Vital sign monitoring wrist strap, monitoring system and method

Abstract

The invention discloses a vital sign monitoring wrist strap, a monitoring system and a monitoring method, and particularly relates to the technical field of vital sign monitoring, which are realized through the following steps: monitoring a plurality of vital sign parameters in real time through a built-in sensor of the wrist strap, uploading the parameters to a cloud end, and generating a data set; based on the data, performing short-term fluctuation and long-term trend evaluation on each parameter to obtain an evaluation result; inputting an evaluation result into the trained machine learning model, outputting a fluctuation risk type and generating an initial classification data set; according to the initial data set and the cross relevance feedback of each parameter, carrying out fluctuation state hierarchical division, and adjusting the sampling frequency of each parameter according to a preset rule to realize efficient and intelligent health monitoring; through a dynamic self-adaptive sampling mechanism, the sampling frequency is intelligently adjusted according to the fluctuation condition of the vital sign parameters and the health risk type, the data real-time performance is ensured, meanwhile, the energy consumption of data collection is effectively reduced, and the service time of equipment is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of vital sign monitoring. More specifically, the present invention relates to a vital sign monitoring wristband, a monitoring system and a method. Background Art

[0002] With the increasing demand for health management in modern society, more and more people begin to pay attention to their health conditions, especially the monitoring of vital signs. Vital signs (such as heart rate, blood pressure, blood oxygen saturation, body temperature, etc.) are important indicators for evaluating the health status of the human body and can reflect the basic physiological state of the body. Traditional vital sign monitoring mostly relies on hospital equipment or occasional handheld measuring tools, and these devices usually can only provide instant single measurements, making it difficult to conduct long-term and continuous health monitoring.

[0003] In recent years, with the development of smart wearable device technology, vital sign monitoring based on wearable devices has become possible. These smart devices are built-in with various sensors, which can collect multiple vital sign data in real time and upload them to the cloud platform for users and medical professionals to analyze. However, the existing smart monitoring devices still face the following challenges in practical applications:

[0004] The contradiction between real-time performance and energy consumption: Real-time data collection and upload require a high frequency, but this will cause the battery of the device to be quickly depleted. How to reduce energy consumption and extend the battery life of the device while ensuring data real-time performance and accuracy is still an important technical problem.

[0005] Insufficient data analysis: Existing health monitoring devices often lack an efficient data analysis mechanism and are difficult to identify health risks in real time, especially in the case of the interaction of multiple vital sign parameters. The lack of precise analysis methods and systems results in the inability to comprehensively evaluate the health risks of individuals. Therefore, a vital sign monitoring wristband, a monitoring system and a method are proposed herein to solve the above problems. Summary of the Invention

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A vital sign monitoring method, comprising the following steps:

[0008] Real-time monitoring of multiple preset vital sign parameters through the built-in sensors of the monitoring wristband and uploading them to the cloud to obtain a vital sign parameter data set;

[0009] Based on the vital sign parameter data set, short-term fluctuation assessment and long-term trend assessment are respectively performed on each preset vital sign parameter to obtain short-term fluctuation assessment results and long-term trend assessment results;

[0010] Substitute the short-term fluctuation assessment results and the long-term trend assessment results into the pre-trained machine learning model to obtain the fluctuation risk type of each preset vital sign parameter, and summarize the fluctuation risk types to obtain the initial classification data group;

[0011] Based on the initial classification data group and the cross-correlation feedback results of each preset vital sign parameter, hierarchically divide the fluctuation state of each preset vital sign parameter, and adjust the sampling frequency of each preset vital sign parameter according to the preset sampling optimization rule.

[0012] In a preferred embodiment, the short-term fluctuation assessment refers to:

[0013] Based on the changes of the preset vital sign parameters within the preset time window, calculate the dynamic fluctuation index to measure its instantaneous fluctuation situation.

[0014] In a preferred embodiment, the long-term trend assessment refers to:

[0015] Based on the historical data of the preset vital sign parameters, calculate the health trend index to evaluate its long-term health status.

