An artificial intelligence-based home-based elderly care intelligent monitoring system and method

Abstract

The present invention discloses an artificial intelligence-based smart monitoring system and method for home-based elderly care, which relates to the field of artificial intelligence technology. The method comprises the following steps: monitoring the physiological and behavioral data of a monitored person through sensors; performing cleaning preprocessing, normalization processing, and synchronous integration on the acquired data; using a model to analyze the behavioral data to identify anomalies, constructing a labeled data set, constructing a regression model for judging the current state of the monitored person through a neural network, and performing a confidence assessment to obtain a confidence level; calculating a quantitative index of transmission credibility, and adjusting the data transmission interval according to the calculated confidence level and the quantitative index of transmission credibility. The present invention realizes smart monitoring for home-based elderly care.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based home-based elderly care intelligent monitoring system and method. Background Art

[0002] With the continuous deepening of population aging and the rapid development of science and technology, home-based elderly care has gradually become an important way of elderly care. The widespread application of smart home technology, sensor technology and artificial intelligence technology has provided strong support for the realization of high-quality smart monitoring for home-based elderly care. In the home-based elderly care scenario, the use of various smart devices to provide all-round monitoring of the elderly under care can not only reduce the care burden of family members, but also improve the quality of life of the elderly and ensure their life, health and safety.

[0003] However, existing home-based elderly care monitoring methods face many problems in practical applications: first, there is a lack of monitoring of the physiological and behavioral problems of the monitored persons, and it is impossible to timely discover potential hidden dangers of the elderly; second, during the monitoring process, the guardians cannot always successfully receive the monitoring reports of the monitored persons, and the stability and reliability of data transmission are insufficient, resulting in the inability to guarantee the timeliness and completeness of the monitoring information. Once the monitored persons have health problems, the opportunity for treatment may be delayed. Therefore, in order to solve the above problems, there is an urgent need for an artificial intelligence-based home-based elderly care intelligent monitoring system and method to achieve comprehensive monitoring of the physiological and behavioral problems of the monitored persons and provide more reliable protection for home-based elderly care. Summary of the Invention

[0004] The purpose of the present invention is to provide an artificial intelligence-based home-based elderly care intelligent monitoring system and method to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an artificial intelligence-based home-based elderly care intelligent monitoring method, the method comprising the following steps:

[0006] Monitor the physiological and behavioral data of the monitored person through sensors;

[0007] Clean, pre-process, normalize and synchronize the acquired data;

[0008] Use the model to analyze behavioral data to identify anomalies, build a labeled data set, and use a neural network to build a regression model for determining the current status of the monitored person, and perform a confidence assessment to obtain the confidence level;

[0009] A quantitative index of transmission reliability is calculated, and a data transmission interval is adjusted according to the calculated confidence and the quantitative index of transmission reliability.

[0010] Sensors are used to monitor the physiological, behavioral, conversational, and facial expression data of the monitored person. The specific steps include:

[0011] Sensors are used to obtain physiological, behavioral, conversational, and facial expression data of the monitored person. Physiological data includes the monitored person's heart rate, blood pressure, and body temperature; behavioral data includes the monitored person's movement trajectory, changes in body posture, and information about the monitored person's interaction with the environment.

[0012] Obtain heart rate, blood pressure, and body temperature data of monitored persons through wearable devices and home medical monitoring devices equipped with sensors;

[0013] Use indoor positioning sensors, inertial sensors, and camera equipment to obtain the activity trajectory and body posture changes of the monitored person;

[0014] Through smart home sensors and camera image recognition technology, information about the interaction between the monitored person and the environment is collected;

[0015] Obtain the conversation content between the monitored person and the smart speaker, including daily conversation content, voice tone, speaking duration, and conversation frequency data;

[0016] Using smart cameras to collect facial expressions of the monitored person, the smart cameras are installed in public areas and recognize the facial expressions of the monitored person through image recognition technology and deep learning algorithms while respecting the privacy of the monitored person;

[0017] The physiological, behavioral, conversational, and facial expression data of the monitored person are collected synchronously at certain time intervals. The acquired data are sorted in chronological order [t1, t2, ..., tw] to form a time series of the physiological, behavioral, conversational, and facial expression data of the monitored person, where t1, t2, ..., tw represent the 1st, 2nd, ..., wth data collection moments, respectively.

[0018] The collected physiological, behavioral, conversational, and facial expression data of the monitored person are cleaned, pre-processed, normalized, and synchronously integrated. The specific steps include:

[0019] The collected physiological, behavioral, conversational, and facial expression data of the monitored individuals are cleaned and preprocessed to remove noise data. When data values at a certain moment are missing, they are filled in using the data values of the adjacent moments using a time series interpolation method. The data format is also standardized, and all data are stored in a unified database table structure.

[0020] The Min-Max normalization algorithm is used to process the physiological data of monitored persons of different magnitudes and uniformly map the data to the [0,1] interval;

[0021] The physiological, behavioral, conversational, and facial expression data of the monitored person are synchronously integrated in chronological order. Based on the synchronously collected time points [t1, t2, ..., tw], the heart rate, blood pressure, body temperature, activity trajectory, changes in body posture, interaction information between the monitored person and the environment, conversational content, and facial expression data at the same moment are correlated and combined to form a complete time series data record.

