Health monitoring system based on wearable device

Abstract

The invention relates to the technical field of health monitoring, in particular to a health monitoring system based on wearable equipment, which comprises a data acquisition module, a health data analysis module, an emotion and sound analysis module and a health prediction feedback module. According to the health monitoring system, the prediction accuracy of the health monitoring system is effectively improved by combining time sequence analysis with a processing strategy of environmental factors, and more accurate health condition early warning, disease prevention and management strategy optimization and emotion and sound analysis integration are provided by comprehensively analyzing long-term health data and real-time environmental changes of the user, so that the user experience is improved. The emotion change is clarified through acoustic feature analysis, the understanding of the emotion state of the user is improved, the health monitoring dimension is enhanced, the participation degree and satisfaction degree of the user are improved through personalized health suggestion and real-time feedback, the practicability and effectiveness of the system are enhanced, and health management is more accurate and has interactivity.

Description

Technical Field

[0001] The present invention relates to the technical field of health monitoring, and in particular to a health monitoring system based on a wearable device. Background Art

[0002] The field of health monitoring technology mainly involves the use of various devices and systems to track and record individual physiological and biomarkers in real time, including the use of wearable devices, mobile applications and remote monitoring systems to collect health data such as heart rate, blood pressure, blood sugar levels, sleep quality and many other vital signs, aiming to improve the convenience and efficiency of health management, so that individuals and medical service providers can better understand health conditions, prevent diseases, and adjust treatment plans in a timely manner. In addition, health monitoring technology also supports long-term disease management and the health of the elderly, improving the accessibility and personalization of medical services.

[0003] Among them, the health monitoring system based on wearable devices can continuously monitor the health status and physical signs of the wearer by integrating sensors, software applications and communication technologies. Its main uses include tracking key health indicators, identifying potential health problems, promoting healthy lifestyles, and supporting clinical decision-making and patient self-management, which helps individuals better understand their health status and provides real-time data for medical professionals to monitor, evaluate and intervene more accurately, especially showing its important value in chronic disease management and preventive medicine.

[0004] Existing technologies often fail to consider the impact of environmental factors and emotional states when monitoring physiological and biomarkers, resulting in an incomplete assessment of health status. For example, failure to correlate climate fluctuations with health data may overlook climate-sensitive health issues. In addition, existing methods are not deep enough in data analysis and often stay at the surface data, which limits the accurate prediction of health trends, leading to delayed preventive measures and inaccurate treatment plans. The lack of attention to emotional monitoring also reduces comprehensive support for mental health, affecting the effective management of chronic diseases and disease recovery. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a health monitoring system based on a wearable device.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A health monitoring system based on a wearable device comprises: The data collection module records the user's physiological parameters through wearable devices, collects temperature, humidity and air quality environmental data, integrates the data, and generates original health and environmental data; The health data analysis module extracts key health indicators based on the original health and environmental data, performs time series analysis on the indicators, identifies key health changes, synchronously processes environmental data based on the identification results, compares the data with the user's historical health records, identifies potential health problems, and generates comprehensive health analysis results; The emotion and sound analysis module obtains user voice input information through the wearable device, extracts acoustic features, analyzes the features to determine changes in the user's emotional state, integrates the analysis results with the comprehensive health analysis results, evaluates the correlation between emotional fluctuations and physical health status, and generates emotional state and health assessment records; The health prediction feedback module is based on the emotional state and health assessment records, combined with the user's long-term health data, and uses trend analysis to predict the user's future health changes. According to the trend analysis results, it designs and sends personalized health advice to the user through wearable devices, performs continuous optimization based on user feedback, and generates health prediction and management plans.

[0007] As a further solution of the present invention, the steps for obtaining the original health and environment data are: The built-in sensors in wearable devices simultaneously record the user's physiological indicators and environmental parameters, including heart rate, blood pressure, body temperature, air temperature, humidity and air quality data, to generate physiological and environmental monitoring data; Based on the physiological and environmental monitoring data, data synchronization processing is performed to detect and eliminate outliers to obtain synchronized physiological and environmental data; The synchronized physiological and environmental data are integrated using the formula, ; Computing quantitative values ​​of the interaction between environmental and physiological data , get the original health and environment data, where Representative The environmental data measured Representative The physiological data measured Indicates the absolute value of the difference between environmental and physiological data.