[0016] In a preferred embodiment, the machine learning model is a convolutional neural network. Substitute the dynamic fluctuation index and the health trend index into the convolutional neural network model. The convolutional neural network outputs a result of 0 or 1. 0 indicates that the fluctuation risk type corresponding to the vital sign parameter is a low-risk type, and 1 indicates that the fluctuation risk type corresponding to the vital sign parameter is a high-risk type. And sort according to the corresponding numbers of the preset vital sign parameters, and summarize all the output results of the convolutional neural network to obtain the initial classification data group.

[0017] In a preferred embodiment, the acquisition logic of the dynamic fluctuation index is:

[0018] Real-time obtain the data of the preset vital sign parameter X within the preset time window from the sensor. Assume that the data collected within the preset time window is a time series. Use the standard deviation as the measure of fluctuation, and at the same time consider introducing the weighted standard deviation to improve the sensitivity to the recent changes:

[0019] wi represents the weight coefficient corresponding to the i-th numbered data in the time series, N represents the total number of data in the time series, and σ(X) represents the fluctuation measurement value of the time series;

[0020] Calculate the change in the fluctuation amplitude as a supplementary index of the dynamic fluctuation:

[0021] Calculate the difference between the maximum value and the minimum value of the data within the current time window to obtain the fluctuation amplitude value A(X); compare the change in the fluctuation amplitude of the current time window with that of the previous time window to obtain the amplitude change value. The calculation formula is as follows:

[0022] ΔA(X) = A(X) - A(Xprevious); A(Xprevious) represents the fluctuation amplitude value corresponding to the previous time window, and ΔA(X) represents the amplitude change value;

[0023] The calculation formula for the dynamic fluctuation index is as follows:

[0024] ∈ represents a preset non-zero constant, and DVI represents the dynamic fluctuation index.

[0025] In a preferred embodiment, the acquisition logic of the weight coefficient corresponding to the i-th numbered data in the time series is as follows:

[0026] Based on the exponential decay factor, it is dynamically adjusted according to the fluctuation rate of the current vital sign parameter. The formula is as follows:

[0027] represents the change rate of the current vital sign parameter sampling data at time ti, β is a preset non-zero influence coefficient for controlling the attenuation of the fluctuation rate, t0 represents the starting point of time of the time series X, ti represents the time point of the current vital sign parameter sampling data in the time series X, and λ is a preset attenuation rate constant.

[0028] In a preferred embodiment, the acquisition logic of the health trend index is as follows:

[0029] Real-time obtain the historical data of the preset vital sign parameter X under the preset historical window from the sensor, and extract the long-term change trend through regression analysis:

[0030] X(j) = a + b·Tj + ε; Tj represents the sampling data corresponding to the time point j of the vital sign parameter X under the historical window, both a and b are regression coefficients, representing the starting value and the change rate of the vital sign parameter X under the historical window, ε is the regression residual, and X(Tj) represents the regression prediction value corresponding to the time point j of the vital sign parameter X under the historical window. Define the trend strength coefficient TSI to measure the change trend of the health status, TSI = |b|. At the same time, introduce the root mean square error between the sampling value and the regression prediction value of the vital sign parameter X under the historical window to obtain the long-term stability coefficient LTSI reflecting the long-term stability of the health parameter;

[0031] The calculation formula for the health trend index is as follows:

[0032] LTSImax is a preset standard value of the long-term stability coefficient, which is used to normalize the long-term stability coefficient LTSI. The normalized long-term stability coefficient LTSI is within the range of [0, 1]. Both f1 and f2 are preset non-zero adjustment coefficients, and the sum of f1 and f2 is one. HTI represents the health trend index.

[0033] In a preferred embodiment, hierarchically dividing the fluctuation state of each preset vital sign parameter based on the initial classification data group and the cross-correlation feedback results of each preset vital sign parameter means:

[0034] Using fuzzy inference, obtaining the result of the preset vital sign parameter X in the initial classification data group, the cross-correlation value between the preset vital sign parameter X and other vital sign parameters calculated by the Pearson correlation coefficient calculation formula, and the preset correlation threshold, all of which are used as input variables of fuzzy logic. Taking the level corresponding to the fluctuation state of the vital sign parameter X as the output variable of fuzzy logic, performing fuzzy processing on the input variables, converting the values of the input variables into fuzzy sets, performing fuzzy processing on the output variables, converting the output variables into fuzzy sets, formulating fuzzy rules to describe the adaptation degree of each level under different combinations of data types, and inferring the fuzzy input variables through the fuzzy rules to obtain the level corresponding to the fluctuation state of the vital sign parameter X.