[0022] The Transformer machine learning model analyzes behavioral data to identify abnormal behaviors, constructs a labeled dataset, adjusts thresholds based on cultural habits and disease characteristics to achieve personalized settings, builds a regression model to determine the current status of the monitored person, and sends the monitored person's status to the monitoring personnel. The specific steps include:

[0023] Based on the pre-processed data, the behavioral data of the monitored persons is analyzed and modeled using the Transformer machine learning model. Through feature extraction and pattern recognition, abnormal behaviors of the monitored persons are identified from the time series and action pattern dimensions of the behavioral data. All identified abnormal behaviors are combined into an abnormal behavior set.

[0024] Build an annotated dataset for distinguishing specific behavioral patterns and abnormal action representations;

[0025] For behavior types corresponding to specific cultural backgrounds, social customs, or long-term personal habits, we extract and describe features in terms of the periodicity, duration, and amplitude of the movements. Based on the analysis of normal samples, we summarize the characteristic patterns of such behaviors under normal conditions.

[0026] For abnormal movement types associated with diseases, characterize them based on the frequency of movement, regularity of movement presentation, and the limb parts involved. By comparing and analyzing them with normal behavior, determine the abnormal characteristic patterns of such movements.

[0027] Personalized threshold setting based on historical user data. For behavior types corresponding to specific cultural backgrounds, social customs, or long-established personal habits, the characteristic thresholds described are adjusted to specific multiples of the initial values in the behavior pattern knowledge base, where the behavior pattern knowledge base is constructed based on historical behavior data.

[0028] For abnormal action types associated with diseases, the characteristic thresholds described are adjusted to specific multiples of the initial values in the behavioral pattern knowledge base;

[0029] A regression model for determining the current status of the monitored person is constructed by integrating a neural network framework that integrates multi-source features. The real-time physiological, behavioral, conversational, and facial expression data of the monitored person, as well as the set alarm threshold, are processed using an embedded coding method. The encoded data is input into a constructed neural network regression model that integrates fully connected layers and convolutional layers. The convolutional layers are used to mine associations in local features of the data, and the fully connected layers are used to integrate global features. By learning and mapping the association patterns between the input data features, the health index score of the current monitored person is output to determine the current status of the monitored person.

[0030] Using a cross-validation method, a confidence evaluation is performed on the health index score output by the regression model for determining the current status of the monitored person to obtain a confidence Z;

[0031] The health index score of the current monitored person is sent to the monitoring person.

[0032] Calculate a quantitative index of transmission reliability, and adjust the data transmission interval based on the calculated confidence and the quantitative index of transmission reliability. The specific steps include:

[0033] The GNN, a graph neural network that integrates spatiotemporal features, predicts the factors that influence the transmission of the monitored person's health index score to the monitoring person. Multi-source data related to the transmission is encoded in the form of a graph structure and input into the constructed GNN model. The model outputs the predicted values of the factors that influence the transmission of the monitored person's health index score by learning the spatial associations and temporal evolution patterns in the data.

[0034] Constructing a transmission credibility evaluation set V={v1,v2,...,vm} based on the predicted values of the influencing factors in the process of transmitting the health index score of the monitored person as an evaluation factor set U={u1,u2,...,un}, wherein u1,u2,...,un represent the 1st, 2nd,...,nth predicted values of the influencing factors in the process of transmitting the health index score of the monitored person, and wherein v1,v2,...,vm represent the 1st, 2nd,...,mth level classification of the transmission credibility;

[0035] According to the statistical analysis of historical data, the membership relationship between the factor ui and each comment in the comment set V is determined, and the fuzzy relationship matrix R is established;

[0036] Specifically, the historical data statistical analysis obtains that the membership of factor u1 to each comment is r11, r12, ..., r1m, and the membership of factor u2 to each comment is r21, r22, ..., r2m, and so on, to obtain the membership of factors u3, ..., un to each comment;

[0037] According to the membership relationship between the evaluation factor set and each comment in the comment set V, a fuzzy relationship matrix R is established;

[0038] The fuzzy relationship matrix R is expressed as: ;

[0039] The Delphi method is used to determine the relative importance of each influencing factor in evaluating transmission credibility, that is, to determine the factor weight vector, which is denoted as A=(a1,a2,...,an), where a1,a2,...,an represents the influence weight of the 1st, 2nd,...,nth factor on transmission credibility;

[0040] The factor weight vector A and the fuzzy relationship matrix R are combined through weighted operation to obtain the fuzzy comprehensive evaluation result vector B. The calculation formula is as follows: B=A*R=(b1,b2,...,bm);

[0041] in, , j = 1, 2, ..., m, b1, b2, ..., bm represent the degree values of transmission credibility determined as levels v1, v2, ..., vm after the factor weight vector A and the fuzzy relationship matrix R are synthesized by weighted operation;

[0042] Assign a corresponding score to each comment in the transmission credibility review set, and calculate the weighted average of the fuzzy comprehensive evaluation result vector B and the corresponding score assigned to each comment in the transmission credibility review set as the quantitative index Q of transmission credibility;

[0043] The time interval for transmitting the health index score of the monitored person to the guardian is readjusted according to the calculated confidence Z and the quantitative index Q of transmission credibility. The time interval adjustment formula is as follows: T=T0+β*Z*Q; where β represents the adjustment coefficient and T0 represents the initially set time interval.

[0044] An artificial intelligence-based home-based elderly care intelligent monitoring system, the system includes a data acquisition module, a data processing module, an anomaly analysis and status assessment module, a transmission credibility assessment module and a transmission control module. The data acquisition module is used to monitor the physiological and behavioral data of the monitored person through sensors; the data processing module is used to clean, pre-process, normalize and synchronously integrate the acquired data; the anomaly analysis and status assessment module is used to use a model to analyze behavioral data to identify anomalies, construct a labeled data set, construct a regression model for judging the current status of the monitored person through a neural network, and perform confidence assessment to obtain confidence; the transmission credibility assessment module is used to calculate a quantitative index of transmission credibility; and the transmission control module adjusts the data transmission interval according to the calculated confidence and the quantitative index of transmission credibility.