[0008] As a further solution of the present invention, the steps of identifying key health changes are: Based on the original health and environmental data, perform data cleaning, fill in missing values, and generate a cleaned data set; Using the cleaned data set, screening and marking health indicators are performed, and a set of key health indicators is output; The time series analysis of the key health indicator set is performed using the formula, ; Calculate in time Health indicator status value , generate key health change records, where, represents a time point in a time series, Indicates Key health indicators in time The absolute value of Indicates The normalization factor of the indicator, is the total number of indicators.

[0009] As a further solution of the present invention, the steps for obtaining the comprehensive health analysis results are: Based on the key health change records, match the timestamps of health data and environmental data, verify the consistency of the recording time of environmental data and health indicators, and generate a complete environmental and health data comparison set; Comparing the complete set of environmental and health data with the user's historical health records, applying statistical analysis to identify statistically significant health changes, revealing the correlation between environmental factors and changes in health status, and generating a record of potential health problems; Based on the potential health problem records, combined with the user's health history and current environmental data, potential health risks are analyzed to generate comprehensive health analysis results.

[0010] As a further solution of the present invention, the step of determining the change in the user's emotional state is: The user inputs voice through the wearable device, and the built-in microphone captures the sound wave information, converts the digital signal, and generates the original sound digital signal; Based on the original sound digital signal, perform spectrum analysis, extract acoustic features, including fundamental frequency and timbre, and generate an acoustic feature data set; The acoustic feature data set is used to perform sentiment analysis, identify and analyze changes in the user's emotional state, and obtain the user's emotional state change results.

[0011] As a further solution of the present invention, the steps of obtaining the emotional state and health assessment records are: Integrate the user's emotional state change results with the comprehensive health analysis results, match the two data sets according to the user identifier and recording time, and generate a fused data set; Based on the fused data set, correlation analysis is performed using the formula, ; Calculate the change in health indicators , and obtain the correlation analysis results, where represents the change in the emotion score, It represents the baseline value of health indicators when there is no change in mood; Based on the correlation analysis results, the relationship between emotional changes and health indicators is recorded, data visualization is performed, and emotional state and health assessment records are constructed.

[0012] As a further solution of the present invention, the steps of predicting the future health changes of the user are: The emotional state and health assessment records are integrated with long-term health information to perform unified data processing to obtain a comprehensive health data framework; Based on the comprehensive health data framework, data normalization processing and time series construction are performed to obtain a preprocessed health time series; Based on the preprocessed health time series, the formula is used: ; Calculate the predicted health status value of the health indicator , generate the prediction result of the user's future health status, where Indicates the previous health status value, is the current health data value, is the health data value at the previous time point.

