Internet-of-things-based disabled health care auxiliary management system

Abstract

The invention discloses a disabled health care auxiliary management system based on Internet of Things, and relates to the technical field of medical health management, and the system comprises the following components: a health Internet of Things acquisition module which connects a wearable health monitoring device, a smart home health monitoring device and an environment sensor with a cloud server through a wireless network, the intelligent health analysis module is introduced, advanced algorithm formulas are combined, real-time physiological index data and living environment data from the health internet-of-things acquisition module are deeply mined and analyzed, and the real-time health risk assessment mechanism can accurately capture tiny fluctuations of physiological indexes of a patient, so that the accuracy of health risk assessment is improved. According to the health risk assessment system and method, the historical health data and living habits of the patient are combined, possible health problems are predicted in advance, early warning is sent to the patient, family members and medical staff in the first time in the modes of short messages, APP pushing and voice prompting, and the accurate and timely health risk assessment provides powerful support for early discovery and intervention of diseases.

Description

Technical Field

[0001] The present invention relates to the technical field of medical health management, and in particular to an auxiliary health care management system for the disabled based on the Internet of Things. Background Art

[0002] With the rapid development of Internet of Things technology and the popularization of smart devices, remote health monitoring and management has become an important trend in modern medical services. The application of wearable health monitoring devices, smart home health monitoring devices and environmental sensors makes it possible to collect and transmit patients' physiological indicators and living environment data in real time, providing a solid technical foundation for building a comprehensive and accurate health management system.

[0003] Although traditional technologies have been able to collect and analyze health data to a certain extent, there are still many shortcomings in real-time health risk assessment. First, traditional systems often rely on simple threshold judgments and statistical models, lack the ability to deeply mine and analyze big data, and find it difficult to accurately capture small fluctuations in patients' physiological indicators and potential health risks. Secondly, when predicting health problems, traditional systems often ignore patients' historical health data and living habits, resulting in inaccurate prediction results and difficulty in providing patients with effective health guidance. In addition, traditional systems also have delays and omissions in the sending of early warning information, and cannot ensure that patients, their families, and medical staff can obtain early warning information in a timely manner and take corresponding intervention measures.

[0004] Therefore, an IoT-based health care auxiliary management system for the disabled was developed. By introducing an intelligent health analysis module and combining it with advanced algorithm formulas to conduct in-depth mining and analysis of the collected data, real-time monitoring and evaluation of the patient's physiological indicators and living environment were achieved. At the same time, the system can accurately predict possible health problems based on the patient's current status and historical data, and issue early warning information in a timely manner through multiple channels to ensure that patients, their families, and medical staff can respond quickly and take appropriate intervention measures. Summary of the invention

[0005] The purpose of this invention is to make up for the shortcomings of the existing technology and to provide a health care auxiliary management system for the disabled based on the Internet of Things. It can accurately predict potential health problems by real-time collection and in-depth analysis of data on the physiological indicators, living habits and environmental factors of the disabled, and immediately issue early warnings to relevant personnel, effectively reducing health risks. At the same time, based on the results of intelligent analysis, it can tailor personalized health management plans for each disabled person to help them improve their living habits and improve their health levels.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a health care auxiliary management system for the disabled based on the Internet of Things, the system includes the following components: Health IoT collection module: connects wearable health monitoring devices, smart home health monitoring devices and environmental sensors to the cloud server through wireless networks, and uploads the physiological index data and living environment data of disabled people in real time; Data integration and processing module: Receives data from the health IoT collection module, cleans, integrates, and stores it, builds a health database for people with disabilities, and implements long-term tracking and comprehensive analysis of data; Intelligent health analysis module: Uses algorithm formulas to deeply mine and analyze the data provided by the data integration and processing module. By analyzing the historical health data, current physiological indicators, living habits and environmental factors of the disabled, it can intelligently identify abnormal fluctuations, predict health problems, and calculate risk levels. Early warning and emergency intervention module: Receives risk information from the intelligent health analysis module and immediately sends early warning information to the disabled person, his / her family members, and medical staff through SMS, APP push, and voice prompts. At the same time, it automatically triggers corresponding intervention measures according to the risk level; Personalized health guidance module: Based on the results of intelligent analysis, a personalized health management plan is customized for each disabled person, and it is regularly pushed through the APP channel to help disabled people improve their living habits and improve their health level.

