CN119791621B - Multifunctional vital sign monitoring method and system - Google Patents

Abstract

The embodiment of the application provides a multifunctional vital sign monitoring method and system. The method comprises the steps of acquiring real-time vital sign signals of a user in different physiological states, combining environment parameters to form a multi-dimensional monitoring data set, dynamically adjusting the monitoring frequency and the accuracy of the real-time vital sign signals according to the multi-dimensional monitoring data set by using an intelligent regulation algorithm, generating a data collection strategy, collecting vital sign data based on the data collection strategy, analyzing and processing the monitored abnormal conditions of the vital signs by using an instant evaluation engine based on the vital sign data to generate health condition evaluation results and personalized health suggestions, and safely synchronizing the health condition evaluation results and the personalized suggestions to a medical monitoring system of the user so as to realize remote medical service and continuous health management. The technical scheme provided by the embodiment of the application improves the accuracy and efficiency of health monitoring.

Description

A multifunctional vital sign monitoring method and system

Technical Field

This application relates to the field of intelligent health monitoring and telemedicine technology, and in particular to a multifunctional vital sign monitoring method and system.

Background Technology

With an aging population and rising incidence of chronic diseases, the demand for continuous health monitoring and telemedicine services is growing. In scenarios such as home care, hospital wards, and rehabilitation centers, real-time acquisition of multi-dimensional vital sign data and environmental parameters is crucial for accurately assessing a user's health status. Furthermore, modern health management not only requires timely warnings of abnormal conditions but also seeks to provide personalized health advice to help users better manage their health. Therefore, a multi-functional vital sign monitoring method is needed that can integrate multi-dimensional data, dynamically adjust monitoring strategies, and achieve remote synchronization.

Currently, various vital sign monitoring devices and technologies exist on the market, including wearable devices, home medical instruments, and hospital-grade monitoring systems. These devices typically collect basic vital sign data and transmit it at preset, fixed frequencies. Some advanced devices also feature preliminary data analysis capabilities, enabling them to identify common anomalies and issue alerts. However, most existing solutions focus primarily on single-dimensional data collection, lacking comprehensive consideration of environmental parameters, and their fixed data collection strategies cannot be flexibly adjusted according to the user's specific condition and needs.

However, while existing technologies meet basic vital sign monitoring needs to some extent, several shortcomings remain. Fixed-frequency data collection may waste resources in low-risk situations while failing to capture critical changes in a timely manner in high-risk situations. Most existing devices rely on general algorithms for data analysis, failing to adequately consider individual differences and environmental factors, resulting in insufficiently personalized and targeted health recommendations. Many devices lack secure synchronization capabilities, hindering efficient telemedicine and continuous health management, and limiting doctors' ability to understand and intervene in patients' health conditions in real time.

In summary, existing vital sign monitoring technologies and equipment still have room for improvement in terms of flexibility, personalization, and remote services. This invention aims to improve user experience and the quality of telemedicine services by introducing intelligent control algorithms and real-time assessment engines, combined with multi-dimensional datasets, to achieve more accurate, efficient, and personalized vital sign monitoring.

Summary of the Invention

This application provides a multifunctional vital sign monitoring method and system to solve the problems of low accuracy and efficiency in health monitoring in the prior art.

In a first aspect, embodiments of this application provide a multifunctional vital sign monitoring method, including:

The system acquires real-time vital signs signals of users under different physiological states and combines them with environmental parameters to form a multi-dimensional monitoring dataset.

Based on the multi-dimensional monitoring dataset, an intelligent control algorithm is applied to dynamically adjust the monitoring frequency and accuracy of real-time vital signs signals, generate a data collection strategy, and collect vital signs data based on the data collection strategy.

Based on the vital signs data, the real-time assessment engine analyzes and processes the detected abnormal vital signs, and combines information such as gender, height, weight, genetic history and past medical history to generate health status assessment results and personalized health recommendations.

The health status assessment results and personalized recommendations are securely synchronized to the user's medical monitoring system to enable remote medical services and continuous health management.

Optionally, the step of dynamically adjusting the monitoring frequency and accuracy of real-time vital sign signals using an intelligent control algorithm based on the multi-dimensional monitoring dataset, generating a data collection strategy, and collecting vital sign data based on the data collection strategy includes:

By utilizing data analysis techniques and the aforementioned multi-dimensional monitoring dataset, we can identify changes in user behavior patterns, activity levels, and the environment they are in, thereby obtaining information that reflects the user's current state.

Based on the information reflecting the user's current state, assess the current state's requirements for monitoring frequency and accuracy, and generate preliminary monitoring requirement assessment results.

By combining historical vital sign data, a personalized monitoring frequency and accuracy strategy is developed, and the preliminary monitoring needs assessment results are optimized to obtain an optimized monitoring strategy.

Using the optimized monitoring strategy, the monitoring frequency and accuracy of the vital signs signal acquisition equipment are dynamically controlled to obtain vital signs data.

Optionally, the step of assessing the monitoring frequency and accuracy requirements under the current state based on the information reflecting the user's current state, and generating a preliminary monitoring requirement assessment result, includes:

Using a behavior pattern recognition algorithm, based on the information reflecting the user's current state, the user's activity type is classified to obtain the classification result of the user's activity type;

Based on the classification results of the user activity types and environmental parameters, the impact of the environmental parameters on the user's health status is evaluated to obtain the environmental adaptability assessment results;

Based on the environmental adaptability assessment results, a pre-trained context-aware model is applied to predict the user's potential health risk level in the current state and generate a health risk prediction report.

Based on the health risk prediction report, determine the current monitoring frequency and accuracy requirements to obtain the monitoring requirement level;

Based on the specific monitoring requirement level, a preliminary monitoring requirement assessment result is generated.

Optionally, by combining historical vital sign data to formulate personalized monitoring frequency and accuracy strategies, and optimizing the preliminary monitoring needs assessment results, an optimized monitoring strategy is obtained, including:

Using the preliminary monitoring needs assessment results, the monitoring needs of users in different scenarios are refined to generate specific monitoring needs descriptions.

Based on the specific monitoring requirements described above, integrate users' long-term behavioral patterns, health status change trends and vital sign data to construct a historical model of user behavior and health.

Based on the historical model of user behavior and health, a personalized prediction model suitable for users is constructed by applying machine learning algorithms. The personalized prediction model can predict the user's future behavior patterns and their impact on health.

Using the personalized prediction model and the specific monitoring requirements, strategies for monitoring frequency and accuracy under different scenarios are formulated, and optimized monitoring strategies are generated.

Optionally, based on the vital sign data, the real-time assessment engine analyzes and processes the detected abnormal vital signs, and combines information such as gender, height, weight, genetic history, and past medical history to generate a health status assessment result and personalized health recommendations, including:

The vital signs data are analyzed and processed in real time using an instant assessment engine to obtain a vital signs abnormality detection report.

The abnormal vital signs detection report is determined based on the fuzzy logic reasoning mechanism, and an abnormal vital signs confirmation result is generated.

Based on the confirmed results of abnormal vital signs, the user's current health status is assessed by comparing the user's long-term health data and similar cases, and a pre-trained health risk prediction model is used to generate a health status assessment result.

Using the health status assessment results, combined with the user's gender, height, weight, genetic history, lifestyle habits, activity level, and past medical history, customized health advice is provided to the user, generating personalized health recommendations.

Optionally, the step of determining the abnormal vital signs detection report based on the fuzzy logic reasoning mechanism and generating a confirmation result of abnormal vital signs includes:

Using fuzzy logic reasoning, the abnormalities in the vital sign abnormality detection report are analyzed to obtain a preliminary list of abnormal parameters.

Extract the key vital signs parameters corresponding to the preliminary abnormal parameter list, compare them with the preset standard range, identify parameter values that exceed the normal range, and generate an abnormal parameter set. The key vital signs parameters include heart rate, blood pressure, and respiratory rate.

Based on the set of abnormal parameters that exceed the standard range, and combined with the fuzzy logic rule base, a weighted set of abnormal parameters is generated by assigning weights to each abnormal parameter.

The weighted abnormal parameter set is comprehensively evaluated using a fuzzy inference algorithm. The score of each abnormal parameter is calculated by using a fuzzy membership function and inference rules. Combined with the overall health status, the result confirming abnormal vital signs is obtained.

Optionally, based on the confirmation of abnormal vital signs, the user's current health status is comprehensively assessed using a pre-trained health risk prediction model by comparing the user's long-term health data and similar cases, generating a health status assessment result, including:

Using a pre-trained health risk prediction model, based on the confirmation results of abnormal vital signs, a preliminary assessment of the user's current health status is performed to obtain a preliminary health risk assessment report.

Based on the preliminary health risk assessment report and combined with the user's long-term health data, an in-depth analysis of the user's health status is conducted to generate personalized health data analysis results. The long-term health data includes past medical history, lifestyle habits, and activity levels.

Based on the personalized health data analysis results, similar cases in the medical database are searched for comparative analysis to generate similar case comparison analysis results.

Using the results of the comparative analysis of similar cases, the parameters of the health risk prediction model are adjusted and the health risk prediction model is optimized to obtain the optimized health risk prediction model.

Based on the optimized health risk prediction model, and taking into account the abnormality confirmation results, long-term health data, and similar case information, the user's current health status is assessed, and a health status assessment result is generated.

Secondly, embodiments of this application provide a multifunctional vital sign monitoring system, including:

The acquisition module is used to acquire real-time vital signs signals of users under different physiological states, and combine them with environmental parameters to form a multi-dimensional monitoring dataset.

The adjustment module is used to dynamically adjust the monitoring frequency and accuracy based on the multi-dimensional monitoring dataset using an intelligent control algorithm, generate a data collection strategy, and minimize energy consumption to obtain vital sign data.

The analysis module is used to analyze and process the monitored abnormalities in vital signs based on the vital sign data using an instant assessment engine, and generate health status assessment results and personalized health recommendations by combining information such as gender, height, weight, genetic history and past medical history.