[0013] As a further solution of the present invention, the steps for obtaining the health prediction and management plan are: Based on the predicted results of the user's future health status, the user's historical health data and real-time monitoring data are integrated to analyze the changing trend of personal health status, and a draft of personalized health recommendations is generated in combination with the user's age, gender, and lifestyle factors; According to the draft of personalized health advice, personalized advice is displayed through a wearable device using a graphical user interface, and the user's health data is collected in real time, and the content and frequency of the advice are adjusted according to the user's activity response, and dynamic optimization is performed to obtain adjusted health advice; Based on the adjusted health recommendations, iterative optimization is performed to optimize user satisfaction, engagement, and health improvement adjustments, and to generate health prediction and management plans.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by combining time series analysis with the processing strategy of environmental factors, the prediction accuracy of the health monitoring system is effectively improved. By comprehensively analyzing the user's long-term health data and real-time environmental changes, a more accurate health status warning is provided, and disease prevention and management strategies are optimized. The integration of emotion and sound analysis clarifies emotional changes through acoustic feature analysis, improves the understanding of the user's emotional state, enhances the dimension of health monitoring, and personalized health suggestions and real-time feedback improve user participation and satisfaction, enhance the practicality and effectiveness of the system, and make health management more accurate and interactive. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a system flow chart of the present invention; Figure 2 The flowchart of obtaining the original health and environmental data of the present invention is as follows; Figure 3 A flow chart for identifying key health changes of the present invention; Figure 4 A flowchart for obtaining comprehensive health analysis results of the present invention; Figure 5 A flow chart for determining a change in a user's emotional state according to the present invention; Figure 6 A flowchart for obtaining emotional state and health assessment records of the present invention; Figure 7 A flowchart for predicting future health changes of users of the present invention; Figure 8 The figure is a flow chart for obtaining the health prediction and management scheme of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0018] See also Figure 1, a health monitoring system based on wearable devices includes: The data collection module records the user's physiological parameters through wearable devices, collects temperature, humidity and air quality environmental data, integrates the data, and generates original health and environmental data; The health data analysis module extracts key health indicators based on the original health and environmental data, performs time series analysis on the indicators, identifies key health changes, synchronously processes environmental data based on the identification results, compares the data with the user's historical health records, identifies potential health problems, and generates comprehensive health analysis results; The emotion and sound analysis module obtains user voice input information through wearable devices, extracts acoustic features, analyzes the features to determine changes in the user's emotional state, integrates the analysis results with the comprehensive health analysis results, evaluates the correlation between emotional fluctuations and physical health status, and generates emotional state and health assessment records; The health prediction and feedback module is based on emotional state and health assessment records, combined with the user's long-term health data, and uses trend analysis to predict the user's future health changes. Based on the trend analysis results, personalized health advice is designed and sent to the user through wearable devices. Continuous optimization is performed based on user feedback to generate health prediction and management plans.

[0019] The original health and environmental data include a set of physiological parameters and environmental data indicators. The comprehensive health analysis results include health indicator change analysis results, environmental impact assessment records and potential health problem records. The emotional state and health assessment records include emotion fluctuation analysis results, health and emotion correlation assessment records and sound feature analysis results. The health prediction and management plan includes health trend analysis results, customized health recommendations and feedback adjustment strategies.

[0020] See also Figure 2 , the steps to obtain the original health and environment data are: The built-in sensors in wearable devices simultaneously record the user's physiological indicators and environmental parameters, including heart rate, blood pressure, body temperature, air temperature, humidity and air quality data, to generate physiological and environmental monitoring data; Wearable devices can record the user's biological indicators and surrounding environmental conditions in real time by integrating a variety of physiological and environmental monitoring sensors. The sensors inside the device are specially designed to capture physiological data such as heart rate, blood pressure and body temperature. At the same time, additional environmental sensors are responsible for measuring air temperature, humidity and air quality index. The data is integrated into a data stream through the device's synchronization function. The data stream is continuously monitored to ensure that all indicators are updated in real time. In this way, the device can provide a comprehensive snapshot of health and environmental conditions at any given time, providing users with real-time health status and environmental quality feedback, so that users can make corresponding life or health adjustments based on these comprehensive data.

[0021] Based on physiological and environmental monitoring data, data synchronization processing is performed to detect and eliminate outliers to obtain synchronized physiological and environmental data; The device's built-in data processing module is responsible for performing synchronization, outlier detection and data rejection to ensure the accuracy and completeness of the data obtained from the sensor before analysis. First, the synchronization operation ensures that all physiological and environmental data are recorded and analyzed according to the same timeline, which is crucial for subsequent trend analysis and pattern recognition. Abnormal data is then identified by comparing data points with the set normal range. Data points are rejected to avoid health monitoring results being interfered with by accidental events or sensor errors, providing users with accurate health and environmental status information.

[0022] The synchronous physiological and environmental data are integrated using the formula, ; Computing quantitative values ​​of the interaction between environmental and physiological data , get the original health and environment data, where Representative The environmental data measured Representative The physiological data measured Indicates the absolute value of the difference between environmental and physiological data; At a specific point in time, the environmental data obtained from the device is (indicates temperature 25℃), physiological data are (indicates a heart rate of 80 beats / minute), then the absolute value of the product in the formula is , the absolute value of the difference is According to the formula, the calculation process is as follows: Assuming there are 10 types of such data points, the sum of the accumulated results is , then the final integrated data The calculation result is: The results show that the processing method of integrated environmental and physiological data can reflect the quantitative value of the interaction between the user's health status and environmental conditions. The higher the value, the more obvious the interaction, which can provide a basis for further health monitoring and environmental management.