[0007] Furthermore, the intelligent health analysis module receives a data set of historical health data, current physiological indicator data, living habit data, and environmental factor data of the disabled from the data integration and processing module, performs data cleaning, feature selection, and normalization operations, builds a model through an algorithm formula, and divides the preprocessed data set into a training set, a validation set, and a test set, initializes the model parameters, trains the model using the training set data, and evaluates the performance of the model using the validation set data. According to the verification results, the model parameters are adjusted, and the currently collected data is input into the trained model to determine whether the physiological indicators are abnormal and whether there are potential health problems, and predict possible risks in the future. According to the model detection results, the health risk level is evaluated through the risk assessment model, and decision support information is provided to medical staff and the disabled person themselves based on the risk level.

[0008] Furthermore, the intelligent health analysis module constructs a model through a multi-dimensional physiological index calculation formula to determine whether the physiological index is abnormal. ,have feature dimensions, respectively , each feature dimension has a corresponding weight ,satisfy , calculate the deviation between the current physiological index value and the mean value of the normal range , assuming the normal range is , the current value is , the calculation formula is: ,right After normalization, we get , so that its value is Calculate the change between the current physiological index and the previous physiological index , assuming the previous moment value is , the formula is: , the normal range of the change trend is defined as ,in and Determined according to actual conditions, the formula for calculating the degree of deviation is: , obtained by normalization , for other feature dimensions , get the normalized deviation degree , the final abnormal judgment indicator for: , Exceeding the set threshold , then the physiological indicators are judged to be abnormal, Exceeding the set threshold , then the physiological indicators are judged to be normal.

[0009] Furthermore, the intelligent health analysis module constructs a model through the intelligent health comprehensive evaluation formula to determine whether there are potential health problems. Suppose the input feature vector is , the health status center is , the potential problem state center is, calculate the distance from the data point to the healthy state center , the calculation formula is: , calculate the distance from the data point to the center of the potential problem state , the calculation formula is: , calculate the angle between the data point and the line connecting the two centers ,vector ,vector , the formula is: , where · represents the vector dot product, and are the vector moduli, defining the health assessment index , the calculation formula is: ,in, is the parameter for adjusting the angle weight, is the threshold, determined by training data, The patient is judged to be healthy and has no potential health problems. , and was judged to have potential health problems.

[0010] Furthermore, the intelligent health analysis module captures the changing trend of physiological indicators over time and predicts possible risks in the future through a dynamic fusion prediction model. Suppose the time series of physiological indicators is , calculate the weighted moving average , the formula is: in, is the number of historical data points considered It is The weight of historical data points, and satisfy The weight is set according to the time, and the change of the weighted moving average between the current moment and the previous moment is calculated. , the formula is: , define the trend indicator , the calculation formula is: ,in It is a small positive number, which prevents the denominator from being zero and predicts the physiological index value at the next moment. , the formula is: ,in is the risk assessment with an adjustment factor. The risk assessment is performed by comparing the predicted value with the normal range ( , ) to determine the predicted value Out of normal range ( , ), and considered to be at higher risk.

[0011] Furthermore, the intelligent health analysis module evaluates the health risk level through the risk assessment model according to the model detection results, organizes and preprocesses the health status results, related physiological indicator data, living habit data and historical health data obtained by the model detection, and determines the importance weight of the physiological indicators through expert evaluation and data analysis methods. ,in It is the number of physiological indicators. For historical health data, the weight is determined based on the frequency and severity of the disease. , for the detection results of the model, the weight is determined according to its accuracy and reliability , based on the determined risk factor weights, calculate the risk assessment index , the risk assessment calculation formula is: ,in, It is The value of physiological indicators, It is a comprehensive indicator of historical health data, obtained by encoding the historical occurrence of diseases. is the detection result of the model, with values ​​of 0 and 1, indicating the absence or presence of potential health problems. The risk level is divided into three levels: low, medium, and high according to the calculation results. The corresponding risk assessment index ranges are , , ,in, and The threshold is determined through data analysis. The sorted data is substituted into the risk assessment formula to calculate the risk assessment index. The value of is determined according to the risk level classification standard to determine the corresponding health risk level.