[0045] The data acquisition module includes a physiological data acquisition unit and a behavioral data acquisition unit. The physiological data acquisition unit is used to obtain the heart rate, blood pressure and body temperature data of the monitored person through sensors. The behavioral data acquisition unit is used to obtain the activity trajectory, body posture changes and interaction information between the monitored person and the environment through sensors.

[0046] The data processing module includes a preprocessing unit, a normalization processing unit and a synchronization integration unit. The preprocessing unit is used to remove noise data from the collected physiological and behavioral data, fill in missing data with a time series-based interpolation method, standardize the data format and store it in a unified database table structure; the normalization processing unit is used to use the Min-Max normalization algorithm to process the physiological data of monitored persons of different magnitudes and uniformly map the data to the [0,1] interval; the synchronization integration unit is used to associate and combine the heart rate, blood pressure, body temperature, activity trajectory, body posture changes, and environmental interaction information at the same time in accordance with the acquisition time sequence and based on the synchronous acquisition time point to form a complete time series data record.

[0047] The abnormal analysis and status assessment module includes an abnormal behavior identification unit, a regression model construction unit and a confidence assessment unit. The abnormal behavior identification unit is used to extract features and perform pattern recognition from the time series and action pattern dimensions of the behavior data based on preprocessed data using the Transformer machine learning model to find out the abnormal behavior of the monitored person and form an abnormal behavior set; the regression model construction unit is used to embed and encode the physiological and behavioral data of the monitored person obtained in real time and the set alarm threshold through a neural network framework that integrates multi-source features, input them into a neural network regression model that integrates a fully connected layer and a convolutional layer, use the convolutional layer to mine local feature associations of the data, and the fully connected layer to integrate global features, learn the association pattern between the input data features, output the health index score of the monitored person, and judge its current status; the confidence assessment unit is used to use a cross-validation method to perform confidence assessment on the health index score output by the regression model to obtain confidence.

[0048] The transmission credibility assessment module includes an influencing factor prediction unit, a fuzzy relationship construction unit and a comprehensive evaluation unit. The influencing factor prediction unit is used to use a graph neural network (GNN) that integrates spatiotemporal features to encode multi-source data related to transmission in a graph structure input model, learn the spatial association and time evolution pattern in the data, and output the predicted value of the influencing factors in the process of transmitting the health index score of the monitored person; the fuzzy relationship construction unit is used to use the predicted value of the influencing factor as the evaluation factor set, construct a comment set of transmission credibility, determine the membership relationship between the factors and each comment in the comment set based on historical data statistical analysis, and establish a fuzzy relationship matrix; the comprehensive evaluation unit is used to synthesize the factor weight vector and the fuzzy relationship matrix through weighted operation to obtain a fuzzy comprehensive evaluation result vector, assign a corresponding score to each comment in the transmission credibility comment set, and calculate the weighted average of the fuzzy comprehensive evaluation result vector and the corresponding score of the comment set as a quantitative indicator of transmission credibility.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. Use graph neural network (GNN) to predict factors affecting data transmission, combine it with fuzzy comprehensive evaluation method to calculate the transmission credibility quantitative index, and dynamically adjust the data transmission interval according to the confidence level and the index to ensure that the monitoring data is transmitted to the monitoring personnel;

[0051] 2. Use artificial intelligence technology to comprehensively analyze the physiological and behavioral data of the monitored person, combine historical data and personalized threshold settings, and identify abnormal behaviors and potential health risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flow chart of an artificial intelligence-based home-based elderly care intelligent monitoring method of the present invention;

[0053] Figure 2 This is a structural diagram of an artificial intelligence-based home-based elderly care intelligent monitoring system of the present invention. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] In the embodiment: Figure 1-Figure 2 As shown, the present invention provides a technical solution, an artificial intelligence-based home-based elderly care intelligent monitoring method, the method comprising the following steps:

[0056] Monitor the physiological and behavioral data of the monitored person through sensors;

[0057] Clean, pre-process, normalize and synchronize the acquired data;

[0058] Use the model to analyze behavioral data to identify anomalies, build a labeled data set, and use a neural network to build a regression model for determining the current status of the monitored person, and perform a confidence assessment to obtain the confidence level;

[0059] A quantitative index of transmission reliability is calculated, and a data transmission interval is adjusted according to the calculated confidence and the quantitative index of transmission reliability.

[0060] Sensors are used to monitor the physiological, behavioral, conversational, and facial expression data of the monitored person. The specific steps include:

[0061] Sensors are used to obtain physiological, behavioral, conversational, and facial expression data of the monitored person. Physiological data includes the monitored person's heart rate, blood pressure, and body temperature; behavioral data includes the monitored person's movement trajectory, changes in body posture, and information about the monitored person's interaction with the environment.