The synchronization module is used to securely synchronize the health status assessment results and personalized suggestions to the user's medical monitoring system, supporting remote medical services and enabling continuous health management.

Thirdly, embodiments of this application provide a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are to be invoked and executed by the processing component to implement a multifunctional vital sign monitoring method as described in the first aspect above.

Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when executed by a computer, implements a multifunctional vital sign monitoring method as described in the first aspect.

In this embodiment, real-time vital sign signals of the user under different physiological states are acquired and combined with environmental parameters to form a multi-dimensional monitoring dataset. Based on the multi-dimensional monitoring dataset, an intelligent control algorithm is applied to dynamically adjust the monitoring frequency and accuracy of the real-time vital sign signals, generate a data collection strategy, and collect vital sign data based on the data collection strategy. Based on the vital sign data, an instant assessment engine is used to analyze and process the detected abnormal vital signs, and combine information such as gender, height, weight, genetic history, and past medical history to generate a health status assessment result and personalized health recommendations. The health status assessment result and personalized recommendations are securely synchronized to the user's medical monitoring system to realize remote medical services and continuous health management.

The technical solution of this application has the following beneficial effects:

This application, by combining multi-dimensional real-time vital signs signals and environmental parameters, can more comprehensively reflect the user's health status, reducing false alarms and missed alarms. The intelligent control algorithm dynamically adjusts the monitoring frequency and accuracy according to actual conditions, avoiding unnecessary high-frequency data collection, thereby extending equipment lifespan and saving energy. The instant assessment engine provides personalized health suggestions, helping users better understand and manage their health status and improving user satisfaction. Securely synchronizing health status assessment results and suggestions to the medical monitoring system enables doctors to obtain the latest patient health information in a timely manner, achieving efficient remote diagnosis and treatment. Through continuous monitoring and data analysis, it provides users with long-term health trend analysis, helping to detect potential health problems early and take preventative measures.

Furthermore, this application also relates to the application of intelligent control algorithms and real-time assessment engines to achieve dynamic monitoring of user vital sign data and real-time assessment of health status. Specifically, by utilizing data analysis techniques and multi-dimensional monitoring datasets, user behavior patterns, activity levels, and environmental changes are identified to obtain information reflecting the user's current state. Based on this information, the required monitoring frequency and accuracy under the current state are assessed, generating preliminary monitoring requirement assessment results. Combined with historical vital sign data optimization strategies, personalized monitoring frequency and accuracy strategies are formulated to dynamically control the monitoring parameters of the acquisition equipment. In addition, the real-time assessment engine is used to analyze and process vital sign data in real time, generating anomaly detection reports and confirming anomalies through fuzzy logic reasoning. Based on the anomaly confirmation results, the user's long-term health data and similar cases are compared, and a pre-trained health risk prediction model is used to assess the user's current health status, ultimately generating a health status assessment result and customized health recommendations.

Through the methods described above, this application significantly improves the accuracy and flexibility of vital sign monitoring. First, the intelligent control algorithm dynamically adjusts the monitoring frequency and accuracy based on the user's real-time status, ensuring the capture of key health information under different physiological conditions while avoiding resource waste. Second, the introduction of the real-time assessment engine enables rapid identification and confirmation of abnormalities, enhancing the timeliness of early warnings. Finally, by combining the user's historical health data and similar cases, personalized health recommendations are generated, improving the targeting and effectiveness of health management. Overall, this method not only optimizes resource utilization but also significantly improves the quality of telemedicine services and user experience, achieving efficient, accurate, and personalized continuous health management.

These or other aspects of this application will become more apparent in the following description of the embodiments.

Attached Figure Description

To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

Figure 1 shows a flowchart of a multifunctional vital sign monitoring method provided in this application;

Figure 2 shows a schematic diagram of the structure of a multifunctional vital signs monitoring system provided in this application;

Figure 3 shows a schematic diagram of the structure of a computing device provided in this application.

Detailed Implementation

To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

Figure 1 is a flowchart of a multifunctional vital sign monitoring method provided in an embodiment of this application. As shown in Figure 1, the method includes:

101. Acquire real-time vital signs signals of users under different physiological states, and combine them with environmental parameters to form a multi-dimensional monitoring dataset;

This step involves collecting real-time vital signs signals from multiple sensors and data sources, including but not limited to heart rate, blood pressure, blood oxygen saturation, and respiratory rate. These physiological parameters reflect the user's current physical condition. Simultaneously, environmental parameters such as temperature, humidity, air quality, and light intensity are also incorporated to provide more comprehensive background information. By integrating these physiological and environmental data, a multi-dimensional monitoring dataset is formed for subsequent analysis and decision-making.

First, the system collects the user's vital signs data in real time through wearable devices (such as smartwatches and wristbands) and other medical instruments (such as blood pressure monitors and pulse oximeters). Simultaneously, environmental sensors (such as temperature and humidity sensors and air quality monitors) record the surrounding environmental conditions. All this data is transmitted to a central processing unit or cloud platform for initial data cleaning and formatting to ensure data integrity and consistency. Then, this multi-dimensional data is stored in a structured database, providing the foundation for subsequent intelligent control algorithms and real-time evaluation engines.

Imagine a home care scenario where an elderly person with chronic obstructive pulmonary disease (COPD) is wearing a multifunctional health monitoring bracelet. This bracelet not only continuously monitors heart rate, blood oxygen saturation, and respiratory rate, but also records indoor temperature and air quality through built-in environmental sensors. When the elderly person is resting at home, the system automatically collects this data and uploads it to a cloud platform. By combining this physiological and environmental data, the system can better understand the elderly person's health status; for example, changes in respiratory rate in poor air quality may indicate a risk of worsening condition.

102. Based on the multi-dimensional monitoring dataset, apply an intelligent control algorithm to dynamically adjust the monitoring frequency and accuracy of real-time vital signs signals, generate a data collection strategy, and collect vital signs data based on the data collection strategy.

This step aims to utilize intelligent control algorithms to analyze multi-dimensional monitoring datasets, identify user behavior patterns, activity levels, and changes in their environment, thereby dynamically adjusting monitoring frequency and accuracy. This not only improves the flexibility and targeting of data collection but also avoids unnecessary resource waste. By learning from users' historical data and behavioral habits, intelligent control algorithms can predict and adapt to different user needs, ensuring that critical information is captured at critical moments.

The intelligent control algorithm first performs in-depth analysis of the multi-dimensional monitoring dataset to identify changes in the user's daily behavior patterns and activity levels. For example, it distinguishes between resting and active states, or identifies different stages in the sleep cycle. Then, based on this information, the algorithm assesses the monitoring frequency and accuracy requirements under the current state, generating a preliminary monitoring needs assessment. Next, it combines the user's historical vital sign data to optimize the preliminary assessment and formulate personalized monitoring frequency and accuracy strategies. Finally, based on the optimized strategy, it dynamically controls the operating mode of the vital sign signal acquisition equipment to ensure efficient and accurate data collection.

Continuing the previous example, in a home care scenario, the intelligent control algorithm learns from the elderly person's historical data that they typically experience a brief increase in heart rate upon waking in the morning, while their heart rate remains relatively stable during sleep at night. Therefore, the system appropriately increases the monitoring frequency of heart rate and blood oxygen saturation in the morning to detect potential abnormalities; while at night, the monitoring frequency is reduced to save power and minimize interference. Furthermore, when indoor air quality deteriorates, the system automatically increases the accuracy of respiratory rate monitoring to promptly detect any abnormal changes and ensure the elderly person's safety.

103. Based on the vital signs data, the real-time assessment engine is used to analyze and process the abnormal vital signs detected, and combined with information such as gender, height, weight, genetic history and past medical history, to generate health status assessment results and personalized health recommendations.

This step utilizes a real-time assessment engine to analyze and process vital sign data in real time, identifying and confirming abnormalities. The real-time assessment engine, combined with fuzzy logic reasoning, can quickly and accurately determine abnormal events and generate detailed anomaly detection reports. Based on these reports, the system further compares the user's long-term health data with similar cases, using a pre-trained health risk prediction model to assess the user's current health status, ultimately generating a health status assessment result and customized health recommendations. This process not only improves the timeliness of alerts but also provides personalized health management guidance.

The real-time assessment engine receives vital sign data from the acquisition device and performs real-time data analysis and anomaly detection. Once an anomaly is detected, the engine uses fuzzy logic reasoning to confirm the anomaly type and severity, generating an anomaly detection report. Subsequently, the system compares the anomaly detection results with the user's long-term health data (such as gender, height, weight, genetic history, past medical history, and lifestyle habits) and searches similar cases in the medical database to find matching reference cases. A pre-trained health risk prediction model uses this information to assess the user's current health status, generating a detailed health assessment result. Finally, the system combines the user's lifestyle habits, activity level, and past medical history to provide customized health recommendations, helping them take appropriate preventative and treatment measures.

Continuing with the home care scenario above, when the real-time assessment engine detects a sudden drop in the elderly person's blood oxygen saturation, the system immediately activates a fuzzy logic reasoning mechanism to confirm that this is an acute hypoxic event. The system then retrieves the elderly person's historical health data, discovering that they have experienced similar hypoxic symptoms in the past, and reviews similar cases, concluding that it is likely due to poor indoor air quality. After assessment, the health risk prediction model determines that there is a high health risk in this situation and recommends that the elderly person immediately improve indoor ventilation and contact a doctor. The system will also remind family members to pay attention to the elderly person's condition and provide specific coping measures, such as adjusting the room ventilation time and preparing oxygen cylinders.

104. Securely synchronize the health status assessment results and personalized recommendations to the user's medical monitoring system to achieve remote medical services and continuous health management.