[0012] Furthermore, the early warning and emergency intervention module receives the early warning information from the intelligent health analysis module, and will immediately generate early warning information containing clear risk warnings, preliminary response measures and emergency contact information according to the preset template, and convey the early warning information to the disabled person, his family members and designated medical staff through text messages, APP push and voice prompts. At the same time, the module will automatically select and trigger corresponding intervention measures from the preset intervention measures library according to the risk level, track the implementation of the intervention measures, collect follow-up data on the health status of the disabled person in real time, evaluate the intervention effect, and automatically adjust the intervention strategy through the adaptive health intervention model if the effect is not good.

[0013] Furthermore, the early warning and emergency intervention module automatically adjusts the intervention strategy through the adaptive health intervention model, assuming that the current health status is The set of intervention measures is , define health status change indicators , the calculation formula is: , represents the sum of the differences between the current health status and the health status at the previous moment. For each intervention measure , calculate its effect index at the current moment , determined by comparing the changes in health status before and after the intervention measures, and the calculation formula is: ,in It is an intervention The weight of reflects its importance to the health status, and the adaptive health intervention adjustment value is calculated to determine the adjustment strategy. The formula is: ,in, Is an indicator of health status The target value of , sgn is the symbol function, according to the comprehensive evaluation index , adjust the interventions, and for each intervention , then increase the intensity of intervention measures, , then reduce the intensity of intervention measures, and adjust the amplitude according to The calculation formula is: ,in, is the adjustment factor.

[0014] Furthermore, the personalized health guidance module conducts a comprehensive and personalized assessment of the health status of each disabled person based on the results of intelligent analysis, and uses a multi-dimensional health customization model to tailor a health management plan for the disabled that covers diet adjustment, exercise plan, psychological guidance, examination reminders, and disease prevention knowledge popularization. The plan and health information are regularly pushed through the APP, and interactive functions are provided to encourage the disabled to record data, share their lives, and provide feedback on the implementation status. These feedbacks are processed in real time to dynamically adjust and optimize the plan. In addition, health data is regularly reviewed and analyzed, the implementation effect of the plan is evaluated, and the plan is adjusted and optimized in combination with user feedback and behavior pattern analysis.

[0015] Furthermore, the personalized health guidance module provides a tailor-made health management plan for the patient through a multi-dimensional health customization model. , the initial score calculation formula is: ,in, , They are the weights of each indicator in the health status, personal preference, living habits, and environmental factors on the diet adjustment dimension. Similarly, the exercise plan dimension can be calculated. , psychological guidance dimension , Check reminder dimensions , the dimension of disease prevention knowledge popularization Initial score of , normalize the initial score of each dimension so that the score is between [0, 1], and the calculation formula is: , similarly, the normalized exercise plan score can be obtained , psychological guidance score , Check reminder score , Disease prevention knowledge popularization score , based on the normalized scores, health management plans are customized for people with disabilities.

[0016] Compared with the existing technology, this disabled health care auxiliary management system based on the Internet of Things has the following beneficial effects: 1. The present invention introduces an intelligent health analysis module and combines it with advanced algorithm formulas to conduct in-depth mining and analysis of real-time physiological indicator data and living environment data from the health IoT acquisition module. This real-time health risk assessment mechanism can accurately capture the slight fluctuations in the patient's physiological indicators, and combine the patient's historical health data and living habits to predict possible health problems in advance, and issue warnings to patients, family members and medical staff at the first time through text messages, APP push, and voice prompts. This accurate and timely health risk assessment provides strong support for early detection and intervention of diseases, and effectively reduces the possibility of worsening health risks.

[0017] 2. The present invention can customize a personalized health management plan for each disabled person through the evaluation results of the intelligent health analysis module, and push it regularly through the APP channel. These personalized plans fully take into account the individual differences of patients, including age, gender, health status and living habits, and aim to help patients gradually improve bad habits and improve their health level. At the same time, through continuous data tracking and analysis, the health management plan can be dynamically adjusted to ensure that it always meets the actual needs of the patient. This personalized and continuous health management strategy not only improves the patient's health awareness, but also significantly enhances the health management effect.