[0062] Obtain heart rate, blood pressure, and body temperature data of monitored persons through wearable devices and home medical monitoring devices equipped with sensors;

[0063] Use indoor positioning sensors, inertial sensors, and camera equipment to obtain the activity trajectory and body posture changes of the monitored person;

[0064] Through smart home sensors and camera image recognition technology, information about the interaction between the monitored person and the environment is collected;

[0065] Obtain the conversation content between the monitored person and the smart speaker, including daily conversation content, voice tone, speaking duration, and conversation frequency data;

[0066] Using smart cameras to collect facial expressions of the monitored person, the smart cameras are installed in public areas and recognize the facial expressions of the monitored person through image recognition technology and deep learning algorithms while respecting the privacy of the monitored person;

[0067] Synchronously collect the physiological, behavioral, conversational, and facial expression data of the monitored person at regular intervals, and sort the acquired data in chronological order [t1, t2, ..., tw] to form a time series of the physiological, behavioral, conversational, and facial expression data of the monitored person, where t1, t2, ..., tw represent the 1st, 2nd, ..., wth data collection moments, respectively;

[0068] Specifically, the physiological data of an elderly person was collected through a smart bracelet with sensors and home medical equipment. Data was collected every 5 minutes between 9 and 10 a.m. on a certain day. At 9 a.m., the heart rate was 75 beats / minute, the blood pressure was 120 / 80 mmHg, and the body temperature was 36.5°C. At the same time, behavioral data was obtained using positioning sensors, inertial sensors and cameras installed indoors. At 9:10 a.m., the elderly person walked from the bedroom to the living room with an upright posture, and the interactive information with the environment showed that the TV in the living room was turned on. The content of the elderly person's conversation was recorded through a smart speaker. During this hour, the elderly person had 5 conversations with the smart speaker, with an average conversation duration of 2 minutes each time. The voice and tone were steady, and the smart camera recognized the facial expression and inferred that the state was calm.

[0069] The collected physiological, behavioral, conversational, and facial expression data of the monitored person are cleaned, pre-processed, normalized, and synchronously integrated. The specific steps include:

[0070] The collected physiological, behavioral, conversational, and facial expression data of the monitored individuals are cleaned and preprocessed to remove noise data. When data values at a certain moment are missing, they are filled in using the data values of the adjacent moments using a time series interpolation method. The data format is also standardized, and all data are stored in a unified database table structure.

[0071] The Min-Max normalization algorithm is used to process the physiological data of monitored persons of different magnitudes and uniformly map the data to the [0,1] interval;

[0072] The physiological, behavioral, conversational, and facial expression data of the monitored person are synchronously integrated in chronological order. Based on the synchronously collected time points [t1, t2, ..., tw], the heart rate, blood pressure, body temperature, activity trajectory, changes in body posture, interaction information between the monitored person and the environment, conversational content, and facial expression data at the same moment are correlated and combined to form a complete time series data record.

[0073] The Transformer machine learning model analyzes behavioral data to identify abnormal behaviors, constructs a labeled dataset, adjusts thresholds based on cultural habits and disease characteristics to achieve personalized settings, builds a regression model to determine the current status of the monitored person, and sends the monitored person's status to the monitoring personnel. The specific steps include:

[0074] Based on the pre-processed data, the behavioral data of the monitored persons is analyzed and modeled using the Transformer machine learning model. Through feature extraction and pattern recognition, abnormal behaviors of the monitored persons are identified from the time series and action pattern dimensions of the behavioral data. All identified abnormal behaviors are combined into an abnormal behavior set.

[0075] Build an annotated dataset for distinguishing specific behavioral patterns and abnormal action representations;

[0076] For behavior types corresponding to specific cultural backgrounds, social customs, or long-term personal habits, we extract and describe features in terms of the periodicity, duration, and amplitude of the movements. Based on the analysis of normal samples, we summarize the characteristic patterns of such behaviors under normal conditions.

[0077] For abnormal movement types associated with diseases, characterize them based on the frequency of movement, regularity of movement presentation, and the limb parts involved. By comparing and analyzing them with normal behavior, determine the abnormal characteristic patterns of such movements.

[0078] Personalized threshold setting based on historical user data. For behavior types corresponding to specific cultural backgrounds, social customs, or long-established personal habits, the characteristic thresholds described are adjusted to specific multiples of the initial values in the behavior pattern knowledge base, where the behavior pattern knowledge base is constructed based on historical behavior data.

[0079] For abnormal action types associated with diseases, the characteristic thresholds described are adjusted to specific multiples of the initial values in the behavioral pattern knowledge base;

[0080] A regression model for determining the current status of the monitored person is constructed by integrating a neural network framework that integrates multi-source features. The real-time physiological, behavioral, conversational, and facial expression data of the monitored person, as well as the set alarm threshold, are processed using an embedded coding method. The encoded data is input into a constructed neural network regression model that integrates fully connected layers and convolutional layers. The convolutional layers are used to mine associations in local features of the data, and the fully connected layers are used to integrate global features. By learning and mapping the association patterns between the input data features, the health index score of the current monitored person is output to determine the current status of the monitored person.

[0081] Using a cross-validation method, a confidence evaluation is performed on the health index score output by the regression model for determining the current status of the monitored person to obtain a confidence Z;

[0082] Send the health index score of the current supervised person to the supervisor;

[0083] Specifically, the Transformer machine learning model was used to analyze behavioral data. Based on the fact that the elderly usually engage in simple indoor activities such as reading or watching TV between 9:00 and 10:00 a.m., their movement patterns are relatively regular. At 9:40 a.m., the model detected that the elderly person remained abnormally still in the living room for a long time, which did not conform to their daily behavior patterns. This behavior was identified as abnormal and a labeled dataset was constructed. For the elderly person's morning exercise behavior, which usually takes place at 9:30 a.m. every day, features were extracted from the periodicity, duration, and amplitude of the movements to determine the characteristic patterns under normal conditions. A personalized threshold was set. The initial threshold for the amplitude of morning exercise movements was 5 (unit: degrees) based on the behavioral pattern knowledge base. Based on the elderly person's historical data, it was adjusted to 8 degrees. A regression model was constructed using a neural network framework that integrated multi-source features. Real-time physiological, behavioral, conversation content, and facial expression data, as well as the set alarm threshold, were embedded and encoded before being input into the model. Based on the processed input data, the model output a health index score of 70 (out of 100). Cross-validation was used to assess the confidence of the health index score, assuming the obtained confidence level Z was 0.8.