[0035] In a preferred embodiment, a vital sign monitoring wristband includes:

[0036] An in-built sensor module for real-time monitoring of multiple preset vital sign parameters;

[0037] A data processing unit for receiving and processing the vital sign parameter data collected by the in-built sensor module and uploading the data to the cloud data processing platform to generate a vital sign parameter data set;

[0038] A short-term fluctuation assessment module for respectively performing short-term fluctuation assessment on each preset vital sign parameter based on the vital sign parameter data set to obtain the short-term fluctuation assessment result of each parameter;

[0039] A long-term trend assessment module for respectively performing long-term trend assessment on each preset vital sign parameter based on the vital sign parameter data set to obtain the long-term trend assessment result of each parameter;

[0040] A machine learning decision module for inputting the short-term fluctuation assessment result and the long-term trend assessment result into a pre-trained machine learning model, outputting the fluctuation risk type of each preset vital sign parameter, and generating an initial classification data group;

[0041] Cross - correlation feedback module, which analyzes the cross - correlation between each preset vital sign parameter;

[0042] Sampling frequency optimization module, which hierarchically divides the fluctuation state of each preset vital sign parameter according to the preliminary classification data set and cross - correlation feedback information, and adjusts the sampling frequency of each vital sign parameter according to the preset sampling optimization rules.

[0043] In a preferred embodiment, a vital sign monitoring system includes:

[0044] Vital sign monitoring terminal, which is built - in with a sensor module for real - time monitoring of multiple preset vital sign parameters, and a data processing unit for receiving and processing the vital sign parameter data collected by the built - in sensor module, and uploading the data to the cloud data processing platform to generate a vital sign parameter data set;

[0045] Cloud data processing platform, which is used to receive the vital sign data set uploaded from the vital sign monitoring wristband and perform the following steps:

[0046] Conduct a short - term fluctuation assessment on each vital sign parameter to obtain a short - term fluctuation assessment result;

[0047] Conduct a long - term trend assessment on each vital sign parameter to obtain a long - term trend assessment result;

[0048] Input the assessment results into a pre - trained machine learning model, output the fluctuation risk type of each vital sign parameter, and hierarchically divide the fluctuation states of each parameter according to the cross - correlation between different parameters;

[0049] Optimize the sampling frequency of each preset vital sign parameter according to the results of the hierarchical division.

[0050] The technical effects and advantages of the present invention:

[0051] Through a dynamic adaptive sampling mechanism, the present invention intelligently adjusts the sampling frequency according to the fluctuation of vital sign parameters and the type of health risk. While ensuring data real - time, it can effectively reduce the energy consumption of data collection and transmission, thus significantly extending the usage time of the device and solving the contradiction between high - frequency data collection and device battery life. The present invention combines short - term fluctuation assessment and long - term trend assessment, and comprehensively analyzes the fluctuation of vital sign parameters by calculating the dynamic fluctuation index and health trend index. The short - term fluctuation assessment can timely capture sudden health changes, while the long - term trend assessment helps to discover the gradual health change trend and early warning of potential health risks. Through the application of the machine learning model, the present invention can accurately judge the fluctuation risk type of vital signs, helping to achieve efficient health monitoring and early warning.

[0052] The present invention can perform personalized health management based on an individual's historical health data and real-time monitoring results. By introducing fuzzy inference and cross-correlation analysis, the system can dynamically adjust health assessment and intervention strategies. For example, for an individual's health characteristics, the system will adjust the monitoring frequency of each vital sign parameter to ensure timely monitoring of high-risk parameters while saving unnecessary resource consumption. The present invention further optimizes health risk assessment by calculating the Pearson correlation coefficient between multiple vital sign parameters and analyzing their cross-correlation. The correlation between different parameters can provide deeper information for health warnings, avoid ignoring the potential impact between certain parameters, and improve the accuracy and comprehensiveness of health monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with the accompanying drawings;

[0054] Figure 1 It is a schematic diagram of a vital sign monitoring method in the present invention.