This step securely synchronizes the health assessment results and personalized recommendations to the user's medical monitoring system, ensuring efficient sharing of this information among doctors, patients, and their families. Encrypted communication technologies and security protocols guarantee the security and privacy of data transmission. Telemedicine services enable doctors to understand the patient's latest health status in real time, conduct remote diagnosis and intervention, while continuous health management helps track the patient's health trends over the long term, providing ongoing personalized support.

The system transmits the generated health status assessment results and personalized recommendations to the user's medical monitoring system via a secure channel. Advanced encryption standards and other encryption technologies are used during this process to ensure that data is not stolen or tampered with during transmission. Upon receiving the data, the medical monitoring system automatically updates the patient's electronic health record and notifies relevant medical staff. Doctors can view the patient's latest health information at any time through mobile applications or web browsers, and provide remote diagnosis and treatment recommendations. In addition, the system regularly sends health reports and suggestions to patients and their families to maintain smooth communication and ensure that patients receive continuous attention and support.

In the aforementioned home care scenario, the system synchronizes the elderly person's health assessment results and personalized suggestions to their attending physician's mobile application via an encrypted channel. The doctor can view the elderly person's latest health data at any time during outpatient visits, understanding the reasons for any decline in blood oxygen saturation and the system's recommendations. Simultaneously, the doctor can also communicate directly with the elderly person and their family through the application, guiding them on how to improve indoor air quality and arranging necessary medical examinations. Furthermore, the system regularly sends health reports to the elderly person and their family, helping them better manage their health and ensuring a continuous improvement in their quality of life.

Through the implementation of steps 101 to 104, this application achieves intelligent management of the entire process from data collection to telemedicine services. First, the construction of a multi-dimensional monitoring dataset ensures the comprehensiveness and accuracy of the data. Second, the intelligent control algorithm dynamically adjusts the monitoring strategy, improving resource utilization efficiency and the targeting of data collection. Third, the introduction of a real-time evaluation engine significantly enhances the ability to identify and warn of abnormal situations. Finally, a secure synchronization mechanism guarantees the effective transmission of information and privacy protection, achieving efficient telemedicine services and continuous health management. Overall, this method not only optimizes each stage of health monitoring but also significantly improves user experience and the quality of medical services, providing users with a comprehensive, intelligent, and efficient health management solution.

Optionally, step 102, which involves dynamically adjusting the monitoring frequency and accuracy of real-time vital signs signals using an intelligent control algorithm based on the multi-dimensional monitoring dataset to generate a data collection strategy and collecting vital sign data based on the data collection strategy, includes: using data analysis techniques and the multi-dimensional monitoring dataset to identify changes in the user's behavior patterns, activity levels, and environment to obtain information reflecting the user's current state; assessing the monitoring frequency and accuracy requirements under the current state based on the information reflecting the user's current state to generate a preliminary monitoring requirement assessment result; combining historical vital sign data to formulate a personalized monitoring frequency and accuracy strategy, optimizing the preliminary monitoring requirement assessment result to obtain an optimized monitoring strategy; and using the optimized monitoring strategy to dynamically control the monitoring frequency and accuracy of the vital sign signal acquisition device to obtain vital sign data.

To address the issue of insufficient flexibility and personalization in monitoring strategies, in some embodiments, step 102, which involves assessing the monitoring frequency and accuracy requirements based on the information reflecting the user's current state, and generating a preliminary monitoring requirement assessment result, includes:

Using a behavior pattern recognition algorithm, the user's activity types are classified based on information reflecting the user's current state, resulting in a classification result. Based on the classification result and environmental parameters, the impact of these parameters on the user's health is assessed, yielding an environmental adaptability assessment result. Based on this assessment result, a pre-trained context-aware model is applied to predict the user's potential health risk level in the current state, generating a health risk prediction report. Based on the health risk prediction report, the required monitoring frequency and accuracy in the current state are determined, resulting in a monitoring requirement level. Based on this specific monitoring requirement level, a preliminary monitoring requirement assessment result is generated.

In this embodiment, a behavior pattern recognition algorithm is used to analyze the user's activity type, such as resting, walking, running, or sleeping. By combining data from sensors (such as accelerometers and gyroscopes), the user's activities can be categorized. This data not only reflects the user's daily behavior patterns but also provides important information about their physical condition. Environmental parameters, including temperature, humidity, air quality, and light intensity, can affect the user's health. For example, high humidity may worsen respiratory problems, while low air quality may increase the risk of cardiovascular disease. Therefore, environmental parameters are an important component of assessing a user's health risk. The context-aware model is a pre-trained machine learning model that predicts potential health risk levels based on the user's behavior patterns and environmental parameters. Trained using extensive historical data and similar cases, it provides accurate health risk predictions and guides adjustments to monitoring strategies. The health risk prediction report, based on the output of the context-aware model, details the user's potential health risks in their current state, including risk level, possible influencing factors, and recommended measures. This report forms the basis for developing specific monitoring needs levels. The monitoring demand level is determined by the health risk prediction report. The system identifies the specific needs for monitoring frequency and accuracy under the current conditions, categorizing them into high demand (frequent and accurate), medium demand (moderate), and low demand (lower frequency and accuracy). This helps optimize resource utilization and ensures that key health information can be captured under different circumstances.

In this embodiment, firstly, the algorithm identifies the user's activity patterns, such as resting, walking, or exercising, by analyzing sensor data (e.g., accelerometer, gyroscope). Secondly, the system considers the user's environmental conditions (e.g., temperature, humidity, air quality) and, in conjunction with the activity type, assesses the potential impact of these environmental factors on the user's health. Next, the model uses historical data and similar cases to predict the user's health risk level under current activity and environmental conditions. Further, based on the content of the health risk prediction report, the system decides whether more frequent and precise data collection is needed, or whether the monitoring intensity can be appropriately reduced to save energy. Finally, the system integrates all assessment information to form a final preliminary monitoring needs assessment result, guiding subsequent data collection strategies.

Here is a specific example:

In a home care scenario, suppose a child with asthma wears a smart bracelet equipped with an accelerometer, gyroscope, and environmental sensors. While the child is playing at home, a behavioral pattern recognition algorithm identifies that he is engaging in vigorous exercise. Simultaneously, the environmental sensors record poor indoor air quality with a high PM2.5 index. Based on this information, the system assesses the impact of these environmental parameters on the child's health and concludes that poor air quality may worsen asthma symptoms.

Next, the context-aware model, based on the child's historical health data and similar cases, predicts a high level of health risk under the current activity and environmental conditions, generating a detailed health risk prediction report. Based on this report, the system determines that the monitoring frequency and accuracy need to be increased in the current state to better capture any abnormal changes. Therefore, the system sets the monitoring demand level to "high demand" and increases the monitoring frequency of heart rate, respiratory rate, and blood oxygen saturation for a period of time.

Finally, the system generates preliminary monitoring needs assessment results to guide subsequent vital sign data collection. This dynamic adjustment not only improves the targeting of monitoring but also ensures that key health information can be captured in a timely manner at critical moments, helping parents and doctors better manage children's health.

To address the issue of insufficient personalization in preliminary monitoring needs assessment results, in some embodiments, step 102 involves combining historical vital sign data to formulate personalized monitoring frequency and accuracy strategies, thereby optimizing the preliminary monitoring needs assessment results and obtaining an optimized monitoring strategy, including:

Using the preliminary monitoring needs assessment results, the monitoring needs of users in different scenarios are refined to generate specific monitoring needs descriptions. Based on these specific monitoring needs descriptions, the user's long-term behavioral patterns, health status trends, and vital sign data are integrated to construct a historical model of user behavior and health. Based on this historical model, machine learning algorithms are applied to construct a personalized prediction model suitable for the user, wherein the personalized prediction model can predict the user's future behavioral patterns and their impact on health. Using the personalized prediction model and the specific monitoring needs descriptions, monitoring frequency and accuracy strategies for different scenarios are formulated to generate optimized monitoring strategies.

In this embodiment, the specific monitoring needs description is a detailed description of the user's monitoring needs in different situations, including specific activity types (such as resting, exercise), environmental conditions (such as temperature, humidity), and corresponding health risk levels. These descriptions are used to guide the subsequent development of monitoring strategies. The historical model of user behavior and health integrates the user's long-term behavioral patterns, health status change trends, and vital sign data, comprehensively reflecting the relationship between the user's daily activities and health status. By analyzing this data, potential factors affecting health can be identified, and future behavioral patterns and health risks can be predicted. The personalized prediction model is a model built based on machine learning algorithms, trained using the user's historical data, capable of predicting the user's future behavioral patterns and their impact on health. This model can provide highly personalized health advice and monitoring strategies based on the user's individual differences. The optimized monitoring strategy is based on the personalized prediction model and the specific monitoring needs description, and the system formulates monitoring frequency and accuracy strategies for different situations. These strategies not only consider the current health status but also anticipate possible future changes, ensuring the flexibility and targeting of monitoring.

In this embodiment, firstly, based on preliminary assessment results, the system details the user's monitoring needs under different activity types and environmental conditions, such as the need for more frequent heart rate monitoring during exercise in high humidity environments. Secondly, the system analyzes the user's long-term data to identify the correlation between behavioral patterns and health status; for example, it discovers that a user's elevated blood pressure during a specific time period is related to work stress. Next, the system uses machine learning technology to train a personalized prediction model based on historical model data, enabling it to accurately predict the user's future behavioral patterns and health impacts. Finally, based on the output of the personalized prediction model and the specific monitoring needs, the system dynamically adjusts the monitoring strategy to ensure that key health information is captured under different circumstances while avoiding resource waste.

Here is a specific example:

In a home care scenario, suppose an elderly patient with hypertension wears a smart health monitoring bracelet. The bracelet collects data in real time, including heart rate, blood pressure, and activity level, and combines this data with environmental parameters (such as temperature and humidity). When the system, based on the initial monitoring needs assessment, finds that the patient's blood pressure typically rises within an hour of waking up in the morning, it refines the monitoring needs for this scenario, generating a specific monitoring requirement description, indicating that the frequency of blood pressure monitoring needs to be increased during this period.