[0084] Calculate a quantitative index of transmission reliability, and adjust the data transmission interval based on the calculated confidence and the quantitative index of transmission reliability. The specific steps include:

[0085] The GNN, a graph neural network that integrates spatiotemporal features, predicts the factors that influence the transmission of the monitored person's health index score to the monitoring person. Multi-source data related to the transmission is encoded in the form of a graph structure and input into the constructed GNN model. The model outputs the predicted values of the factors that influence the transmission of the monitored person's health index score by learning the spatial associations and temporal evolution patterns in the data.

[0086] Constructing a transmission credibility evaluation set V={v1,v2,...,vm} based on the predicted values of the influencing factors in the process of transmitting the health index score of the monitored person as an evaluation factor set U={u1,u2,...,un}, wherein u1,u2,...,un represent the 1st, 2nd,...,nth predicted values of the influencing factors in the process of transmitting the health index score of the monitored person, and wherein v1,v2,...,vm represent the 1st, 2nd,...,mth level classification of the transmission credibility;

[0087] According to the statistical analysis of historical data, the membership relationship between the factor ui and each comment in the comment set V is determined, and the fuzzy relationship matrix R is established;

[0088] Specifically, the historical data statistical analysis obtains that the membership of factor u1 to each comment is r11, r12, ..., r1m, and the membership of factor u2 to each comment is r21, r22, ..., r2m, and so on, to obtain the membership of factors u3, ..., un to each comment;

[0089] According to the membership relationship between the evaluation factor set and each comment in the comment set V, a fuzzy relationship matrix R is established;

[0090] The fuzzy relationship matrix R is expressed as: ;

[0091] The Delphi method is used to determine the relative importance of each influencing factor in evaluating transmission credibility, that is, to determine the factor weight vector, which is denoted as A=(a1,a2,...,an), where a1,a2,...,an represents the influence weight of the 1st, 2nd,...,nth factor on transmission credibility;

[0092] The factor weight vector A and the fuzzy relationship matrix R are combined through weighted operation to obtain the fuzzy comprehensive evaluation result vector B. The calculation formula is as follows: B=A*R=(b1,b2,...,bm);

[0093] in, , j = 1, 2, ..., m, b1, b2, ..., bm represent the degree values of transmission credibility determined as levels v1, v2, ..., vm after the factor weight vector A and the fuzzy relationship matrix R are synthesized by weighted operation;

[0094] Assign a corresponding score to each comment in the transmission credibility review set, and calculate the weighted average of the fuzzy comprehensive evaluation result vector B and the corresponding score assigned to each comment in the transmission credibility review set as the quantitative index Q of transmission credibility;

[0095] The time interval for transmitting the monitored person's health index score to the guardian is readjusted based on the calculated confidence Z and the quantitative index Q of transmission credibility. The time interval adjustment formula is as follows: T=T0+β*Z*Q; where β represents the adjustment coefficient and T0 represents the initially set time interval;

[0096] Specifically, the factors affecting data transmission are predicted by the graph neural network (GNN) that integrates spatiotemporal features. Assuming that the influencing factors include network signal strength, device load, data flow, etc., the predicted values of these factors are used as the evaluation factor set U, and the transmission credibility comment set V is constructed as {high credibility, medium credibility, low credibility}. Based on the statistical analysis of historical data, the membership relationship between the factors and the comment set is determined. The membership of the network signal strength factor u1 to high credibility, medium credibility, and low credibility is 0.6, 0.3, and 0.1, respectively, and the fuzzy relationship matrix R is established;

[0097] The Delphi method was used to determine the factor weight vector A = (0.4, 0.3, 0.3). The factor weight vector A was then combined with the fuzzy relationship matrix R through a weighted operation to obtain the fuzzy comprehensive evaluation result vector B: B = A * R = (0.52, 0.35, 0.13). Each comment in the transmission credibility review set was assigned a corresponding score, with high credibility being 90, medium credibility being 70, and low credibility being 50. The weighted average of the fuzzy comprehensive evaluation result vector B and the corresponding scores of the review set was calculated as the quantitative indicator Q of transmission credibility. The calculated value Q = 78.4.

[0098] Transmission control stage: Based on the initially set time interval T0 of 10 minutes and the adjustment coefficient β of 0.5, according to the time interval adjustment formula, it is calculated that the time interval for transmitting the health index score of the monitored person to the guardian after adjustment is approximately 41 minutes.

[0099] An artificial intelligence-based home-based elderly care intelligent monitoring system, the system includes a data acquisition module, a data processing module, an anomaly analysis and status assessment module, a transmission credibility assessment module and a transmission control module. The data acquisition module is used to monitor the physiological and behavioral data of the monitored person through sensors; the data processing module is used to clean, pre-process, normalize and synchronously integrate the acquired data; the anomaly analysis and status assessment module is used to use a model to analyze behavioral data to identify anomalies, construct a labeled data set, construct a regression model for judging the current status of the monitored person through a neural network, and perform confidence assessment to obtain confidence; the transmission credibility assessment module is used to calculate a quantitative index of transmission credibility; and the transmission control module adjusts the data transmission interval according to the calculated confidence and the quantitative index of transmission credibility.