[0055] Figure 2 It is a schematic diagram of a vital sign monitoring wristband in the present invention.

[0056] Figure 3 It is a schematic diagram of a vital sign monitoring system in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative work shall fall within the protection scope of the present invention.

[0058] Refer to Figure 1 - Figure 3 The following embodiments are obtained:

[0059] Embodiment 1:

[0060] With the continuous development of technology and the improvement of people's health awareness, vital sign monitoring technology has become an important part in the fields of daily health management, disease prevention, and elderly care. Vital signs such as heart rate, blood pressure, blood oxygen saturation, body temperature, etc. reflect the basic health status of the human body and are important indicators for judging disease risk and health status. Traditional vital sign monitoring mostly relies on hospital equipment or relatively simple handheld measuring instruments, which usually have disadvantages such as low usage frequency, long measurement cycle, and poor immediacy.

[0061] In recent years, with the popularization of smart wearable devices, the monitoring of vital signs based on devices such as smart bracelets and watches has gradually emerged. These devices are usually equipped with sensors that can collect vital sign data in real time and upload it to the cloud, facilitating long-term health management for users. However, there are still some problems with the current monitoring technology in the following aspects:

[0062] The contradiction between real-time performance and energy consumption: High-frequency data collection and transmission consume a large amount of electrical energy, resulting in a shortened battery life of the device. Reducing the sampling frequency may affect the real-time performance of the data, reducing the accuracy and response speed of the monitoring.

[0063] Insufficient data analysis and processing: Although modern wearable devices can collect vital sign data in real time, how to efficiently analyze this data and adjust the sampling frequency in a timely manner according to its fluctuations and identify abnormal fluctuations is still a difficult point. Most current monitoring methods rely on single-parameter evaluation, lacking cross-correlation analysis of multiple vital sign parameters and a dynamic adaptive adjustment mechanism.

[0064] Lack of personalization in fluctuation risk assessment: Most monitoring devices cannot dynamically adjust the sampling strategy according to the individual differences of users and only sample according to fixed rules. This makes it impossible for the device to accurately judge the fluctuation risk according to the user's health status in some specific scenarios.

[0065] Problems with data fusion and optimization: Traditional vital sign monitoring systems often ignore the internal correlation between vital signs and fail to effectively utilize the cross-information between various vital sign parameters, resulting in low monitoring efficiency and inability to achieve intelligent sampling frequency adjustment and health assessment.

[0066] The present invention proposes a vital sign monitoring method, aiming to solve the contradictions between real-time performance and energy consumption, insufficient data analysis and processing, lack of personalized fluctuation risk assessment, and data fusion and optimization in the prior art. Through an innovative multi-level adaptive sampling mechanism, combining short-term fluctuation assessment and long-term trend assessment, introducing a machine learning model to classify and evaluate the fluctuation risk, and dividing the fluctuation state levels through cross-correlation feedback, the present invention dynamically adjusts the sampling frequency, reduces the energy consumption of the device while ensuring data real-time performance, and improves the battery life of the device. According to the individual's vital sign fluctuations and long-term trends, combined with the user's historical health data, a personalized fluctuation risk assessment is carried out to ensure that the monitoring results are more targeted and accurate. By analyzing the cross-correlation between multiple vital sign parameters, comprehensively evaluating the fluctuation states of each parameter, providing more basis for dynamically adjusting the sampling frequency, and optimizing the monitoring effect. Based on the classification of the vital sign fluctuation risk types by the machine learning model, combined with the cross-correlation feedback mechanism, different fluctuation states are divided into levels, and the sampling frequency is intelligently adjusted according to the level changes, so as to realize the dynamic optimization of the adaptive sampling strategy.