[0018] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 This is a flowchart of a health care auxiliary management system for the disabled based on the Internet of Things; Figure 2 This is the workflow diagram of the smart health analysis module. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Embodiment 1: This embodiment provides a health care auxiliary management system for the disabled based on the Internet of Things. The system integrates five modules: health Internet of Things collection, data integration and processing, intelligent health analysis, early warning and emergency intervention, and personalized health guidance. The health data of the disabled are monitored through wearable devices and smart homes. The system can track and comprehensively analyze their health status in real time, intelligently identify abnormal fluctuations, predict health risks, and automatically send early warning information to relevant personnel. At the same time, the system can customize personalized health management plans, push health guidance through APP, and improve the quality of life of the disabled. In addition, the system also has adaptive intervention adjustment capabilities, automatically optimizes intervention measures according to changes in health status, and ensures the effectiveness of health management.

[0023] In the specific implementation, the health IoT acquisition module connects wearable health monitoring devices, smart home health monitoring devices and environmental sensors to each other through wireless network technology and completes the initial configuration with the cloud server to ensure that it can identify and access the designated cloud service platform. After the configuration is completed, each device and sensor automatically collects the physiological indicator data of the disabled, such as heart rate, blood pressure, sleep quality, and living environment data, such as indoor temperature and humidity, air quality, and light intensity, according to the preset collection frequency and rules, while ensuring the accuracy, integrity and real-time of the data. The collected data is encrypted and encapsulated using a protocol to ensure security. It is uploaded to the cloud server in real time through the established wireless network connection. During the transmission process, it supports breakpoint resumption and automatic reconnection functions to ensure stability and reliability. After receiving the uploaded data, the cloud server performs integrity verification and decryption, and stores the data in the corresponding database after confirming the validity of the data for subsequent data processing and analysis.

[0024] The data integration and processing module receives the original data of the physiological indicators and living environment of the disabled from the health Internet of Things acquisition module in real time through preset interfaces and protocols to ensure integrity and timeliness, and cleans the received original data, including removing duplicate data, correcting erroneous data, filling missing values, and identifying and processing outliers, to improve data quality and ensure accuracy and reliability. The cleaned data is integrated according to preset rules and formats, the data format is unified, the timestamp is calibrated, and data from different sources are associated to form a structured data set to ensure data consistency and relevance. The integrated data is stored in the disabled health database. The database design takes into account scalability, security and query efficiency, supports large-scale data storage and rapid retrieval, and establishes a data backup and recovery mechanism to ensure security and recoverability. Indexes are established for the data in the database to improve retrieval efficiency, and a flexible query interface is provided to support users in data retrieval and analysis according to their needs.

[0025] The intelligent health analysis module receives the data set from the data integration and processing module, which contains the historical health data, current physiological index data, living habits data and environmental factor data of the disabled, cleans the data, removes noise and outliers, performs feature selection and normalization processing, and builds a model through a multi-dimensional physiological index calculation formula to determine whether the physiological index is abnormal according to the analysis requirements. The model is built through an intelligent health comprehensive evaluation formula to determine whether there are potential health problems. The dynamic fusion prediction model captures the changing trend of physiological indicators over time and predicts possible risks in the future. The preprocessed data set is divided into a training set, a validation set and a test set. The model is trained using the training set data. During the training process, the validation set data is regularly used to evaluate the model performance. The model structure and parameters are adjusted according to the evaluation results until the model performance reaches the preset standard. The currently collected data is input into the trained model, and abnormal fluctuations in physiological indicators are intelligently identified, possible health problems are predicted, and the corresponding risk probability is calculated. The health risk level is further evaluated based on the risk probability predicted by the model and combined with the risk assessment model. According to the health risk level assessment results, decision support information is provided to medical staff and the disabled person themselves.