[0100] The data acquisition module includes a physiological data acquisition unit and a behavioral data acquisition unit. The physiological data acquisition unit is used to obtain the heart rate, blood pressure and body temperature data of the monitored person through sensors. The behavioral data acquisition unit is used to obtain the activity trajectory, body posture changes and interaction information between the monitored person and the environment through sensors.

[0101] The data processing module includes a preprocessing unit, a normalization processing unit and a synchronization integration unit. The preprocessing unit is used to remove noise data from the collected physiological and behavioral data, fill in missing data with a time series-based interpolation method, standardize the data format and store it in a unified database table structure; the normalization processing unit is used to use the Min-Max normalization algorithm to process the physiological data of monitored persons of different magnitudes and uniformly map the data to the [0,1] interval; the synchronization integration unit is used to associate and combine the heart rate, blood pressure, body temperature, activity trajectory, body posture changes, and environmental interaction information at the same time in accordance with the acquisition time sequence and based on the synchronous acquisition time point to form a complete time series data record.

[0102] The abnormal analysis and status assessment module includes an abnormal behavior identification unit, a regression model construction unit and a confidence assessment unit. The abnormal behavior identification unit is used to extract features and perform pattern recognition from the time series and action pattern dimensions of the behavior data based on preprocessed data using the Transformer machine learning model to find out the abnormal behavior of the monitored person and form an abnormal behavior set; the regression model construction unit is used to embed and encode the physiological and behavioral data of the monitored person obtained in real time and the set alarm threshold through a neural network framework that integrates multi-source features, input them into a neural network regression model that integrates a fully connected layer and a convolutional layer, use the convolutional layer to mine local feature associations of the data, and the fully connected layer to integrate global features, learn the association pattern between the input data features, output the health index score of the monitored person, and judge its current status; the confidence assessment unit is used to use a cross-validation method to perform confidence assessment on the health index score output by the regression model to obtain confidence.

[0103] The transmission credibility assessment module includes an influencing factor prediction unit, a fuzzy relationship construction unit and a comprehensive evaluation unit. The influencing factor prediction unit is used to use a graph neural network (GNN) that integrates spatiotemporal features to encode multi-source data related to transmission in a graph structure input model, learn the spatial association and time evolution pattern in the data, and output the predicted value of the influencing factors in the process of transmitting the health index score of the monitored person; the fuzzy relationship construction unit is used to use the predicted value of the influencing factor as the evaluation factor set, construct a comment set of transmission credibility, determine the membership relationship between the factors and each comment in the comment set based on historical data statistical analysis, and establish a fuzzy relationship matrix; the comprehensive evaluation unit is used to synthesize the factor weight vector and the fuzzy relationship matrix through weighted operation to obtain a fuzzy comprehensive evaluation result vector, assign a corresponding score to each comment in the transmission credibility comment set, and calculate the weighted average of the fuzzy comprehensive evaluation result vector and the corresponding score of the comment set as a quantitative indicator of transmission credibility.

[0104] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. An artificial intelligence-based home-based elderly care intelligent monitoring method, characterized by: The method comprises the following steps: Monitor the physiological and behavioral data of the monitored person through sensors; Clean, pre-process, normalize and synchronize the acquired data; Use the model to analyze behavioral data to identify anomalies, build a labeled data set, and use a neural network to build a regression model for determining the current status of the monitored person, and perform a confidence assessment to obtain the confidence level; Calculating a quantitative index of transmission reliability, and adjusting a data transmission interval according to the calculated confidence level and the quantitative index of transmission reliability; The GNN, a graph neural network that integrates spatiotemporal features, predicts the factors that influence the transmission of the monitored person's health index score to the monitoring person. Multi-source data related to the transmission is encoded in the form of a graph structure and input into the constructed GNN model. The model outputs the predicted values of the factors that influence the transmission of the monitored person's health index score by learning the spatial associations and temporal evolution patterns in the data. Constructing a transmission credibility evaluation set V={v1,v2,...,vm} based on the predicted values of the influencing factors in the process of transmitting the health index score of the monitored person as an evaluation factor set U={u1,u2,...,un}, wherein u1,u2,...,un represent the 1st, 2nd,...,nth predicted values of the influencing factors in the process of transmitting the health index score of the monitored person, and wherein v1,v2,...,vm represent the 1st, 2nd,...,mth level classification of the transmission credibility; According to the statistical analysis of historical data, the membership relationship between the factor ui and each comment in the comment set V is determined, and the fuzzy relationship matrix R is established; Specifically, the historical data statistical analysis obtains that the membership of factor u1 to each comment is r11, r12, ..., r1m, and the membership of factor u2 to each comment is r21, r22, ..., r2m, and so on, to obtain the membership of factors u3, ..., un to each comment; According to the membership relationship between the evaluation factor set and each comment in the comment set V, a fuzzy relationship matrix R is established; The fuzzy relationship matrix R is expressed as: ; The Delphi method is used to determine the relative importance of each influencing factor in evaluating transmission credibility, that is, to determine the factor weight vector, which is denoted as A=(a1,a2,...,an), where a1,a2,...,an represents the influence weight of the 1st, 2nd,...,nth factor on transmission credibility; The factor weight vector A and the fuzzy relationship matrix R are combined through weighted operation to obtain the fuzzy comprehensive evaluation result vector B. The calculation formula is as follows: B=A*R=(b1,b2,...,bm); in, , b1, b2, ..., bm represent the degree values of transmission credibility determined as levels v1, v2, ..., vm after the factor weight vector A and the fuzzy relationship matrix R are synthesized through weighted operation, j = 1, 2, ..., m; Assign a corresponding score to each comment in the transmission credibility review set, and calculate the weighted average of the fuzzy comprehensive evaluation result vector B and the corresponding score assigned to each comment in the transmission credibility review set as the quantitative index Q of transmission credibility; The time interval for transmitting the health index score of the monitored person to the guardian is readjusted according to the calculated confidence Z and the quantitative index Q of transmission credibility. The time interval adjustment formula is as follows: T=T0+β*Z*Q; where β represents the adjustment coefficient and T0 represents the initially set time interval.