[0067] The present invention proposes a vital sign monitoring method, comprising the following steps:

[0068] Real-time monitor multiple preset vital sign parameters through the built-in sensors of the monitoring wristband and upload them to the cloud to obtain a vital sign parameter data set; this step is the basis of the entire vital sign monitoring process. The built-in sensors of the wristband generate a continuous data stream by measuring multiple vital sign parameters (such as heart rate, blood pressure, blood oxygen, etc.) in real time. Uploading the data to the cloud enables remote monitoring and data storage, so that the user's health data can be stored for a long time and analyzed more comprehensively. Real-time monitoring can capture the immediate changes in the health status, and uploading to the cloud ensures that the data is not lost and can be accessed or analyzed at any time.

[0069] Based on the vital sign parameter data set, conduct short-term fluctuation assessment and long-term trend assessment on each preset vital sign parameter respectively to obtain short-term fluctuation assessment results and long-term trend assessment results; Short-term fluctuation assessment: Aims to evaluate the changes in vital signs in a short period of time, such as rapid changes in heart rate, short-term fluctuations in blood pressure, etc. These changes may reflect abnormalities in the instantaneous health status or responses to external stimuli. Long-term trend assessment: By analyzing the vital sign data over a period of time, evaluate its long-term trend, such as whether the heart rate continues to increase, whether the blood sugar has an upward trend, etc. This helps to judge the individual's health change trend and predict potential health risks. The combination of these two assessment results can more comprehensively describe the individual's health status, revealing both instantaneous fluctuations and providing a perspective on long-term health changes.

[0070] Substitute the short-term fluctuation assessment results and the long-term trend assessment results into the pre-trained machine learning model to obtain the fluctuation risk types of each preset vital sign parameter, and summarize the fluctuation risk types to obtain the initial classification data group; this step uses the machine learning model to comprehensively analyze the results of short-term fluctuation assessment and long-term trend assessment. The machine learning model (such as a convolutional neural network) can understand the relationship between different fluctuations and health risks through the pre-trained data. Through the output of the model, each vital sign parameter is classified as high risk or low risk, providing a reference for subsequent health interventions and decisions. This step can automatically identify and classify the health risk types of each parameter, thus reducing the need for manual intervention and improving the judgment accuracy.

[0071] Based on the initial classification data group and the cross-correlation feedback results of each preset vital sign parameter, perform a hierarchical division of the fluctuation state of each preset vital sign parameter, and adjust the sampling frequency of each preset vital sign parameter according to the preset sampling optimization rule. In this step, the system not only relies on a single vital sign data but also considers the cross-correlation between multiple vital sign parameters. For example, there may be a correlation between heart rate and blood oxygen, and the fluctuations of blood pressure and body temperature may also affect each other. The cross-correlation feedback evaluates the comprehensive health status by calculating the mutual relationship between each vital sign parameter, making the risk assessment more comprehensive and accurate. Through this hierarchical division, the system can refine the grading of the health status and further optimize the risk management strategy. This makes the monitoring more flexible and allows for different levels of intervention on the parameter fluctuations according to the specific fluctuation state. Based on the previous assessment results and hierarchical division, the system can dynamically adjust the sampling frequency. For example, if a certain vital sign parameter is evaluated as a high-risk fluctuation state, the system will increase the sampling frequency of this parameter to more finely monitor its changes; while for parameters with smaller fluctuations, the system can reduce the sampling frequency, thereby reducing energy consumption. This adaptive sampling mechanism can optimize the energy consumption of the device, extend the battery life, and ensure the efficiency of health monitoring while ensuring the real-time nature of the data. This step is the core of the intelligence of the entire monitoring system, which can intelligently adjust the strategy according to the data feedback and avoid the resource waste caused by the traditional fixed sampling frequency.