[0026] The early warning and emergency intervention module receives information from the intelligent health analysis module, and immediately generates early warning information containing clear risk warnings, preliminary response measures and emergency contact information based on the preset template and the specific type, level and urgency of the risk. The generated early warning information is simultaneously conveyed to the disabled person, his or her family and designated medical staff through multiple channels, including but not limited to SMS notifications, APP push messages and voice prompt calls. According to the risk level determined in the early warning information, the early warning and emergency intervention module automatically retrieves and triggers corresponding remote medical consultation appointments, emergency rescue service requests, and automatically adjusts home environment settings to reduce health risks from the preset intervention measures library. The module continuously tracks the implementation of the triggered intervention measures, collects follow-up data on the health status of the disabled in real time, and uses the adaptive health intervention model to evaluate the effectiveness of the intervention measures based on the collected follow-up health data.

[0027] The adaptive health intervention model is used to evaluate the effect of intervention measures. Assuming the current health status is ,in is the number of health status indicators, and the set of intervention measures is ,in is the number of interventions, defining the health status change indicator , the calculation formula is: , represents the sum of the differences between the current health status and the health status at the previous moment. For each intervention measure , calculate its effect index at the current moment , determined by comparing the changes in health status before and after the intervention measures, and the calculation formula is: ,in It is an intervention The weight of reflects its importance to the health status, and the adaptive health intervention adjustment value is calculated to determine the adjustment strategy. The formula is: ,in, Is an indicator of health status The target value of sgn is a sign function. If the current health status index value is greater than the target value, the sign is negative, indicating that the intervention measures corresponding to the index need to be reduced. If the current health status index value is less than the target value, the sign is positive, indicating that the intervention measures corresponding to the index need to be increased. According to the comprehensive evaluation index , adjust the interventions, and for each intervention , then increase the intensity of intervention measures, , then reduce the intensity of intervention measures, and adjust the amplitude according to The calculation formula is: ,in, is the adjustment coefficient, which is determined through experiments. If the evaluation results show that the expected goals have not been achieved, the module will automatically adjust the intervention strategy and provide detailed feedback on the intervention effect to the disabled and their families, including the measures taken, the effect evaluation results and follow-up suggestions.

[0028] The personalized health guidance module receives the comprehensive evaluation results from the intelligent health analysis module and uses the preset multi-dimensional health evaluation framework to conduct a comprehensive and in-depth personalized evaluation of the health status of the disabled. Based on the evaluation results, the module uses a multi-dimensional health customization model to tailor a personalized health management plan for each disabled person. , the initial score calculation formula is: ,in, , They are the weights of each indicator in the health status, personal preference, living habits, and environmental factors on the diet adjustment dimension. Similarly, the exercise plan dimension can be calculated. , psychological guidance dimension , Check reminder dimensions , the dimension of disease prevention knowledge popularization Initial score of Normalize the initial score of each dimension so that the score is between [0, 1], and the calculation formula is: , similarly, the normalized exercise plan score can be obtained , psychological guidance score , Check reminder score , Disease prevention knowledge popularization score According to the normalized scores, a health management plan is customized for the disabled. The content of the plan covers but is not limited to diet adjustment, exercise plan, psychological guidance, examination reminder, and popularization of disease prevention knowledge. Customized health management plans and related health information are regularly pushed to the disabled through a dedicated APP channel. A user feedback mechanism is set up to process feedback in real time to dynamically adjust and optimize the plan. Health data is regularly reviewed and analyzed to evaluate the implementation effect of the plan. The plan is adjusted and optimized based on user feedback and behavior pattern analysis, and the results are displayed using visualization technology.

[0029] Embodiment 2: In this embodiment, a health care auxiliary management system for the disabled based on the Internet of Things is particularly emphasized that the intelligent health analysis module receives and pre-processes the health data of the disabled, identifies abnormal physiological indicators and potential health problems through algorithm model analysis, and evaluates the health risk level.

[0030] Receive data sets from the data integration and processing module, including historical health data, current physiological indicator data, lifestyle data, and environmental factor data, and perform data cleaning, feature selection, and normalization to ensure data quality and the standardization of model input.