[0023] See also Figure 3 , the steps to identify key health changes are: Based on the original health and environmental data, perform data cleaning, fill in missing values, and generate a cleaned data set; Based on the original health and environmental data, the first thing to focus on is the handling of missing values. Usually, different filling strategies are chosen according to the type and distribution of the data. For example, for continuous variables, the mean or median of the variable is used for filling, while for categorical variables, the mode (the most frequently occurring value) may be selected as the filling value; next, different identification methods are selected according to the distribution characteristics of the data, such as the Z-score-based method to identify and process data points that deviate from the mean by multiple standard deviations. The processing methods include removing outliers or replacing them with appropriate statistical values ​​to ensure the consistency and reliability of the data set; in addition, data standardization is also an important step in preprocessing. Through data standardization, data of different magnitudes or units are converted to the same scale. Commonly used methods include minimum-maximum standardization and Z-score standardization, which together ensure the comprehensiveness and in-depth nature of data preprocessing and provide a stable and standardized basis for subsequent analysis.

[0024] Use the cleaned data set to screen and mark health indicators and output a set of key health indicators; When screening key health indicators, the random forest algorithm is applied for processing. First, the algorithm's operating parameters are set, including the number of decision trees and the depth of the trees, to adjust the complexity and fit of the model. The random forest trains the data set by constructing multiple trees. Each tree is constructed on a different random subset of the data set, thereby increasing the generalization ability of the model. During the training process, each tree will evaluate the importance of each feature, that is, the ability of each feature to improve the accuracy of the model at the split point in the decision tree. By accumulating the evaluation results of all trees, the overall feature importance score is obtained. Finally, features with high importance scores are selected as key health indicators, which are usually strongly correlated with the target variable, that is, health status. This method can not only accurately identify the main factors affecting health, but also effectively deal with feature selection problems in a big data environment, thereby improving the model's explanatory power and prediction accuracy.

[0025] A time series analysis of a set of key health indicators is performed using the formula, ; Calculate in time Health indicator status value , generate key health change records, where, represents a time point in a time series, Indicates Key health indicators in time The absolute value of Indicates The normalization factor of the indicator, is the total number of indicators; , time point The index value , , the corresponding normalization factor , .

[0026] Calculate the adjusted value of each indicator: Sum: Take the square root: Divide by the number of indicators: Therefore, at the time point Health indicators status It is approximately 3.496, which means that after taking the weight into account, the comprehensive assessment of health status at this point in time is above average. The value can help medical experts quickly understand the patient's health status at a specific moment.

[0027] See also Figure 4 , the steps to obtain the comprehensive health analysis results are: Based on key health change records, match the timestamps of health data and environmental data, verify the consistency of the recording time of environmental data and health indicators, and generate a complete set of environmental and health data comparisons; When synchronizing environmental data with health data, the process involves importing environmental data and health records into data processing software, such as Python or R, first sorting the data using the timestamp field, and then applying data merging techniques to ensure that each health indicator record is aligned with the corresponding environmental data record. This is usually achieved through SQL's JOIN operation or Python's merge function. In addition, for data with incompletely corresponding timestamps, interpolation methods are used to estimate environmental data at missing time points to ensure the integrity and consistency of the data set. Through detailed steps, the accuracy and effectiveness of the data preparation stage are ensured, laying a solid foundation for subsequent in-depth analysis.

[0028] Compare the complete set of environmental and health data with the user's historical health records, apply statistical analysis, identify statistically significant health changes, reveal the correlation between environmental factors and changes in health status, and generate records of potential health problems; Comparative analysis of synchronized environmental and health data with historical health records is accomplished through multi-level data processing. First, data cleaning techniques are used to remove any outliers or inconsistent data, and then advanced statistical methods, such as multivariate regression analysis, are used to explore the relationship between environmental variables and health indicators. In particular, libraries such as Python's Pandas and SciPy are used to process data and perform statistical tests to determine which environmental factors are significantly correlated with changes in health indicators. Further, time series analysis is used to identify long-term trends in health changes, and cluster analysis is used to group similar health events. The analysis can help accurately identify how environmental factors affect health status, thereby generating scientifically valid health risk assessments.