[0072] Short-term fluctuation assessment refers to: calculating a dynamic fluctuation index based on the changes of preset vital sign parameters within a preset time window to measure its instantaneous fluctuation. Long-term trend assessment refers to: calculating a health trend index based on the historical data of preset vital sign parameters to evaluate its long-term health status. The short-term fluctuation assessment aims to capture the instantaneous fluctuations of vital sign parameters within a short time window. This process calculates the dynamic fluctuation index to quantify the change range of parameters (such as heart rate, blood pressure, blood oxygen, etc.). This assessment helps to identify immediate health abnormalities or sudden fluctuations, such as a sudden increase in heart rate, a sudden rise in blood pressure, etc. These fluctuations may be caused by factors such as exercise, emotional fluctuations, external environment, etc., and may even be warning signals of potential health problems. Through this real-time assessment method, the immediate changes in health status can be effectively tracked, providing a basis for subsequent intervention and treatment.

[0073] Based on the historical data of preset vital sign parameters, the long-term trend assessment can help evaluate the health change trend of an individual by calculating the health trend index, revealing the potential long-term changes in the health status. The goal of this process is to identify long-term persistent health problems or improvement trends, such as early signs of chronic diseases, tracking of the health recovery process, etc. The health trend index reflects the long-term stability of vital signs, helping to understand the health performance of an individual over a longer period of time.

[0074] The machine learning model is a convolutional neural network. The dynamic fluctuation index and the health trend index are both input into the convolutional neural network model. The convolutional neural network outputs a result of 0 or 1. 0 indicates that the fluctuation risk type corresponding to the vital sign parameter is a low-risk type, and 1 indicates that the fluctuation risk type corresponding to the vital sign parameter is a high-risk type. And they are sorted according to the corresponding numbers of the preset vital sign parameters. The output results of all convolutional neural networks are summarized to obtain an initial classification data group.

[0075] In the present invention, the convolutional neural network model is based on a hierarchical structure, in which the input data is gradually processed through multiple convolutional layers. The main function of the convolutional layer is to extract local features from the input data, and these features represent certain regularities of the data in space or time. In vital sign monitoring, these convolutional layers can capture the subtle fluctuations of vital sign parameters over time, such as the instantaneous change trends of parameters such as heart rate and blood pressure.

[0076] The convolution operation calculates the weighted sum of a local area by sliding a small window over the input data, thereby generating a new feature map. This process can effectively preserve the local correlations in the data and reduce the computational complexity. As the number of layers increases, the convolutional neural network gradually progresses from simple local feature extraction to more complex pattern recognition, ultimately forming a deep understanding of the entire dataset. In the present invention, these feature maps help the model identify the patterns between the vital sign parameters, thereby evaluating the fluctuation degree of each parameter.

[0077] After the convolutional layer, the pooling layer performs a dimensionality reduction operation on the extracted features, further compressing the data size while retaining the most important feature information. Finally, the fully connected layer combines the features extracted by the convolutional layer and the pooling layer to perform the final classification or regression task. Through this hierarchical learning method, the convolutional neural network can efficiently analyze the short-term fluctuations and long-term trends of the vital sign parameters and ultimately output the risk level of each parameter, helping to achieve more accurate health assessment and monitoring.

[0078] The convolutional neural network takes the dynamic fluctuation index and the health trend index as inputs, analyzes the complex patterns and potential health risks in the vital sign data. Through the trained machine learning model, it can efficiently and automatically classify the risk of each vital sign parameter, outputting a result of 0 or 1, indicating the type of fluctuation risk of the parameter. 0 represents low risk and 1 represents high risk. This binary classification can quickly judge the health status of the vital signs and provide timely feedback. By summarizing the output results of each parameter, an initial classification data group is obtained, which can synthesize the risk types of each vital sign parameter and provide an effective basis for subsequent health interventions, data analysis, and decision-making.

[0079] The acquisition logic of the dynamic fluctuation index is as follows:

[0080] Real-time obtain the data of the preset vital sign parameter X within the preset time window from the sensor. Assume that the data collected within the preset time window is a time series. Use the standard deviation as the measure of fluctuation, and at the same time consider introducing a weighted standard deviation to improve the sensitivity to recent changes:

[0081] wi represents the weight coefficient corresponding to the i-th numbered data in the time series, N represents the total number of data in the time series, and σ(X) represents the fluctuation measure value of the time series;

[0082] Calculate the change in the fluctuation amplitude as a supplementary indicator of dynamic fluctuation:

[0083] Calculate the difference between the maximum and minimum values of the data within the current time window to obtain the fluctuation amplitude value A(X); compare the change in the fluctuation amplitude of the current time window with that of the previous time window to obtain the amplitude change value. The calculation formula is as follows:

[0084] ΔA(X) = A(X) - A(Xprevious); A(Xprevious) represents the fluctuation amplitude value corresponding to the previous time window, and ΔA(X) represents the amplitude change value.