[0031] Select the multi-dimensional physiological index calculation formula to build a model to determine whether the physiological index is abnormal. Suppose the physiological index is ,have feature dimensions, respectively , each feature dimension has a corresponding weight , and satisfies , calculate the deviation between the current physiological index value and the mean value of the normal range , assuming the normal range is , the current value is , the calculation formula is: ,right After normalization, we get , so that its value is The calculation formula is: , calculate the change between the current physiological index and the previous moment's physiological index , assuming the previous moment value is , the formula is: , the normal range of the change trend is defined as ,in and Determined according to actual conditions, the formula for calculating the degree of deviation is: , obtained by normalization , for other feature dimensions , get the normalized deviation degree , the final abnormal judgment indicator for: , Exceeding the set threshold , then the physiological indicators are judged to be abnormal, Exceeding the set threshold , then the physiological indicators are judged to be normal.

[0032] The model is constructed by the intelligent health comprehensive evaluation formula to determine whether there are potential health problems. Suppose the input feature vector is ,in is the number of features, and the health status center is , the potential problem status center is , calculate the distance from the data point to the center of the health state , the calculation formula is: , calculate the distance from the data point to the center of the potential problem state , the calculation formula is: , calculate the angle between the data point and the line connecting the two centers ,vector ,vector , the formula is: , where · represents the vector dot product, and are the vector moduli, defining the health assessment index , the calculation formula is: ,in, is the parameter for adjusting the angle weight, is the threshold, determined by training data, The patient is judged to be healthy and has no potential health problems. , and was judged to have potential health problems.

[0033] The dynamic fusion prediction model is used to capture the changing trend of physiological indicators over time and predict possible risks in the future. Assume that the time series of physiological indicators is ,in Indicates Physiological index values ​​at the moment, calculate the weighted moving average , the formula is: in, is the number of historical data points considered It is The weight of historical data points, and satisfy The weight is set according to the time, and the change of the weighted moving average between the current moment and the previous moment is calculated. , the formula is: , define the trend indicator , obtained by normalizing the variation, the calculation formula is: ,in It is a small positive number, which prevents the denominator from being zero and predicts the physiological index value at the next moment. , the formula is: ,in is the adjustment factor, which is determined experimentally based on actual data. Risk assessment is performed by comparing the predicted value with the normal range ( , ) to determine the predicted value Out of normal range ( , ), and considered to be at higher risk.

[0034] The preprocessed data set is divided into training set, validation set and test set, the model parameters are initialized, the model is trained using the training set data, the model performance is evaluated using the validation set data, the model parameters are adjusted according to the validation results, the current collected data is input into the trained model, and it is determined whether the physiological indicators are abnormal, whether there are potential health problems, and whether possible risks in the future are predicted.

[0035] According to the model detection results, the health risk level is evaluated through the risk assessment model, and the health status results, relevant physiological indicator data, living habit data and historical health data obtained by the model detection are sorted and preprocessed, and the importance weight of the physiological indicators is determined through expert evaluation and data analysis methods. ,in It is the number of physiological indicators. For historical health data, the weight is determined based on the frequency and severity of the disease. , for the detection results of the model, the weight is determined according to its accuracy and reliability , based on the determined risk factor weights, calculate the risk assessment index , the risk assessment calculation formula is: ,in, It is The value of physiological indicators, It is a comprehensive indicator of historical health data, obtained by encoding the historical occurrence of diseases. is the detection result of the model, with values ​​of 0 and 1, indicating the absence or presence of potential health problems. The risk level is divided into three levels: low, medium, and high according to the calculation results. The corresponding risk assessment index ranges are , , ,in, and The threshold is determined through data analysis. The sorted data is substituted into the risk assessment formula to calculate the risk assessment index. The value of is used to determine the corresponding health risk level according to the risk level classification standard, and decision support information is provided to medical staff and disabled people themselves based on the risk level.

[0036] Through the above embodiments, the intelligent health analysis module can effectively process and analyze the health data of disabled people, and provide timely and accurate health risk assessment and decision support information for medical staff and disabled people themselves.