[0029] Based on the records of potential health problems, combined with the user's health history and current environmental data, potential health risks are analyzed to generate comprehensive health analysis results; Apply machine learning techniques to integrate environmental and health data to predict future health risks. The process includes building and training decision trees and neural network models. Specific operations include selecting appropriate feature variables, adjusting model parameters such as learning rate and number of iterations, and using a large number of historical data sets for training to ensure model generalization ability and prediction accuracy. Cross-validation methods are used to avoid overfitting, and model performance is evaluated through indicators such as model accuracy and recall. In addition, after the model training is completed, new or real-time data is used for prediction verification. Through machine learning algorithms, potential health problems can be analyzed and obtained, and the pertinence and effectiveness of preventive measures can be improved.

[0030] See also Figure 5 , the steps to determine the change of user emotional state are: The user inputs voice through the wearable device, and the built-in microphone captures the sound wave information, converts the digital signal, and generates the original sound digital signal; When a user inputs voice through a wearable device, the highly sensitive microphone in the device first captures the sound waves in the environment, which include the user's voice and possible background noise. The microphone converts the sound waves into analog electrical signals, which are then transmitted to the analog-to-digital converter (ADC). The function of the ADC is to convert continuous analog signals into digital signals for computer processing. During this process, the ADC sampling frequency and bit depth settings ensure that the converted digital signal can retain the characteristics of the original sound wave as much as possible, such as the pitch and intensity of the voice. The converted digital signal saves all the necessary information of the voice, including pitch, intensity and rhythm. The data provides the basis for subsequent acoustic analysis and emotion recognition to generate the original digital signal.

[0031] Based on the original sound digital signal, perform spectrum analysis, extract acoustic features, including fundamental frequency and timbre, and generate an acoustic feature data set; The steps of extracting acoustic features from raw digital signals involve complex signal processing techniques. First, the signal is analyzed using Fast Fourier Transform (FFT) to identify the fundamental frequency, harmonics, and other frequency components of the sound. FFT conversion helps reveal the frequency distribution in speech signals, which is essential for subsequent acoustic pattern recognition. Based on the frequency information, acoustic models are further applied to analyze the time domain and frequency domain characteristics of the signal, such as volume changes, speaking speed, and pitch fluctuations. These features are very helpful for understanding the emotional color of speech. After that, the acoustic features are encoded into feature vectors and stored in the dataset. These feature vectors provide training data for the emotion recognition model based on machine learning, generating an acoustic feature dataset to support high-precision emotional state analysis.

[0032] Use acoustic feature datasets to perform sentiment analysis, identify and analyze changes in users’ emotional states, and obtain results of changes in users’ emotional states; After obtaining a detailed acoustic feature dataset, the data is used to identify the user's emotional changes through the sentiment analysis model. The first step in the process is to select an appropriate machine learning algorithm, such as a support vector machine (SVM) or a deep neural network, which can learn and recognize complex emotional patterns from the training data. The acoustic feature dataset is input into the trained model, and the model analyzes the data to identify the emotional state of the voice, such as happiness, sadness, anger, etc. During this analysis process, each acoustic feature is assigned a different weight, and the weight is adjusted during the model training phase based on the contribution of the feature to the emotion classification. The output of the model is a prediction of the user's emotional state, which can be directly used to improve user interaction or conduct emotional health monitoring, and finally determine and obtain the results of the user's emotional state changes to provide real-time feedback to the user.

[0033] See also Figure 6 , the steps to obtain emotional state and health assessment records are: Integrate the results of user emotional state changes with comprehensive health analysis results, match the two data sets according to user identification and recording time, and generate a fused data set; Integrate the results of the user's emotional state changes with the comprehensive health analysis results. In this process, first locate the relevant entries in the two data sets by identifying the user's unique identifier, and then use timestamp alignment technology to ensure that the matching data comes from the same recording period. Use advanced data fusion techniques, such as database join operations and time series analysis, to handle inconsistent data, such as time deviations and missing values, and estimate missing data points through interpolation or neighbor-based best estimation methods. After completing the steps, the generated fused data set will contain comprehensive information from emotional changes to physical health status, providing an accurate and comprehensive data foundation for the next step of correlation analysis, ensuring the reliability and practicality of the analysis results.