Next, the system integrates patients' long-term behavioral patterns (such as the habit of taking a walk every morning), trends in health status changes (such as larger fluctuations in blood pressure in the morning), and vital sign data to construct a historical model of user behavior and health. By analyzing this data, the system identifies that morning walks may be one of the causes of elevated blood pressure.

Then, the system applies machine learning algorithms to build a personalized predictive model for the patient based on historical models. This model can not only predict the patient's future behavioral patterns (such as whether they will continue to take morning walks), but also assess their impact on health (such as the specific impact of morning walks on blood pressure).

Finally, the system utilizes personalized predictive models and specific monitoring needs to develop strategies for monitoring frequency and accuracy in the context of morning walks. For example, it increases the frequency of blood pressure monitoring half an hour before the walk and maintains high-frequency monitoring during the walk. This optimized monitoring strategy not only improves the targeting of monitoring but also helps doctors better understand patients' health status in daily life, thereby providing more personalized health management recommendations.

In this way, the system not only solves the problem of insufficient personalization in the initial monitoring needs assessment results, but also achieves forward-looking and flexible management of users' health status, significantly improving the quality of telemedicine services and user experience.

To address the issue of insufficient accuracy in detecting abnormal vital signs and providing health advice, in some embodiments, step 103 involves analyzing and processing the detected abnormal vital signs based on the vital sign data using a real-time assessment engine. This analysis, combined with information on gender, height, weight, genetic history, and past medical history, generates a health status assessment result and personalized health advice, including...

Using an instant assessment engine, the vital sign data is analyzed and processed in real time to obtain a vital sign abnormality detection report. Based on a fuzzy logic reasoning mechanism, the vital sign abnormality detection report is determined, generating a vital sign abnormality confirmation result. Based on the vital sign abnormality confirmation result, by comparing the user's long-term health data and similar cases, a pre-trained health risk prediction model is used to assess the user's current health status, generating a health status assessment result. Using the health status assessment result, combined with the user's lifestyle habits, activity level, and past medical history, customized health suggestions are provided to the user, generating personalized health advice.

In this embodiment, the real-time assessment engine is a real-time processing system capable of rapidly analyzing vital sign data (such as heart rate, blood pressure, and blood oxygen saturation) from sensors to identify potential anomalies. Through complex algorithms and rule sets, the engine can generate detailed vital sign anomaly detection reports in a short time. These reports meticulously record any anomalies detected during monitoring, including the specific values of abnormal parameters, the time of occurrence, and their severity. They not only provide a snapshot of the current state but also lay the foundation for further analysis. The fuzzy logic reasoning mechanism is a mathematical method for handling uncertainty and fuzzy information. In this solution, it is used to confirm anomalies in the vital sign anomaly detection reports, considering multiple factors (such as data fluctuations and environmental influences) to determine the authenticity and severity of the anomalies. The vital sign anomaly confirmation results are the final confirmed vital sign anomalies after processing by the fuzzy logic reasoning mechanism. These confirmation results form the basis for further health assessments, ensuring the accuracy and reliability of anomaly detection. The health risk prediction model is a pre-trained machine learning model capable of predicting the risk level of the user's current health status based on the user's long-term health data and similar cases. The model is trained using extensive historical data and similar cases to improve predictive accuracy. Personalized health recommendations are based on health status assessments, combined with the user's lifestyle habits, activity levels, and medical history, providing customized health advice. These recommendations aim to help users take appropriate preventative and treatment measures to improve their health.

In this embodiment, firstly, the real-time assessment engine receives and processes data from sensors, identifies any anomalies, and generates a detailed anomaly detection report. Secondly, the system applies a fuzzy logic reasoning mechanism, comprehensively considering multiple factors (such as data fluctuations and environmental influences), to confirm the authenticity and severity of the anomaly, generating a final confirmation result of abnormal vital signs. Next, the system utilizes a health risk prediction model, combined with the user's historical health data and similar cases, to comprehensively assess the current health status, generating a detailed health status assessment result. Finally, based on the health status assessment result, the system provides specific health recommendations to help the user take appropriate preventative and treatment measures to improve their health status.

Here is a specific example:

In a home care scenario, suppose an elderly person with chronic obstructive pulmonary disease wears a multi-functional health monitoring bracelet. The bracelet continuously monitors heart rate, blood oxygen saturation, and respiratory rate, and transmits the data to an instant assessment engine in real time.

While the elderly man was resting at home, the real-time assessment engine detected that his blood oxygen saturation suddenly dropped to 85%, below the normal range. The system immediately generated a vital signs anomaly detection report, recording the specific values of the abnormality and the time of occurrence.

Next, the system applies a fuzzy logic reasoning mechanism, taking into account the elderly person's daily activity patterns and the current indoor air quality (low humidity, high PM2.5 index), to confirm that this drop in blood oxygen saturation is an acute hypoxic event, generating a confirmation result of abnormal vital signs.

The system then uses a pre-trained health risk prediction model, combined with the elderly person's historical health data (such as past medical history and lifestyle habits) and similar cases, to assess their current health status. The model predicts a high health risk in this situation, especially considering the elderly person's past experience with similar hypoxia symptoms.

Finally, based on the health assessment results, the system provides the elderly with customized health recommendations. For example, it may suggest immediately improving indoor ventilation, increasing air humidity, and contacting a doctor for further examination. Simultaneously, the system reminds family members to monitor the elderly person's condition and prepare necessary medical equipment (such as oxygen cylinders). Furthermore, the system regularly sends health reports to the elderly person and their family to help them better manage their health.

In this way, the system not only solves the problem of inaccurate detection of abnormal vital signs and health advice, but also realizes real-time monitoring and personalized management of users' health status, significantly improving the quality of telemedicine services and user experience.

To address the issue of insufficient accuracy in confirming abnormalities in vital sign abnormality detection reports, in some embodiments, step 103, which involves determining the vital sign abnormality detection report based on a fuzzy logic reasoning mechanism and generating a vital sign abnormality confirmation result, includes:

Using a fuzzy logic reasoning mechanism, the abnormalities in the vital sign anomaly detection report are analyzed to obtain a preliminary list of abnormal parameters. Key vital sign parameters corresponding to the preliminary list of abnormal parameters are extracted and compared with preset standard ranges to identify parameter values exceeding the normal range, generating an abnormal parameter set. These key vital sign parameters include heart rate, blood pressure, and respiratory rate. Based on the abnormal parameter set exceeding the standard range, and combined with a fuzzy logic rule base, each abnormal parameter is weighted to generate a weighted abnormal parameter set. Using a fuzzy inference algorithm, the weighted abnormal parameter set is comprehensively evaluated. Through fuzzy membership functions and inference rules, a score for each abnormal parameter is calculated, and combined with the overall health status, a confirmation result of the abnormal vital signs is obtained.

In this embodiment, fuzzy logic reasoning is a mathematical method for handling uncertainty and fuzzy information, suitable for describing and processing imprecise data. In this scheme, fuzzy logic reasoning is used to analyze anomalies in vital sign anomaly detection reports, considering multiple factors (such as data fluctuations and environmental influences) to determine the authenticity and severity of the anomalies. The preliminary list of abnormal parameters is obtained after initial screening of all anomalies in the vital sign anomaly detection reports. This list contains all possible abnormal parameters and their corresponding values, providing a foundation for further analysis. Key vital sign parameters are those closely related to health status, including heart rate, blood pressure, and respiratory rate. They have significant reference value in assessing a user's health status and can provide key clues about potential health problems. The anomaly parameter set is a set generated by extracting key vital sign parameters from the preliminary anomaly parameter list and comparing them with preset standard ranges. It only contains parameter values that exceed the normal range, used for subsequent weighting and comprehensive evaluation. The weighted anomaly parameter set is a set generated by assigning weights to each anomaly parameter according to a fuzzy logic rule base. The weights reflect the degree of influence of different parameters on the overall health status, ensuring the accuracy and personalization of the evaluation. A fuzzy membership function is a mathematical tool used to map input variables to membership values within the interval [0, 1], representing the degree or probability of a condition. In this scheme, the fuzzy membership function is used to evaluate the severity of each anomalous parameter. Fuzzy inference rules are a set of predefined rules that guide the fuzzy inference algorithm on how to calculate the output based on the input data. In this scheme, fuzzy inference rules help the system combine the overall health status of multiple anomalous parameters to arrive at the final confirmation result of abnormal vital signs.

In this embodiment, firstly, the system identifies and lists all possible abnormal parameters and their corresponding values as the basis for further analysis. Secondly, the system retains only key vital sign parameters (such as heart rate, blood pressure, and respiratory rate) and compares them with preset standard ranges to identify parameter values that exceed the normal range. Next, the system assigns appropriate weights to each abnormal parameter based on a fuzzy logic rule base to reflect its impact on overall health. Finally, the system uses a fuzzy inference algorithm to comprehensively evaluate the scores of all abnormal parameters based on fuzzy membership functions and inference rules, ultimately generating a vital sign abnormality confirmation result to ensure the accuracy and reliability of the assessment.

Here is a specific example:

In a home care scenario, suppose an elderly person with chronic obstructive pulmonary disease wears a multi-functional health monitoring bracelet. The bracelet continuously monitors heart rate, blood oxygen saturation, and respiratory rate, and transmits the data to an instant assessment engine in real time.

While the elderly man was resting at home, the real-time assessment engine detected that his blood oxygen saturation suddenly dropped to 85%, below the normal range. The system immediately generated a vital signs anomaly detection report, recording the specific values of the abnormality and the time of occurrence.

Next, the system uses fuzzy logic reasoning to perform a preliminary analysis of all the anomalies in the anomaly detection report, generating a preliminary list of abnormal parameters, including parameters such as blood oxygen saturation, heart rate, and respiratory rate.