[0037] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A health care auxiliary management system for the disabled based on the Internet of Things, characterized in that: The system consists of the following components: Health IoT collection module: connects wearable health monitoring devices, smart home health monitoring devices and environmental sensors to the cloud server through wireless networks, and uploads the physiological index data and living environment data of disabled people in real time; Data integration and processing module: Receives data from the health IoT collection module, cleans, integrates, and stores it, builds a health database for people with disabilities, and implements long-term tracking and comprehensive analysis of data; Intelligent health analysis module: Uses algorithm formulas to deeply mine and analyze the data provided by the data integration and processing module. By analyzing the historical health data, current physiological indicators, living habits and environmental factors of the disabled, it can intelligently identify abnormal fluctuations, predict health problems, and calculate risk levels. Early warning and emergency intervention module: Receives risk information from the intelligent health analysis module and immediately sends early warning information to the disabled person, his / her family members, and medical staff through SMS, APP push, and voice prompts. At the same time, it automatically triggers corresponding intervention measures according to the risk level; Personalized health guidance module: Based on the results of intelligent analysis, a personalized health management plan is customized for each disabled person, and it is regularly pushed through the APP channel to help disabled people improve their living habits and improve their health level.

2. According to claim 1, a disabled person health care auxiliary management system based on the Internet of Things is characterized in that: The intelligent health analysis module receives the data set of historical health data, current physiological index data, living habit data and environmental factor data of the disabled from the data integration processing module, performs data cleaning, feature selection and normalization operations, builds a model through an algorithm formula, and divides the preprocessed data set into a training set, a validation set and a test set, initializes the model parameters, trains the model using the training set data, and evaluates the performance of the model using the validation set data. According to the verification results, the model parameters are adjusted, and the currently collected data is input into the trained model to determine whether the physiological indicators are abnormal and whether there are potential health problems, and predict possible risks in the future. According to the model detection results, the health risk level is evaluated through the risk assessment model, and decision support information is provided to medical staff and the disabled person himself based on the risk level.

3. According to claim 2, a disabled person health care auxiliary management system based on the Internet of Things is characterized in that: The intelligent health analysis module constructs a model through a multi-dimensional physiological index calculation formula to determine whether the physiological index is abnormal. ,have feature dimensions, respectively , each feature dimension has a corresponding weight ,satisfy , calculate the deviation between the current physiological index value and the mean value of the normal range , assuming the normal range is , the current value is , the calculation formula is: ,right After normalization, we get , so that its value is Calculate the change between the current physiological index and the previous physiological index , assuming the previous moment value is , the formula is: , the normal range of the change trend is defined as ,in and Determined according to actual conditions, the formula for calculating the degree of deviation is: , obtained by normalization , for other feature dimensions , get the normalized deviation degree , the final abnormal judgment indicator for: , Exceeding the set threshold , then the physiological indicators are judged to be abnormal, Exceeding the set threshold , then the physiological indicators are judged to be normal.

4. According to claim 2, the disabled health care auxiliary management system based on the Internet of Things is characterized in that: The intelligent health analysis module builds a model through the intelligent health comprehensive evaluation formula to determine whether there are potential health problems. Suppose the input feature vector is , the health status center is , the potential problem state center is, calculate the distance from the data point to the healthy state center , the calculation formula is: , calculate the distance from the data point to the center of the potential problem state , the calculation formula is: , calculate the angle between the data point and the line connecting the two centers ,vector ,vector , the formula is: , where · represents the vector dot product, and are the vector moduli, defining the health assessment index , the calculation formula is: ,in, is the parameter for adjusting the angle weight, is the threshold, determined by training data, The patient is judged to be healthy and has no potential health problems. , and was judged to have potential health problems.

5. According to claim 2, the disabled health care auxiliary management system based on the Internet of Things is characterized in that: The intelligent health analysis module captures the changing trend of physiological indicators over time and predicts possible risks in the future through a dynamic fusion prediction model. Suppose the time series of physiological indicators is , calculate the weighted moving average , the formula is: in, is the number of historical data points considered It is The weight of historical data points, and satisfy The weight is set according to the time, and the change of the weighted moving average between the current moment and the previous moment is calculated. , the formula is: , define the trend indicator , the calculation formula is: ,in It is a small positive number, which prevents the denominator from being zero and predicts the physiological index value at the next moment. , the formula is: ,in is the risk assessment with an adjustment factor. The risk assessment is performed by comparing the predicted value with the normal range ( , ) to determine the predicted value Out of normal range ( , ), and considered to be at higher risk.