[0085] The calculation formula for the dynamic fluctuation index is as follows:

[0086] ∈ represents a preset non-zero constant, and DVI represents the dynamic fluctuation index. By introducing weights, the model can better focus on recent data changes. Especially in the case of rapid fluctuations, it can immediately reflect the impact of these changes on the health status. By comparing the current and past changes in the fluctuation amplitude, it can better judge whether there are abnormal fluctuations or change trends, thus providing more accurate health warnings. Based on the comprehensive consideration of fluctuation measurement and amplitude change, the dynamic fluctuation index can more comprehensively evaluate health fluctuations and provide more accurate data support for sampling frequency adjustment. The larger the dynamic fluctuation index, the stronger the volatility of the vital sign data. That is to say, the vital sign parameters change greatly in a short time, which may imply greater health risks or instability of the health status. For example, drastic fluctuations in heart rate, rapid changes in blood pressure, etc. may indicate potential health problems and may require further monitoring.

[0087] The acquisition logic of the weight coefficient corresponding to the i-th numbered data in the time series is as follows:

[0088] Based on the exponential decay factor, it is dynamically adjusted according to the fluctuation rate of the current vital sign parameter. The formula is as follows:

[0089] represents the change rate of the current vital sign parameter sampling data at time ti, β is a preset non-zero influence coefficient that controls the attenuation of the fluctuation rate, t0 represents the starting time of the time series X, ti represents the time point of the current vital sign parameter sampling data in the time series X, and λ is a preset decay rate constant. The exponential decay factor makes the influence of earlier data gradually decrease over time. The decay rate constant λ controls the rate of this process. This part shows that as the amplitude of the change in vital sign parameters (such as heart rate, blood pressure, etc.) increases, the weight of the data at that moment is enhanced, which increases the sensitivity to sharp parameter changes. t0 represents the starting time point of the time series and is used to calculate the time difference. Based on the time decay factor, the model pays more attention to the most recent data, thus making a rapid response to the current fluctuations in vital signs. Especially when there are large fluctuations, it is dynamically adjusted according to the rate of change of vital sign parameters. In the case of rapid or violent changes, more weight can be given to more accurately monitor and evaluate the current health status. This formula, by considering the feedback mechanism of the fluctuation rate and the change amplitude, helps to timely capture the abnormal fluctuations of vital signs and provides a more accurate basis for subsequent sampling frequency adjustment and health intervention.

[0090] The acquisition logic of the health trend index is as follows:

[0091] Real-time obtain the historical data of the preset vital sign parameter X in the preset historical window from the sensor, and extract the long-term change trend through regression analysis:

[0092] X(j) = a + b·Tj + ε; Tj represents the sampling data of the vital sign parameter X corresponding to the time point j in the historical window. Both a and b are regression coefficients, representing the starting value and the change rate of the vital sign parameter X in the historical window. ε is the regression residual, and X(Tj) represents the regression prediction value of the vital sign parameter X corresponding to the time point j in the historical window. Define the trend strength coefficient TSI to measure the change trend of the health status, TSI = |b|. At the same time, introduce the root mean square error between the sampling value and the regression prediction value of the vital sign parameter X in the historical window to obtain the long-term stability coefficient LTSI reflecting the long-term stability of the health parameter.

[0093] The calculation formula of the health trend index is:

[0094] LTSImax is the preset standard value of the long-term stability coefficient, used to normalize the long-term stability coefficient LTSI. The normalized long-term stability coefficient LTSI is in the range of [0, 1]. Both f1 and f2 are preset non-zero adjustment coefficients, and the sum of f1 and f2 is one. HTI represents the health trend index.