2. The method for intelligent home-based elderly care monitoring based on artificial intelligence according to claim 1, characterized in that: The physiological and behavioral data of the monitored person are monitored through sensors. The specific steps include: Sensors are used to obtain physiological and behavioral data of the monitored person. Physiological data includes the monitored person's heart rate, blood pressure, and body temperature; behavioral data includes the monitored person's movement trajectory, changes in body posture, and information about the monitored person's interaction with the environment. The physiological and behavioral data of the monitored person are synchronously collected at certain time intervals, and the acquired data are sorted in chronological order [t1, t2, ..., tw] to form a time series of the physiological and behavioral data of the monitored person, where t1, t2, ..., tw represent the 1st, 2nd, ..., wth data collection moments respectively.

3. The method for intelligent home-based elderly care monitoring based on artificial intelligence according to claim 2, characterized in that: The collected physiological and behavioral data of the monitored persons are cleaned, pre-processed, normalized, and synchronously integrated. The specific steps include: The collected physiological and behavioral data of the monitored persons are cleaned and preprocessed to remove noise data. When the data value at a certain moment is missing, the data value at the previous and next moments is filled in using the time series interpolation method. The data format is standardized and all data are stored according to a unified database table structure. The Min-Max normalization algorithm is used to process the physiological data of monitored persons of different magnitudes and uniformly map the data to the [0,1] interval; The physiological and behavioral data of the monitored person are synchronously integrated in the order of collection time. Based on the synchronous collection time points [t1, t2, ..., tw], the heart rate, blood pressure, body temperature, activity trajectory, body posture changes, and interaction information between the monitored person and the environment at the same moment are associated and combined to form a complete time series data record.

4. The method for intelligent home-based elderly care monitoring based on artificial intelligence according to claim 3 is characterized by: The Transformer machine learning model analyzes behavioral data to identify abnormal behaviors, constructs a labeled dataset, adjusts thresholds based on cultural habits and disease characteristics to achieve personalized settings, builds a regression model to determine the current status of the monitored person, and sends the monitored person's status to the monitoring personnel. The specific steps include: Based on the pre-processed data, the behavioral data of the monitored persons is analyzed and modeled using the Transformer machine learning model. Through feature extraction and pattern recognition, abnormal behaviors of the monitored persons are identified from the time series and action pattern dimensions of the behavioral data. All identified abnormal behaviors are combined into an abnormal behavior set. Build an annotated dataset for distinguishing specific behavioral patterns and abnormal action representations; For behavior types corresponding to specific cultural backgrounds, social customs, or long-term personal habits, we extract and describe features in terms of the periodicity, duration, and amplitude of the movements. Based on the analysis of normal samples, we summarize the characteristic patterns of such behaviors under normal conditions. For abnormal movement types associated with diseases, characterize them based on the frequency of movement, regularity of movement presentation, and the limb parts involved. By comparing and analyzing them with normal behavior, determine the abnormal characteristic patterns of such movements. Personalized threshold setting based on historical user data. For behavior types corresponding to specific cultural backgrounds, social customs, or long-established personal habits, the characteristic thresholds described are adjusted to specific multiples of the initial values in the behavior pattern knowledge base, where the behavior pattern knowledge base is constructed based on historical behavior data. For abnormal action types associated with diseases, the characteristic thresholds described are adjusted to specific multiples of the initial values in the behavioral pattern knowledge base; A regression model for determining the current status of the monitored person is constructed through a neural network framework that integrates multi-source features. Real-time physiological and behavioral data of the monitored person, as well as the set alarm thresholds, are processed using embedded coding. The encoded data is then input into a neural network regression model that integrates fully connected and convolutional layers. The convolutional layers mine associations in local features of the data, while the fully connected layers integrate global features. By learning and mapping the association patterns between input data features, the health index score of the current monitored person is output. Using a cross-validation method, a confidence evaluation is performed on the health index score output by the regression model for determining the current status of the monitored person to obtain a confidence Z; The health index score of the current monitored person is sent to the monitoring person.