6. The disabled health care auxiliary management system based on the Internet of Things according to claim 2 is characterized in that: The intelligent health analysis module evaluates the health risk level through the risk assessment model according to the model detection results, organizes and preprocesses the health status results, related physiological indicator data, living habit data and historical health data obtained by the model detection, and determines the importance weight of the physiological indicators through expert evaluation and data analysis methods. ,in It is the number of physiological indicators. For historical health data, the weight is determined based on the frequency and severity of the disease. , for the detection results of the model, the weight is determined according to its accuracy and reliability , based on the determined risk factor weights, calculate the risk assessment index , the risk assessment calculation formula is: ,in, It is The value of physiological indexes, It is a comprehensive indicator of historical health data, obtained by encoding the historical occurrence of diseases. is the detection result of the model, with values ​​of 0 and 1, indicating the absence or presence of potential health problems. The risk level is divided into three levels: low, medium, and high according to the calculation results. The corresponding risk assessment index ranges are , , ,in, and The threshold is determined through data analysis. The sorted data is substituted into the risk assessment formula to calculate the risk assessment index. The value of is determined according to the risk level classification standard to determine the corresponding health risk level.

7. According to the Internet of Things-based health care auxiliary management system for the disabled according to claim 1, it is characterized in that: The early warning and emergency intervention module receives the early warning information from the intelligent health analysis module, and will immediately generate early warning information containing clear risk warnings, preliminary response measures and emergency contact information according to the preset template, and convey the early warning information to the disabled person, his family members and designated medical staff through text messages, APP push and voice prompts. At the same time, the module will automatically select and trigger corresponding intervention measures from the preset intervention measures library according to the risk level, track the implementation of the intervention measures, collect follow-up data on the health status of the disabled in real time, evaluate the intervention effect, and automatically adjust the intervention strategy through the adaptive health intervention model if the effect is not good.

8. The Internet of Things-based health care auxiliary management system for the disabled according to claim 7 is characterized in that: The early warning and emergency intervention module automatically adjusts the intervention strategy through the adaptive health intervention model. , the set of intervention measures is , define health status change indicators , the calculation formula is: , represents the sum of the differences between the current health status and the health status at the previous moment. For each intervention measure , calculate its effect index at the current moment , determined by comparing the changes in health status before and after the intervention measures, and the calculation formula is: ,in It is an intervention measure The weight of reflects its importance to the health status, and the adaptive health intervention adjustment value is calculated to determine the adjustment strategy. The formula is: ,in, Is an indicator of health status The target value of , sgn is the symbol function, according to the comprehensive evaluation index , adjust the interventions, and for each intervention , then increase the intensity of intervention measures, , then reduce the intensity of intervention measures, and adjust the amplitude according to The calculation formula is: ,in, is the adjustment factor.

9. The Internet of Things-based health care auxiliary management system for the disabled according to claim 1 is characterized in that: Based on the results of intelligent analysis, the personalized health guidance module conducts a comprehensive and personalized assessment of the health status of each disabled person, and uses a multi-dimensional health customization model to tailor a health management plan for the disabled that covers diet adjustment, exercise plan, psychological guidance, examination reminders, and disease prevention knowledge popularization. The plan and health information are regularly pushed through the APP, and interactive functions are provided to encourage the disabled to record data, share their lives, and provide feedback on the implementation status. These feedbacks are processed in real time to dynamically adjust and optimize the plan. In addition, health data is regularly reviewed and analyzed, the implementation effect of the plan is evaluated, and the plan is adjusted and optimized in combination with user feedback and behavior pattern analysis.

10. The disabled health care auxiliary management system based on the Internet of Things according to claim 9 is characterized in that: The personalized health guidance module uses a multi-dimensional health customization model to tailor a health management plan for the patient. , the initial score calculation formula is: ,in, , They are the weights of each indicator in the health status, personal preference, living habits, and environmental factors on the diet adjustment dimension. Similarly, the exercise plan dimension can be calculated. , psychological guidance dimension , check reminder dimensions , the dimension of disease prevention knowledge popularization Initial score of , normalize the initial score of each dimension so that the score is between [0, 1], and the calculation formula is: , similarly, the normalized exercise plan score can be obtained , psychological guidance score , Check reminder score , Disease prevention knowledge popularization score , based on the normalized scores, health management plans are customized for people with disabilities.

Images