[0034] Based on the fused data set, correlation analysis is performed using the formula, ; Calculate the change in health indicators , and obtain the correlation analysis results, where represents the change in the emotion score, It represents the baseline value of health indicators when there is no change in mood; represents the change in sentiment score, Changes in health indicators, such as changes in heart rate or blood pressure. represents the expected baseline value of the health indicator when there is no change in mood. The part of the brain that increases sensitivity to emotional changes, especially when the emotional changes are small (i.e. When the value is low, the square root function increases the speed of response to small changes. For example, if (emotional change), then , assuming ,but . It means that under a given change in mood, the health index is expected to change by 20 units. The results show the direct relationship between changes in mood and changes in health indicators, and can be used to further analyze the specific impact of mood fluctuations on physical health.

[0035] Based on the results of correlation analysis, record the relationship between emotional changes and health indicators, perform data visualization, and construct emotional state and health assessment records; According to the results of the correlation analysis, in the process of constructing the emotional state and health assessment record, how each health indicator changes with emotional fluctuations is recorded in detail. Use data visualization tools, such as bar charts, line charts, and heat maps, to depict the relationship between emotional changes and health indicators. For example, through the line chart, you can clearly observe the specific performance of indicators such as heart rate and blood pressure during emotional fluctuations, and perform trend analysis at the same time to identify which health indicators are most strongly associated with emotional changes. In addition, the generated assessment record is not just a simple display of data, but also includes an interpretation of data trends, and provides suggestions for future health management directions, making the record instructive for both users and health professionals, and providing a scientific basis for the implementation of personalized health interventions.

[0036] See also Figure 7 , the prediction steps for future health changes of users are: Comprehensive emotional state and health assessment records, combined with long-term health information, unified data processing to obtain a comprehensive health data framework; First, the data collection module collects the user's emotional state records, health assessment data and long-term health information, including the user's heart rate, blood pressure, sleep quality, and emotional state reports obtained through questionnaires. Then the data is imported through a unified data interface and standardized using the ETL (extraction, transformation, loading) process. For example, heart rate and blood pressure data are converted to times per minute and millimeters of mercury units to ensure that all data conform to the standard format for health data analysis. In addition, the data needs to be detected for outliers and missing values, and gaps are filled using interpolation or mean replacement methods to ensure the integrity of the data set. Finally, the data is integrated into a comprehensive health data framework through the database management system. This data framework supports instant updates and efficient queries, providing a reliable foundation for subsequent data analysis to obtain a comprehensive health data framework.

[0037] Based on the comprehensive health data framework, data normalization and time series construction are performed to obtain the preprocessed health time series; After creating a comprehensive health data framework, data preprocessing is performed. The key is to eliminate potential biases between different data sources. The technical means to achieve this is normalization, which converts all health monitoring indicators such as heart rate, blood pressure, and weight to the same measurement for effective comparison. At the same time, a time series model is constructed, focusing on the pattern of data changes over time, including statistics on the average and fluctuation range of daily heart rate, and adjustments for seasonal factors, such as physiological differences between winter and summer. In addition, the construction of the time series also involves determining the periodicity and trend of the data, and using the autoregressive moving average (ARMA) model to analyze trends, which provides a clear and unified data basis for trend analysis, and produces preprocessed health time series that can provide accurate input data for further health trend analysis and prediction.

[0038] Based on the preprocessed health time series, the formula is used: ; Calculate the predicted health status value of the health indicator , generate the prediction result of the user's future health status, where Indicates the previous health status value, is the current health data value, is the health data value at the previous time point; and is the health data value of two consecutive time points, such as the average heart rate value of two consecutive days. bpm (heart rate for the day), bpm (heart rate of the previous day), first calculate the absolute value of the difference between the heart rates of the two days, that is, bpm, and then taking the square root of the difference, we get , calculate the normalized denominator, , and finally calculate the fraction , if the previous prediction state Assuming 75 bpm, the final predicted health status will be The results show that based on the heart rate changes over two consecutive days, it can be predicted that the heart rate will increase slightly on the third day, which reflects the daily fluctuation of heart rate and the accuracy of prediction.