The system then extracted key vital signs (heart rate, blood pressure, and respiratory rate) and compared them with preset standard ranges. The results showed that blood oxygen saturation (85%) was significantly lower than the normal range (90%-100%), while heart rate and respiratory rate were also within the borderline range. The system generated an abnormal parameter set containing only blood oxygen saturation, which was outside the normal range.

Next, the system assigns weighted values to blood oxygen saturation based on the set of abnormal parameters and a fuzzy logic rule base. According to the rule base, blood oxygen saturation has a relatively high weight because it has a direct impact on the health status of patients with chronic obstructive pulmonary disease. The system generates a weighted set of abnormal parameters, in which blood oxygen saturation is assigned a higher weight value.

Finally, the system uses a fuzzy inference algorithm to comprehensively evaluate the weighted abnormal parameter set. Through fuzzy membership functions and inference rules, the system calculates the blood oxygen saturation score and, combined with the elderly person's overall health status (such as past medical history and lifestyle habits), arrives at the final confirmation result of the abnormal vital signs—an acute hypoxic event. The system confirms the authenticity and severity of this abnormal situation and generates a detailed abnormality confirmation report.

In this way, the system not only solves the problem of insufficient accuracy in confirming abnormalities in vital sign abnormality detection reports, but also achieves rapid and accurate identification of abnormalities, significantly improving the quality of telemedicine services and user experience. This meticulous assessment method helps doctors better understand the patient's current health status, thereby enabling them to take timely and effective intervention measures.

To address the issue of insufficient comprehensiveness and personalization in health status assessments, in some embodiments, step 103, based on the confirmation of abnormal vital signs, uses a pre-trained health risk prediction model to comprehensively assess the user's current health status by comparing the user's long-term health data and similar cases, generating a health status assessment result, including:

Using a pre-trained health risk prediction model, based on the confirmed results of abnormal vital signs, a preliminary assessment of the user's current health status is performed, resulting in a preliminary health risk assessment report. Based on this report and the user's long-term health data, a deep analysis of the user's health status is conducted, generating personalized health data analysis results. This long-term health data includes past medical history, lifestyle habits, and activity levels. Based on these personalized health data analysis results, similar cases are searched in a medical database for comparative analysis, generating similar case comparison analysis results. Using these results, the parameters of the health risk prediction model are adjusted to optimize it, resulting in an optimized model. Finally, based on this optimized model, and considering the confirmed results of abnormalities, long-term health data, and similar case information, the user's current health status is assessed, generating a health status assessment result.

In this embodiment, the pre-trained health risk prediction model is a machine learning model trained on a large amount of historical data and similar cases. It can quickly assess a user's preliminary health risk based on the input vital sign abnormality confirmation results. This model is trained using advanced algorithms (such as neural networks and decision trees) to ensure the accuracy and reliability of the assessment. The preliminary health risk assessment report is the first assessment report generated based on the vital sign abnormality confirmation results, detailing the user's preliminary risk level under their current health condition. This report provides foundational information for subsequent in-depth analysis. Long-term health data includes the user's past medical history, lifestyle habits, and activity levels, providing long-term background information on the user's health status. Long-term health data is crucial for identifying potential health problems and developing personalized health management strategies. The personalized health data analysis result is generated by deeply analyzing the user's long-term health data, resulting in a detailed personalized health data analysis. This result not only reflects the user's current health condition but also reveals their health trends and potential risk factors. The similar case comparison analysis result is generated by searching for similar cases in the medical database and comparing them with the user's situation. Similar cases provide additional reference information, helping to more accurately assess the user's health risk. The optimized health risk prediction model is based on personalized health data analysis and comparative analysis of similar cases. The pre-trained model has been systematically adjusted and optimized to better adapt to individual user differences. The optimized model improves prediction accuracy and better reflects the user's actual health status.

In this embodiment, firstly, the system inputs the anomaly confirmation result into a pre-trained health risk prediction model to generate a preliminary health risk assessment report, recording the user's current health risk level. Secondly, the system deeply analyzes the user's long-term health data, identifies potential factors that may affect health, and generates detailed personalized health data analysis results. Next, the system searches for cases similar to the user's situation, conducts comparative analysis, identifies common problems and solutions in similar cases, and generates comparative analysis results for similar cases. Furthermore, based on the information from similar cases, the system adjusts the parameters of the health risk prediction model to better adapt to individual differences in users and improve prediction accuracy. Finally, the system uses the optimized health risk prediction model, combined with all relevant information, to generate the final health status assessment result, ensuring the comprehensiveness and personalization of the assessment.

Here is a specific example:

In a home care scenario, suppose an elderly person with chronic obstructive pulmonary disease wears a multi-functional health monitoring bracelet. The bracelet continuously monitors heart rate, blood oxygen saturation, and respiratory rate, and transmits the data to an instant assessment engine in real time.

While the elderly man was resting at home, the system detected that his blood oxygen saturation suddenly dropped to 85%, below the normal range. The system immediately generated a vital signs abnormality detection report and, through a fuzzy logic reasoning mechanism, confirmed that this was an acute hypoxic event, generating a vital signs abnormality confirmation result.

Next, the system uses a pre-trained health risk prediction model to conduct a preliminary assessment of the elderly person's current health status based on this acute hypoxia event, generating a preliminary health risk assessment report that indicates a high level of acute health risk.

The system then combines the elderly person's long-term health data (such as medical history, lifestyle habits, and activity level) to conduct an in-depth analysis of their health status. The system discovered that the elderly person had a history of chronic obstructive pulmonary disease and had recently increased their outdoor activity time, potentially exposing them to environments with poor air quality. The system generated a detailed, personalized health data analysis, revealing the impact of these factors on the elderly person's health.

Next, the system searched the medical database for similar cases and found several patients with chronic obstructive pulmonary disease who experienced acute hypoxic events in similar environments. Through comparative analysis, the system found that these patients shared common characteristics: poor indoor air quality and excessive exercise, generating a comparative analysis of similar cases.

Subsequently, the system used comparative analysis of similar cases to adjust the parameters of the health risk prediction model, optimizing the model to better adapt to individual differences among the elderly. The optimized model improved the accuracy of predicting acute hypoxic events.

Finally, based on the optimized health risk prediction model, and taking into account anomaly confirmation results, long-term health data, and information from similar cases, the system comprehensively assesses the elderly person's current health status and generates a detailed health assessment report. The system recommends that the elderly person immediately improve indoor ventilation, reduce outdoor activities, and contact a doctor for further examination. In addition, the system will regularly send health reports to the elderly person and their family to help them better manage their health.

In this way, the system not only solves the problems of insufficient comprehensiveness and personalization in health assessment, but also achieves accurate assessment and personalized management of users' health status, significantly improving the quality of telemedicine services and user experience. This meticulous assessment method helps doctors better understand patients' current health status, thereby enabling them to take timely and effective intervention measures.

This application considers that with the increasing focus on health management, especially the needs of patients with chronic diseases and the elderly, accurate and personalized health monitoring has become particularly important. Traditional vital sign monitoring methods often use fixed-frequency data collection, which cannot be flexibly adjusted according to the user's real-time status, leading to wasted resources or missed key health information. Therefore, a dynamic monitoring strategy based on intelligent control algorithms has been developed. This strategy aims to accurately assess the monitoring frequency and accuracy requirements of the user's current state through behavioral pattern recognition, environmental parameter evaluation, and context-aware models. Thus, a new alternative solution is proposed, which includes:

Based on the information reflecting the user's current state, assess the monitoring frequency and accuracy requirements under the current state, and generate preliminary monitoring requirement assessment results, including:

Using a behavior pattern recognition algorithm, based on the information reflecting the user's current state, the user's activity type is classified to obtain the classification result of the user's activity type;

The weighted activity type classification accuracy A _{_acc} is calculated using the following formula:

Where A _{_acc} represents the weighted activity type classification accuracy, N represents the total number of samples, C_{_i} represents the classification result of the i-th sample, and T _{_i} represents the true label of the i-th sample. This is an indicator function that returns 1 when the condition is true and 0 otherwise. _{w_act,i} represents the activity weight of the i-th sample, used to emphasize the importance of different activity types.

Based on the classification results of user activity types and combined with environmental parameters, a comprehensive assessment is conducted on how these factors affect user health, resulting in an environmental adaptability assessment.

The environmental adaptability score E _score is calculated using the following formula:

Where E _score represents the environmental adaptability score, M represents the number of environmental parameters, _wj represents the weight of the j-th environmental parameter, _Ej represents the influence value of the j-th environmental parameter (such as temperature, humidity, etc.), and _βj is the exponential factor of the j-th environmental parameter, used to adjust the influence intensity of each parameter.

Based on the environmental adaptability assessment results, a pre-trained context-aware model is applied to predict the potential health risk level of users in the current state and generate a health risk prediction report.

The health risk probability P_{_risk} is calculated using the following formula:

P _risk =σ(W ^T x+b+λ·log(E _score ))

Where P _{_risk} represents the probability of health risk, σ is the Sigmoid function, which is used to map the result of the linear combination to the interval [0, 1], representing the likelihood of health risk, W is the weight vector of the context-aware model, x is the input feature vector, b is the bias term, and λ is the adjustment coefficient, which is used to adjust the impact of environmental adaptability score on health risk prediction.

Based on the health risk prediction report, the specific requirements for monitoring frequency and accuracy under the current conditions are determined. In high-risk situations, more frequent and more accurate data collection is required; while in low-risk or stable situations, the monitoring intensity can be appropriately reduced to save energy, thus obtaining the specific monitoring requirement level.

The specific monitoring requirement _level D is obtained using the following formula:

Where D _level represents the monitoring demand level, P _threshold is the preset risk threshold used to distinguish between high demand and low demand states, δ is the time sensitivity coefficient, and Δt is the time interval since the last assessment, used to consider the increase in risk over time.

