated memory materials from the personal full-dimensional knowledge base according to the emotion requirement identification results, generating multi-mode accompanying content through AIGC, constructing a structured interaction script based on the health risk assessment results and the multi-mode accompanying content, and executing health diagnosis assistance and accompanying content in a cross-device scheduling mode. As a preferable scheme of the AIGC-based old person health diagnosis and accompanying method, the multi-source heterogeneous data comprises physiological data, behavioral data, environmental data and emotion-rich interaction data; The preprocessing includes data cleaning, denoising, calibration, and timing alignment. As a preferable scheme of the AIGC-based senior health diagnosis and accompanying method, the method for diagnosing and accompanying senior citizens of the invention comprises the steps of fusing a structured multi-modal time sequence data set, pre-stored life experience data and health files, Performing time alignment and semantic matching on the structured multi-mode time sequence data set and pre-stored life experience data to obtain a life experience fusion result; and comparing and analyzing the structured multi-modal time sequence data set with the pre-stored health files, and calculating the difference value between the current value of the physiological parameter and the personal health baseline index to obtain a health file fusion result. As a preferable scheme of the AIGC-based old person health diagnosis and accompanying method, the method constructs and dynamically updates a personal full-dimensional knowledge base, and comprises the following steps, Constructing a experience map