[0039] See also Figure 8 , the steps to obtain the health prediction and management plan are: Based on the prediction results of the user's future health status, the user's historical health data and real-time monitoring data are integrated to analyze the changing trend of personal health status, and a draft of personalized health recommendations is generated based on the user's age, gender, and lifestyle factors; Based on the predicted results of the user's future health status, the process of developing personalized health recommendations first involves a comprehensive evaluation of user data, including statistical analysis of historical health data and real-time monitoring data. Statistical analysis uses machine learning technology to identify health trends and predict potential health risks. For example, by analyzing factors such as the user's heart rate data, sleep quality, and activity level, it can identify behavioral patterns that may lead to health problems. Then, the model will make improvement suggestions based on the analysis results. This is achieved by comparing the correlation between user behavior and health status. This not only improves the personalization of the recommendations, but also enhances the practicality and timeliness of the recommendations, thereby making health management more accurate and effective, and generating a preliminary draft of personalized health recommendations.

[0040] Based on the draft of personalized health recommendations, personalized recommendations are displayed through wearable devices using a graphical user interface, and the user's health data is collected in real time. The content and frequency of recommendations are adjusted according to the user's activity response, and dynamic optimization is performed to obtain adjusted health recommendations; In the process of implementing personalized health advice through smart wearable devices, the advice first needs to be encoded into a format that can be read by the device, and then sent to the user's device through wireless communication technology. The application on the device will adjust the display method and frequency of the advice based on the user's real-time health data. For example, if the user's heart rate monitoring shows abnormal activity, the advice system will adjust the advice intensity and immediately remind the user to reduce strenuous activities. At the same time, the system will monitor the user's response to the advice, such as the number of steps walked or the duration of exercise, and adjust subsequent advice content based on feedback. The dynamic adjustment process ensures that the advice always remains relevant and effective, thereby optimizing the user's health management experience and obtaining adjusted health advice.

[0041] Based on the adjusted health recommendations, iterative optimization is performed to optimize user satisfaction, engagement, and health improvement, and generate health prediction and management plans; In the process of collecting user feedback and continuously optimizing health recommendations, the focus is on analyzing users' acceptance and implementation of current recommendations, including evaluating user satisfaction and health improvements through data collected through online questionnaires and health monitoring devices. Based on the data, statistical and machine learning methods are used to re-evaluate and adjust the health recommendation module. For example, model parameters are adjusted to better reflect the user's latest health data and feedback. At the same time, new predictive indicators such as stress levels or sleep quality are introduced to enhance the model's predictive ability. The data-based iterative optimization process not only improves the personalization and accuracy of recommendations, but also enhances the program's ability to adapt to changes in user health, ultimately generating a comprehensive health prediction and management plan.

[0042] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A health monitoring system based on wearable devices, characterized in that: The system comprises: The data collection module records the user's physiological parameters through wearable devices, collects temperature, humidity and air quality environmental data, integrates the data, and generates original health and environmental data; The health data analysis module extracts key health indicators based on the original health and environmental data, performs time series analysis on the indicators, identifies key health changes, synchronously processes environmental data based on the identification results, compares the data with the user's historical health records, identifies potential health problems, and generates comprehensive health analysis results; The emotion and sound analysis module obtains user voice input information through the wearable device, extracts acoustic features, analyzes the features to determine changes in the user's emotional state, integrates the analysis results with the comprehensive health analysis results, evaluates the correlation between emotional fluctuations and physical health status, and generates emotional state and health assessment records; The health prediction feedback module is based on the emotional state and health assessment records, combined with the user's long-term health data, and uses trend analysis to predict the user's future health changes. According to the trend analysis results, it designs and sends personalized health advice to the user through wearable devices, performs continuous optimization based on user feedback, and generates health prediction and management plans.