By integrating the specific monitoring requirement levels, a preliminary monitoring requirement assessment result is generated.

The following is a detailed explanation of each parameter:

N: Total number of samples. This refers to the total number of samples in the dataset used to evaluate behavior pattern recognition algorithms. It is usually collected over a period of time using data acquisition devices (such as smart bracelets, sensors).

C _{_i} : The classification result of the i-th sample. This is the result of classifying each sample by a behavior pattern recognition algorithm, usually derived through a machine learning model or rule engine.

T _{_i} : The true label of the i-th sample. This is the actual activity type label for each sample, usually provided by manual annotation or known standard data.

I(C _{_i} = T_{_i} ): Indicator function. This is a binary function that returns 1 if the condition is true, and 0 otherwise. It is used to determine whether the classification result is correct.

w _{_act,i} : The activity weight of the i-th sample. Different activity types have different health importance and are therefore assigned different weights. These weights can be determined through expert knowledge or historical data analysis.

M: Number of environmental parameters. This refers to the number of environmental factors considered, such as temperature, humidity, and air quality, which are usually collected in real time by environmental sensors.

w_{_j} : The weight of the j-th environmental parameter. Different environmental parameters have different degrees of impact on health, and therefore are assigned different weights. These weights can be determined through expert knowledge or historical data analysis.

E_{_j} : The influence value of the j-th environmental parameter. This is the specific measured value of each environmental parameter, such as a temperature of 22℃ and a humidity of 60%.

α_{_j} : The exponential factor for the j-th environmental parameter. Used to adjust the influence intensity of each environmental parameter, so that certain parameters (such as air quality) have a greater impact under specific conditions. The exponential factor can be set based on experimental data or expert opinions.

σ: Sigmoid function. Used to map the result of a linear combination to the interval [0, 1], representing the probability of health risk. The Sigmoid function ensures that the output value is within a reasonable range, making it easy to interpret and apply.

W: The weight vector of the context-aware model. These are the parameters of the pre-trained context-aware model, typically learned from a large amount of historical data using machine learning algorithms.

x: Input feature vector. This is a feature vector containing information about the user's current state, such as vital signs data, activity type classification results, etc.

b: Bias term. This is a constant term in the context-aware model used to adjust the model's baseline predictions.

λ: Adjustment coefficient. Used to adjust the impact of the environmental adaptability score (E _score) on health risk prediction. This coefficient can be adjusted according to actual circumstances to balance the importance of environmental factors and other characteristics.

log(E _score ): The logarithmic value of the environmental adaptability score. The logarithmic form is used to smooth out the influence of environmental parameters and prevent extreme values from having an excessive impact on the prediction results.

P _threshold : A preset risk threshold. This is a benchmark value that distinguishes between high and low demand states, and is usually set based on clinical experience and historical data.

δ: Time sensitivity coefficient. Used to adjust for the increase in risk over time, reflecting the impact of time on health risks. This coefficient can be set according to the user's health condition and lifestyle.

Δt: The time interval since the last assessment. This refers to the length of time since the last assessment, usually in hours or days.

Here is a specific example:

Imagine a home care scenario where an elderly patient with hypertension is wearing a multi-functional health monitoring bracelet. The bracelet continuously monitors vital signs such as heart rate, blood pressure, and respiratory rate, and records environmental parameters such as temperature and humidity using environmental sensors. The specific implementation steps are as follows:

The system first uses a behavior pattern recognition algorithm to classify the user's activity type based on information reflecting the user's current state (such as accelerometer and gyroscope data). For example, over a period of time, the system collects N=100 sample data points, which include classification results and real labels for activity types such as resting, walking, and running.

The weighted activity type classification accuracy A _{_acc} is calculated using the following formula:

Suppose that _Ci and _Ti represent the classification result and true label of the i-th sample, respectively, I( _Ci = _Ti ) is an indicator function that returns 1 when the classification is correct and 0 otherwise, w _act,i is the activity weight of the i-th sample.

Substitute the numerical values into the calculation:

The weighted activity type classification accuracy rate was 95%, indicating that the behavior pattern recognition algorithm performed well in this time period.

Next, based on the classification results of user activity types and combined with environmental parameters (such as temperature and humidity), the system comprehensively assesses how these factors affect user health, resulting in an environmental adaptability assessment. For example, the indoor temperature is 22℃ and the humidity is 60%.

The environmental adaptability score (E _score) is calculated using the following formula:

Assume M = 2, representing two environmental parameters (temperature and humidity), _w1 = 0.6 and _w2 = 0.4 are the weights of temperature and humidity respectively, _E1 = 22 and _E2 = 60 are the influence values of temperature and humidity respectively, and _α1 = 1.2 and _α2 = 1.5 are the exponential factors of temperature and humidity respectively.

Substitute the numerical values into the calculation:

E _score = 0.6·22 ^1.2 + 0.4·60 ^1.5 ≈ 0.6·74.86 + 0.4·288.44 ≈ 44.92 + 115.38 = 160.30. The system uses a pre-trained context-aware model to predict the user's potential health risk level in the current state and generate a health risk prediction report. Assuming the input feature vector x includes the user's historical health data and current vital sign data, the Sigmoid function is used to map the linear combination result to the interval [0, 1], representing the probability of health risk.

The probability of health risk, P_{_risk} , is calculated using the following formula:

P _risk =σ(W ^T x+b+λ·log(E _score ))

Assume WT x+b = -0.5, which is the output of the context-aware model, and λ = 0.01, where the adjustment coefficient is used to adjust the impact of the environmental adaptability score on health risk prediction.

Substitute the numerical values into the calculation:

P _risk ＝σ(-0.5+0.01·log(160.30))≈σ(-0.5+0.01·2.20)≈σ(-0.478)≈0.38

That is, the probability of health risk is about 38%, indicating that there is a certain health risk in the current state, but it has not yet reached a high risk level;

Based on the health risk prediction report, the system determines the specific requirements for monitoring frequency and accuracy under the current conditions. Assume a preset risk threshold P _threshold = 0.5, a time sensitivity coefficient δ = 0.05, and a time interval Δt = 2 hours since the last assessment.

The specific monitoring requirement _level D is obtained using the following formula:

Substitute the numerical values into the calculation:

P _threshold ·(1+δ·Δt)=0.5·(1+0.05·2)=0.5·1.1=0.55

Since P _risk = 0.38 < 0.55, the system's monitoring requirement level is "low requirement", which means that the monitoring intensity can be appropriately reduced in the current state to save energy.

By applying the above formula in practice, the system can not only accurately identify the user's activity type, but also comprehensively assess the impact of environmental parameters on health status and predict potential health risks through a context-aware model. Ultimately, the system determines the current monitoring need level based on the probability of health risks, achieving rational resource allocation. This refined monitoring strategy helps improve the efficiency of health management and user experience, ensuring that key health information is captured under different circumstances while avoiding unnecessary resource waste.

This application recognizes that with the development of medical technology and the increasing emphasis on health management, accurate and personalized health assessments have become particularly important. Traditional health assessment methods often rely on doctors' experience and limited data, making it difficult to comprehensively consider individual differences and long-term health data. Therefore, a comprehensive assessment system based on a pre-trained health risk prediction model and comparative analysis of similar cases has been developed. This system aims to achieve a comprehensive assessment of a user's current health status through multi-level data analysis. Thus, a new alternative solution is proposed, which includes:

Based on the confirmed abnormal vital signs, and by comparing the user's long-term health data with similar cases, a pre-trained health risk prediction model is used to comprehensively assess the user's current health status, generating a health status assessment result, including:

Using a pre-trained health risk prediction model, based on the confirmed results of abnormal vital signs, a preliminary assessment of the user's current health status is performed to obtain a preliminary health risk score.

Calculate the initial health risk score _{R_initial} using the following formula:

Where R_{_initial} represents the initial health risk score, N represents the number of outlier parameters, w_{_risk,i} represents the risk weight of the i-th outlier parameter, A_{_i} represents the risk score of the i-th outlier parameter, and β _{_i} is the exponential factor of the i-th outlier parameter, used to adjust the influence intensity of each parameter.

Based on the preliminary health risk assessment report and combined with the user's long-term health data, an in-depth analysis of the user's health status is conducted to generate a personalized health score;

The personalized health score _{S_personal} is calculated using the following formula:

Where _{S_personal} represents the personalized health score, α, β, and γ are the weight coefficients of each factor, H represents the past medical history score, L represents the lifestyle score, A represents the activity level score, and _γH , _γL , and _γA are the index factors of each factor, used to adjust the importance of different factors.

Based on the results of personalized health data analysis, similar cases are searched in the medical database to find cases that match the user's situation. Comparative analysis is then conducted to generate a comprehensive assessment and a final health status assessment score.

_{The similarity} score C is calculated using the following formula:

Where C _similarity represents the similarity score, S _user represents the user's personalized health score, S _case represents the health score of the similar case, and σ is the standard deviation, used to adjust the sensitivity of the similarity function;

By using the results of comparative analysis of similar cases, the parameters of the health risk prediction model are adjusted and the model is optimized to better adapt to individual differences among users, resulting in an optimized health risk prediction model.

Based on the optimized health risk prediction model, and taking into account the abnormality confirmation results, long-term health data and similar case information, a comprehensive assessment of the user's current health status is conducted to generate a health status assessment score.

The health assessment score F _final is calculated using the following formula:

Where _Ffinal represents the health status assessment score, _θ1 , _θ2 , and _θ3 are the weight coefficients of each part, representing the degree of influence of the preliminary health risk score, the personalized health score, and the similar case score, respectively, and _δR , _δS , and _δC are the index factors of each part, used to adjust the influence intensity of different factors.