5. An artificial intelligence-based home-based elderly care intelligent monitoring system, characterized by: The system includes a data acquisition module, a data processing module, an anomaly analysis and status assessment module, a transmission credibility assessment module and a transmission control module. The data acquisition module is used to monitor the physiological and behavioral data of the monitored person through sensors; the data processing module is used to clean, pre-process, normalize and synchronously integrate the acquired data; the anomaly analysis and status assessment module is used to use a model to analyze behavioral data to identify anomalies, build a labeled data set, build a regression model for judging the current status of the monitored person through a neural network, and perform confidence assessment to obtain confidence; the transmission credibility assessment module is used to calculate a quantitative index of transmission credibility; the transmission credibility assessment module includes an influencing factor prediction unit, a fuzzy relationship construction unit and a comprehensive evaluation unit. The influencing factor prediction unit is used to use a graph neural network (GNN) that integrates spatiotemporal features to encode multi-source data related to transmission in the form of a graph structure and input it into the model, learn the spatial association and time evolution pattern in the data, and output the predicted value of the influencing factors in the process of transmitting the health index score of the monitored person; The fuzzy relationship construction unit is used to use the predicted values of the influencing factors as the evaluation factor set to construct a comment set for transmission credibility, determine the membership relationship between the factors and each comment in the comment set based on the statistical analysis of historical data, and establish a fuzzy relationship matrix; the comprehensive evaluation unit is used to combine the factor weight vector and the fuzzy relationship matrix through a weighted operation to obtain a fuzzy comprehensive evaluation result vector, assign a corresponding score to each comment in the transmission credibility comment set, and calculate the weighted average of the fuzzy comprehensive evaluation result vector and the corresponding score of the comment set as a quantitative indicator of transmission credibility; The transmission credibility assessment module predicts the factors affecting the transmission of the monitored person's health index score to the guardian through a graph neural network (GNN) that integrates spatiotemporal features. The multi-source data related to the transmission is encoded in the form of a graph structure and input into the constructed GNN model. The model outputs the predicted values of the factors affecting the transmission of the monitored person's health index score by learning the spatial association and temporal evolution patterns in the data. Constructing a transmission credibility evaluation set V={v1,v2,...,vm} based on the predicted values of the influencing factors in the process of transmitting the health index score of the monitored person as an evaluation factor set U={u1,u2,...,un}, wherein u1,u2,...,un represent the 1st, 2nd,...,nth predicted values of the influencing factors in the process of transmitting the health index score of the monitored person, and wherein v1,v2,...,vm represent the 1st, 2nd,...,mth level classification of the transmission credibility; According to the statistical analysis of historical data, the membership relationship between the factor ui and each comment in the comment set V is determined, and the fuzzy relationship matrix R is established; Specifically, the historical data statistical analysis obtains that the membership of factor u1 to each comment is r11, r12, ..., r1m, and the membership of factor u2 to each comment is r21, r22, ..., r2m, and so on, to obtain the membership of factors u3, ..., un to each comment; According to the membership relationship between the evaluation factor set and each comment in the comment set V, a fuzzy relationship matrix R is established; The fuzzy relationship matrix R is expressed as: ; The Delphi method is used to determine the relative importance of each influencing factor in evaluating transmission credibility, that is, to determine the factor weight vector, which is denoted as A=(a1,a2,...,an), where a1,a2,...,an represents the influence weight of the 1st, 2nd,...,nth factor on transmission credibility; The factor weight vector A and the fuzzy relationship matrix R are combined through weighted operation to obtain the fuzzy comprehensive evaluation result vector B. The calculation formula is as follows: B=A*R=(b1,b2,...,bm); in, , b1, b2, ..., bm represent the degree values of transmission credibility determined as levels v1, v2, ..., vm after the factor weight vector A and the fuzzy relationship matrix R are synthesized through weighted operation, j = 1, 2, ..., m; each comment in the transmission credibility comment set is assigned a corresponding score, and the weighted average of the fuzzy comprehensive evaluation result vector B and the corresponding score assigned to each comment in the transmission credibility comment set is calculated as the quantitative index Q of transmission credibility; The transmission control module adjusts the data transmission interval according to the calculated confidence level and the quantitative index of transmission credibility. The transmission control module readjusts the time interval for transmitting the health index score of the monitored person to the guardian according to the calculated confidence level Z and the quantitative index of transmission credibility Q. The time interval adjustment formula is as follows: T=T0+β*Z*Q; wherein β represents the adjustment coefficient and T0 represents the initially set time interval.

6. The artificial intelligence-based home-based elderly care intelligent monitoring system according to claim 5, characterized in that: The data acquisition module includes a physiological data acquisition unit and a behavioral data acquisition unit. The physiological data acquisition unit is used to obtain the heart rate, blood pressure and body temperature data of the monitored person through sensors. The behavioral data acquisition unit is used to obtain the activity trajectory, body posture changes and interaction information between the monitored person and the environment through sensors.

7. The artificial intelligence-based home-based elderly care intelligent monitoring system according to claim 6, characterized in that: The data processing module includes a preprocessing unit, a normalization processing unit and a synchronization integration unit. The preprocessing unit is used to remove noise data from the collected physiological and behavioral data, fill in missing data with a time series-based interpolation method, standardize the data format and store it in a unified database table structure; the normalization processing unit is used to use the Min-Max normalization algorithm to process the physiological data of monitored persons of different magnitudes and uniformly map the data to the [0,1] interval; the synchronization integration unit is used to associate and combine the heart rate, blood pressure, body temperature, activity trajectory, body posture changes, and environmental interaction information at the same time in accordance with the acquisition time sequence and based on the synchronous acquisition time point to form a complete time series data record.

8. The artificial intelligence-based home-based elderly care intelligent monitoring system according to claim 7, characterized in that: The abnormal analysis and status assessment module includes an abnormal behavior identification unit, a regression model construction unit, and a confidence assessment unit. The abnormal behavior identification unit is used to extract features and perform pattern recognition from the time series and action pattern dimensions of the behavior data based on preprocessed data using the Transformer machine learning model to identify abnormal behaviors of the monitored person and form an abnormal behavior set. The regression model construction unit is used to embed and encode the physiological and behavioral data of the monitored person obtained in real time and the set alarm threshold through a neural network framework that integrates multi-source features, input the data into a neural network regression model that integrates a fully connected layer and a convolutional layer, use the convolutional layer to mine local feature associations of the data, and use the fully connected layer to integrate global features, learn the association pattern between input data features, output the health index score of the monitored person, and judge his current status; the confidence assessment unit is used to use a cross-validation method to perform a confidence assessment on the health index score output by the regression model to obtain confidence.

Images