2. The wearable device-based health monitoring system according to claim 1, characterized in that: The steps for obtaining the original health and environment data are as follows: The built-in sensors in wearable devices simultaneously record the user's physiological indicators and environmental parameters, including heart rate, blood pressure, body temperature, air temperature, humidity and air quality data, to generate physiological and environmental monitoring data; Based on the physiological and environmental monitoring data, data synchronization processing is performed to detect and eliminate outliers to obtain synchronized physiological and environmental data; The synchronized physiological and environmental data are integrated using the formula, Computing quantitative values ​​of the interaction between environmental and physiological data , get the original health and environment data, where Representative The environmental data measured Representative The physiological data measured Indicates the absolute value of the difference between environmental and physiological data.

3. The health monitoring system based on wearable devices according to claim 2, characterized in that: The steps for identifying the key health changes are: Based on the original health and environmental data, perform data cleaning, fill in missing values, and generate a cleaned data set; Using the cleaned data set, screening and marking health indicators are performed, and a set of key health indicators is output; The time series analysis of the key health indicator set is performed using the formula, Calculate in time Health indicator status value , generate key health change records, where, represents a time point in a time series, Indicates Key health indicators in time The absolute value of Indicates The normalization factor of the indicator, is the total number of indicators.

4. The health monitoring system based on wearable devices according to claim 3, characterized in that: The steps for obtaining the comprehensive health analysis results are: Based on the key health change records, match the timestamps of health data and environmental data, verify the consistency of the recording time of environmental data and health indicators, and generate a complete environmental and health data comparison set; Comparing the complete set of environmental and health data with the user's historical health records, applying statistical analysis to identify statistically significant health changes, revealing the correlation between environmental factors and changes in health status, and generating a record of potential health problems; Based on the potential health problem records, combined with the user's health history and current environmental data, potential health risks are analyzed to generate comprehensive health analysis results.

5. The wearable device-based health monitoring system according to claim 1, characterized in that: The steps of determining the change of the user's emotional state are: The user inputs voice through the wearable device, and the built-in microphone captures the sound wave information, converts the digital signal, and generates the original sound digital signal; Based on the original sound digital signal, perform spectrum analysis, extract acoustic features, including fundamental frequency and timbre, and generate an acoustic feature data set; The acoustic feature data set is used to perform sentiment analysis, identify and analyze changes in the user's emotional state, and obtain the user's emotional state change results.

6. The health monitoring system based on wearable devices according to claim 5, characterized in that: The steps for obtaining the emotional state and health assessment records are as follows: Integrate the user's emotional state change results with the comprehensive health analysis results, match the two data sets according to the user identifier and recording time, and generate a fused data set; Based on the fused data set, correlation analysis is performed using the formula, Calculate the change in health indicators , and obtain the correlation analysis results, where represents the change in the emotion score, It represents the baseline value of health indicators when there is no change in mood; Based on the correlation analysis results, the relationship between emotional changes and health indicators is recorded, data visualization is performed, and emotional state and health assessment records are constructed.

7. The wearable device-based health monitoring system according to claim 6, characterized in that: The steps for predicting the user's future health changes are: The emotional state and health assessment records are integrated with long-term health information to perform unified data processing to obtain a comprehensive health data framework; Based on the comprehensive health data framework, data normalization processing and time series construction are performed to obtain a preprocessed health time series; Based on the preprocessed health time series, the formula is used: Calculate the predicted health status value of the health indicator , generate the prediction result of the user's future health status, where Indicates the previous health status value, is the current health data value, is the health data value at the previous time point.

8. The wearable device-based health monitoring system according to claim 7, characterized in that: The steps for obtaining the health prediction and management plan are: Based on the predicted results of the user's future health status, the user's historical health data and real-time monitoring data are integrated to analyze the changing trend of personal health status, and a draft of personalized health recommendations is generated in combination with the user's age, gender, and lifestyle factors; According to the draft of personalized health advice, personalized advice is displayed through a wearable device using a graphical user interface, and the user's health data is collected in real time, and the content and frequency of the advice are adjusted according to the user's activity response, and dynamic optimization is performed to obtain adjusted health advice; Based on the adjusted health recommendations, iterative optimization is performed to optimize user satisfaction, engagement, and health improvement adjustments, and to generate health prediction and management plans.

Images