The following is a detailed explanation of each parameter:

N: Number of abnormal parameters. This refers to the number of parameters in the vital signs data that are identified as abnormal, such as heart rate and blood pressure. These parameters are usually obtained through real-time monitoring devices (such as smart bracelets and sensors).

w _{_risk,i} : The risk weight of the i-th outlier parameter. Different parameters have different degrees of influence on health risk, and therefore are assigned different weights. These weights can be determined through expert knowledge or historical data analysis.

A_{_i} : Risk score for the i-th outlier parameter. This risk score is calculated based on the specific value of each outlier parameter, usually obtained through pre-defined scoring rules or machine learning models.

β_{_i} : An exponential factor for the i-th abnormal parameter. Used to adjust the influence strength of each abnormal parameter, so that certain parameters (such as heart problems) have a greater impact under specific circumstances. The exponential factor can be set based on experimental data or expert opinion.

α, β, γ: Weighting coefficients for each factor. Different factors (past medical history, lifestyle habits, activity level) have varying degrees of influence on health status, and therefore are assigned different weights. These weights can be determined through expert knowledge or historical data analysis.

H: Past Medical History Score. This score is calculated based on the user's historical medical records, typically derived from medical records or information provided by the user.

L: Lifestyle Habits Score. This score is calculated based on the user's daily habits (such as diet, exercise, smoking, etc.), and is usually derived from questionnaires or user-input data.

A: Activity Level Score. This score is calculated based on a user's daily activity level (such as steps taken, exercise time, etc.) and is usually obtained through smart bracelets or other monitoring devices.

_γH , _γL , _γA : Exponential factors for each factor. These are used to adjust the importance of different factors, allowing certain factors (such as past medical history) to have a greater impact in specific situations. Exponential factors can be set based on experimental data or expert opinions.

_{S_user} : The user's personalized health score. This is a score of the user's current health status calculated using the formula above.

S _case : Health score of similar cases. This is the health score of cases that match the user's situation and are found in a medical database, usually derived through search algorithms and comparative analysis.

σ: Standard deviation. Used to adjust the sensitivity of the similarity function, ensuring that the similarity assessment is neither too lenient nor too strict. The standard deviation can be adjusted according to the specific circumstances.

_θ1 , _θ2 , _θ3 : Weighting coefficients for each component. Different assessment components (preliminary health risk score, personalized health score, similar case score) have varying degrees of influence on the final health status assessment, and therefore are assigned different weights. These weights can be determined through expert knowledge or historical data analysis.

_δR , _δS , _δC : Index factors for each component. Used to adjust the influence strength of different assessment components, so that certain components (such as the initial health risk score) have a greater impact under specific circumstances. Index factors can be set based on experimental data or expert opinions.

The following explains the rationale behind each sub-item design:

The impact of past medical history on current health status is considered. The weight α emphasizes the importance of past medical history, while the exponential factor _γH flexibly adjusts its degree of influence.

The impact of lifestyle habits on current health status is considered. The weight β emphasizes the importance of lifestyle habits, while the exponential factor _γL flexibly adjusts the degree of their influence.

Consider the impact of activity levels on current health status. The weight γ emphasizes the importance of activity levels, while the exponential factor _γA flexibly adjusts the degree of its impact.

The weighted indexes of each factor are summed to comprehensively consider the contribution of all factors to health status, generating a holistic, personalized health score. This approach ensures that each factor is appropriately reflected in the final score while maintaining the interpretability of the assessment results.

Consider the impact of the preliminary health risk score on current health status. The weight _θ1 emphasizes the importance of the preliminary health risk score, while the exponential factor _δR flexibly adjusts its degree of influence.

Consider the impact of personalized health scores on current health status. The weight _θ² emphasizes the importance of personalized health scores, while the exponential factor _δS flexibly adjusts the degree of their impact.

The impact of similarity scores on current health status is considered. The weight _θ3 emphasizes the importance of similarity scores, while the exponential factor _δC flexibly adjusts the degree of its influence.

The weighted indexes of each component are summed to comprehensively consider the contribution of all assessment components to the health status, generating a holistic health status assessment score. This approach ensures that each assessment component is appropriately reflected in the final assessment result, while maintaining the interpretability and accuracy of the results.

The overall design of this formula aims to achieve a comprehensive assessment of a user's current health status. By introducing a multi-dimensional assessment method, the system can comprehensively consider the user's current status, long-term health data, and similar case information to generate personalized health status assessment results. Specifically, through a comprehensive assessment of three dimensions—preliminary health risk score, personalized health score, and similar case score—the system can comprehensively capture the user's health status, ensuring the comprehensiveness and accuracy of the assessment results. The introduction of weighting coefficients and index factors allows the system to flexibly adjust the importance of different factors and assessment components, ensuring that the assessment results meet the needs of different users. The health risk prediction model is optimized using the results of similar case comparative analysis, enabling the system to continuously improve and adapt to individual user differences, thereby increasing the accuracy of the assessment. Through comprehensive assessment, the system can dynamically adjust monitoring strategies, ensuring the effectiveness of health monitoring while rationally allocating resources and avoiding unnecessary waste.

This approach not only improves the efficiency and accuracy of health management but also enhances the user experience, ensuring that key health information is captured in different situations and helping users better manage their health.

Here is a specific example:

Imagine a home care scenario where an elderly patient with hypertension is wearing a multi-functional health monitoring bracelet. The bracelet continuously monitors vital signs such as heart rate, blood pressure, and respiratory rate, and records environmental parameters such as temperature and humidity using environmental sensors. The specific implementation steps are as follows:

The system first uses a pre-trained health risk prediction model to conduct a preliminary assessment of the user's current health status based on the confirmation results of abnormal vital signs, and obtains a preliminary health risk score.

_{The initial} health risk score R is calculated using the following formula:

Assume N = 3, representing three abnormal parameters (heart rate, blood pressure, and blood oxygen saturation), w _{_risk,1} = 0.4, w_{_risk,2} = 0.3, and w_{_risk,3} = 0.3, which are the risk weights of each abnormal parameter, A _₁ = 8, A _₂ = 7, and A_₃ = 6, which are the risk scores of each abnormal parameter, and β _₁ = 1.2, β_₂ = 1.1, and β _₃ = 1.0, which are the exponential factors of each abnormal parameter.

Substitute the numerical values into the calculation:

Next, the system combines the user's long-term health data (such as past medical history, lifestyle habits, and activity level) to conduct in-depth analysis of the user's health status and generate a personalized health score.

The personalized health score S_{_personal} is calculated using the following formula:

Assume that α = 0.4, β = 0.3, γ = 0.3 are the weight coefficients of each factor, H = 7, L = 6, A = 8 are the past medical history score, lifestyle habit score, and activity level score, respectively, and γ _H = 1.2, γ _L = 1.1, γ _A = 1.0 are the exponential factors of each factor.

Substitute the numerical values into the calculation:

S _personal ＝0.4·7 ^1.2 +0.3·6 ^1.1 +0.3·8 ^1.0 ≈0.4·12.58+0.3·6.6+0.3·8

≈5.03 + 1.98 + 2.4 ≈ 9.41

Based on personalized health data analysis results, the system searches for similar cases in the medical database, finds cases that match the user's situation, conducts comparative analysis, and generates similar case scores.

_{The similarity} score C is calculated using the following formula:

Assume that _{S_user} = 9.41, the user's personalized health score, _{S_case} = 9.3, the health scores of similar cases, and σ = 0.5, the standard deviation used to adjust the sensitivity of the similarity function;

Substitute the numerical values into the calculation:

Finally, based on the optimized health risk prediction model, the system comprehensively assesses the user's current health status by taking into account the abnormality confirmation results, long-term health data, and similar case information, and generates a health status assessment score.

The health assessment score _Ffinal is calculated using the following formula:

Assume that _θ1 = 0.4, _θ2 = 0.3, _θ3 = 0.3 are the weight coefficients of each part, and _δR = 1.2, _δS = 1.1, _δC = 1.0 are the exponential factors of each part;

Substitute the numerical values into the calculation:

F _final ＝0.4·9.17 ^1.2 +0.3·9.41 ^1.1 +0.3·0.976 ^1.0

≈0.4·13.25+0.3·10.24+0.3·0.976

≈5.3 + 3.07 + 0.29 ≈ 8.66

By applying the above formula in practice, the system can not only accurately assess the impact of abnormal vital signs on the current health status, but also conduct in-depth analysis by combining the user's long-term health data and similar cases to generate a personalized health score. Ultimately, by comprehensively considering all factors, the system generated a health status assessment score F _final = 8.66, indicating that the user's current health status has some risk, but is generally still within a controllable range.

This refined assessment method helps doctors and family members better understand a patient's health condition, enabling them to take timely and effective interventions. For example, the system recommends that seniors pay attention to controlling their blood pressure, maintaining healthy lifestyle habits, and undergoing regular health checkups. Furthermore, the system regularly sends health reports to seniors and their families to help them better manage their health. In this way, the system achieves comprehensive assessment and personalized management of users' health conditions, significantly improving the quality of telemedicine services and the user experience.

Figure 2 is a schematic diagram of a multifunctional vital signs monitoring system provided in an embodiment of this application. As shown in Figure 2, the system includes:

The acquisition module 21 is used to acquire real-time vital signs signals of users under different physiological states, and combine them with environmental parameters to form a multi-dimensional monitoring dataset.

Adjustment module 22 is used to dynamically adjust the monitoring frequency and accuracy based on the multi-dimensional monitoring dataset using an intelligent control algorithm, generate a data collection strategy, minimize energy consumption, and obtain vital sign data.

Analysis module 23 is used to analyze and process the monitored abnormalities in vital signs based on the vital sign data using an instant assessment engine, and generate health status assessment results and personalized health recommendations by combining information such as gender, height, weight, genetic history and past medical history.

The synchronization module 24 is used to securely synchronize the health status assessment results and personalized suggestions to the user's medical monitoring system, supporting remote medical services and realizing continuous health management.

The multifunctional vital signs monitoring system shown in Figure 2 can execute the multifunctional vital signs monitoring method described in the embodiment shown in Figure 1. Its implementation principle and technical effects will not be elaborated further. The specific methods by which each module and unit of the multifunctional vital signs monitoring system in the above embodiments perform operations have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

In one possible design, the multifunctional vital signs monitoring system of the embodiment shown in FIG2 can be implemented as a computing device, as shown in FIG3, which may include a storage component 31 and a processing component 32.

The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

The processing component 32 is used in the multifunctional vital sign monitoring method of the embodiment described in FIG1 above.

The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

Of course, computing devices may also include other components, such as input/output interfaces, display components, communication components, etc.

Input/output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can implement the multifunctional vital sign monitoring method shown in Figure 1 above.

Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims (8)

1. A method of multifunctional vital sign monitoring, comprising:

Acquiring real-time vital sign signals of a user in different physiological states, and combining environmental parameters to form a multi-dimensional monitoring data set;

according to the multi-dimensional monitoring data set, an intelligent regulation algorithm is applied to dynamically adjust the monitoring frequency and accuracy of the real-time vital sign signals, a data collection strategy is generated, and vital sign data are collected based on the data collection strategy;

Based on the vital sign data, analyzing and processing the monitored abnormal condition of the vital sign by utilizing an instant evaluation engine, and generating a health condition evaluation result and personalized health advice by combining the information of gender, height, weight, genetic medical history and past medical history;

the health condition assessment result and the personalized advice are safely synchronized to a medical monitoring system of the user so as to realize remote medical service and continuous health management;

the method for dynamically adjusting the monitoring frequency and accuracy of the real-time vital sign signals by using an intelligent regulation algorithm according to the multi-dimensional monitoring data set, generating a data collection strategy, and collecting vital sign data based on the data collection strategy comprises the following steps:

Identifying the behavior mode, the activity level and the change condition of the environment of the user by utilizing a data analysis technology and the multi-dimensional monitoring data set to obtain information reflecting the current state of the user;

according to the information reflecting the current state of the user, the requirements of the current state on the monitoring frequency and the accuracy are evaluated, and a preliminary monitoring requirement evaluation result is generated;

Combining historical vital sign data, formulating an individualized monitoring frequency and accuracy strategy, and optimizing the preliminary monitoring demand evaluation result to obtain an optimized monitoring strategy;

dynamically controlling the monitoring frequency and the accuracy of the vital sign signal acquisition equipment by using the optimized monitoring strategy to obtain vital sign data;

according to the information reflecting the current state of the user, the requirement of the current state on the monitoring frequency and the accuracy is evaluated, and a preliminary monitoring requirement evaluation result is generated, which comprises the following steps:

classifying the activity types of the user based on the information reflecting the current state of the user by using a behavior pattern recognition algorithm to obtain classification results of the activity types of the user;

calculating the weighted activity type classification accuracy by the following formula :

;

Wherein, the Representing the weighted activity type classification accuracy,Representing the total number of samples,Represent the firstThe result of the classification of the individual samples,Represent the firstThe true labels of the individual samples are then displayed,Is an indication function, returns 1 when the condition is true, otherwise returnsRepresent the firstActivity weights for the individual samples to emphasize the importance of different activity types;

According to the classification result of the user activity type and by combining with the environmental parameters, comprehensively evaluating how to influence the health condition of the user, and obtaining an environmental adaptability evaluation result;

Calculating an environmental suitability score by the following formula :

;

Wherein, the The environmental suitability score is represented by a score,Representing the number of environmental parameters that are to be included,Represent the firstThe weight of the individual environmental parameter(s),Represent the firstThe influence values of the individual environmental parameters,Is the firstAn exponential factor for each environmental parameter for adjusting the impact strength of each parameter;

Based on an environmental adaptability evaluation result, a pre-trained context awareness model is applied to predict potential health risk levels of the user in the current state, and a health risk prediction report is generated;

calculating health risk probability by the following formula :

;

Wherein, the The probability of health risk is indicated,Is a Sigmoid function for mapping the linearly combined result toWithin the interval, representing the likelihood of health risk,Is a weight vector of the context aware model,Is an input feature vector that is used to determine the feature vector,Is a bias term that is used to determine,Is an adjustment factor for adjusting the impact of the environmental suitability score on the health risk prediction;

according to the health risk prediction report, determining the specific requirements on the monitoring frequency and accuracy in the current state, and under the condition of high risk, more frequent and more accurate data acquisition is required;

the specific monitoring demand level is obtained by the following formula :

;

Wherein, the Indicating the level of the monitored demand,Is a preset risk threshold, for distinguishing between high demand and low demand conditions,Is the time sensitivity coefficient of the sample,Is the time interval since the last evaluation for taking into account the increased risk over time;

and integrating the specific monitoring demand level to generate a preliminary monitoring demand evaluation result.

2. The method of claim 1, wherein the combining historical vital sign data to formulate a personalized monitoring frequency and accuracy policy optimizes the preliminary monitoring demand assessment results to obtain an optimized monitoring policy comprises:

carrying out refinement treatment on the monitoring demands of the user in different environments by utilizing the preliminary monitoring demand assessment result to generate specific monitoring demand description;

According to the specific monitoring requirement specification, integrating a long-term behavior mode, a health condition change trend and vital sign data of a user, and constructing a historical model of user behavior and health;

Based on the historical model of the user behavior and the health, a machine learning algorithm is applied to construct a personalized prediction model suitable for the user, wherein the personalized prediction model can predict a future behavior mode of the user and influence on the health;

And formulating monitoring frequency and precision strategies aiming at different environments by utilizing the personalized prediction model and the specific monitoring requirement description, and generating an optimized monitoring strategy.

3. The method of claim 1, wherein analyzing the monitored vital sign anomalies using the instant assessment engine based on the vital sign data, and generating health assessment results and personalized health advice in combination with information on gender, height, weight, genetic medical history, and past medical history, comprises:

The vital sign data is analyzed and processed in real time by utilizing an instant evaluation engine to obtain a vital sign abnormality detection report;

Determining the vital sign abnormality detection report based on a fuzzy logic reasoning mechanism, and generating a vital sign abnormality confirmation result;

based on the vital sign abnormality confirmation result, the current health condition of the user is evaluated by comparing long-term health data of the user with similar cases and using a pre-trained health risk prediction model, and a health condition evaluation result is generated;

And providing customized health advice for the user by using the health condition assessment result and combining the gender, the height and the weight of the user, the genetic medical history, the living habit, the activity level and the prior medical history, and generating personalized health advice.

4. The method of claim 3, wherein the determining the vital sign abnormality detection report based on a fuzzy logic inference mechanism, generating a vital sign abnormality confirmation result, comprises:

analyzing abnormal conditions in the vital sign abnormal detection report by utilizing a fuzzy logic reasoning mechanism to obtain a preliminary abnormal parameter list;

extracting key vital sign parameters corresponding to the preliminary abnormal parameter list, comparing the key vital sign parameters with a preset standard range, identifying parameter values exceeding a normal range, and generating an abnormal parameter set, wherein the key vital sign parameters comprise heart rate, blood pressure and respiratory rate;

based on the abnormal parameter set exceeding the normal range, combining a fuzzy logic rule base, carrying out weight assignment on each abnormal parameter to generate a weighted abnormal parameter set;

and comprehensively evaluating the weighted abnormal parameter set by using a fuzzy inference algorithm, calculating the score of each abnormal parameter by using a fuzzy membership function and an inference rule, and combining the whole health condition to obtain a vital sign abnormal confirmation result.

5. The method of claim 3, wherein the generating a health assessment result based on the vital sign anomaly confirmation result by comprehensively assessing the current health condition of the user using a pre-trained health risk prediction model by comparing long-term health data of the user with similar cases, comprises:

Performing preliminary evaluation processing on the current health condition of the user based on the vital sign abnormality confirmation result by using a pre-trained health risk prediction model to obtain a preliminary health risk evaluation report;

According to the preliminary health risk assessment report, carrying out deep analysis on the health condition of the user by combining with long-term health data of the user to generate a personalized health data analysis result, wherein the long-term health data comprises past medical history, life habits and activity level;

Based on the personalized health data analysis result, searching similar cases in a medical database for comparison analysis processing, and generating a similar case comparison analysis result;

Adjusting parameters of the health risk prediction model by using the analysis results of the similar case comparison, and optimizing the health risk prediction model to obtain an optimized health risk prediction model;

And according to the optimized health risk prediction model, comprehensively considering the abnormal confirmation result, the long-term health data and the similar case information, evaluating the current health condition of the user, and generating a health condition evaluation result.

6. A multifunctional vital sign monitoring system for performing a multifunctional vital sign monitoring method according to any of claims 1-5, comprising:

The acquisition module is used for acquiring real-time vital sign signals of the user in different physiological states and combining environmental parameters to form a multi-dimensional monitoring data set;

the adjusting module is used for dynamically adjusting the monitoring frequency and the accuracy by applying an intelligent regulation algorithm according to the multi-dimensional monitoring data set to generate a data collection strategy, and simultaneously, minimizing energy consumption to obtain vital sign data;

The analysis module is used for analyzing and processing the monitored abnormal condition of the vital sign by utilizing the instant evaluation engine based on the vital sign data, and generating a health condition evaluation result and personalized health advice by combining the information of gender, height, weight, inheritance and past medical history;

And the synchronization module is used for safely synchronizing the health condition evaluation result and the personalized suggestion to a medical monitoring system of the user, supporting remote medical service and realizing continuous health management.

7. A computing device, comprising a processing component and a storage component, the storage component storing one or more computer instructions for execution by the processing component, the one or more computer instructions for implementing a multi-functional vital sign monitoring method according to any of claims 1-5.

8. A computer storage medium, wherein a computer program is stored, which, when executed by a computer, implements a multi-functional vital sign monitoring method according to any of claims 1-5.