[0095] The Health Trend Index predicts the future change trend of vital signs through regression analysis. Regression analysis helps to find out the change rules of data, ensuring that the trend of health changes can be effectively extracted from historical data. For example, the starting value, change rate, etc. of vital signs are estimated through regression, providing a mathematical basis for future health prediction, and being able to identify the long-term change rules of health parameters, rather than just instantaneous fluctuations. Through the evaluation of long-term trends, it can help to judge the gradual changes in an individual's health status and timely detect potential chronic diseases or long-term unstable health conditions.

[0096] The Fluctuation Intensity Coefficient (TSI) is introduced into the Health Trend Index, which reflects the frequency and intensity of the changes in vital signs. When the vital signs fluctuate greatly, the health risk may also be higher. Therefore, it is necessary to weight and evaluate the health risk through the fluctuation intensity. This helps the system to more sensitively capture the fluctuations of health parameters and make a rapid response to health changes. The significance of this design lies in ensuring that in the face of large health fluctuations, the monitoring system can respond to these changes in real time and timely adjust the health assessment and sampling strategies.

[0097] By measuring the long-term stability of health parameters, it can effectively reflect the stability and change trend of the health status. The introduction of the Long-Term Stability Coefficient helps to judge whether the health status is in a long-term unstable state and provides a prediction of chronic diseases or potential health problems. It can evaluate the health trend from a long-term perspective and identify chronic health risks or gradually deteriorating health conditions.

[0098] The Health Trend Index comprehensively evaluates the health status by combining the fluctuation intensity and the long-term stability coefficient and using preset weighting coefficients. In this way, while ensuring the accuracy of health assessment, the system can flexibly adjust the weights of different factors and pay more or less attention to different parameters according to the current health status. The weighted synthesis makes the health assessment not just an assessment of a single indicator, but a dynamic and all-round assessment process that can make adaptive adjustments according to different health conditions.

[0099] Regression analysis is usually performed by least squares fitting. Especially in such problems, least squares method is widely used in linear regression analysis. The least squares method fits the data by minimizing the difference between the actual data points and the predicted values of the regression model, and obtains the optimal regression coefficients. Specifically, by adjusting the parameters of the model (such as the starting value and the rate of change) to minimize the sum of the squares of the prediction errors, the trend prediction of the vital sign data is achieved. The larger the health trend index, the more significant the changes in vital signs in the long-term trend, and there may be a sharp change or deterioration in the health condition. For patients with chronic diseases, a drastic change in the long-term health trend may mean the aggravation of the condition or an unstable state. A large value of the health trend index may mean poor long-term stability of the health state and large fluctuations, which usually indicates that the individual's health condition is relatively unstable and requires more monitoring and intervention.

[0100] Based on the initial classification data set and the cross-correlation feedback results of each preset vital sign parameter, the hierarchical division of the fluctuation state of each preset vital sign parameter refers to:

[0101] Using fuzzy inference, obtain the result of the preset vital sign parameter X in the initial classification data set, the cross-correlation value between the preset vital sign parameter X and other vital sign parameters calculated by the Pearson correlation coefficient calculation formula, and the preset correlation threshold, and use them together as the input variables of fuzzy logic. Take the level corresponding to the fluctuation state of the vital sign parameter X as the output variable of fuzzy logic. Perform fuzzy processing on the input variables, convert the values of the input variables into fuzzy sets, perform fuzzy processing on the output variables, convert the output variables into fuzzy sets, formulate fuzzy rules to describe the adaptation degree of each level under different data type combinations, and perform inference on the fuzzy input variables through the fuzzy rules to obtain the level corresponding to the fluctuation state of the vital sign parameter X.

[0102] In health monitoring, cross-correlation refers to the mutual relationship between different vital sign parameters. For example, there may be a certain positive correlation between heart rate and blood pressure, and body temperature and blood oxygen saturation may show a negative correlation. By calculating the Pearson correlation coefficient, the correlation between these parameters can be quantitatively described: the Pearson correlation coefficient is a value between -1 and 1, indicating the degree of linear relationship between two variables. When the correlation coefficient is close to 1, it means there is a strong positive correlation between these two variables; when it is close to -1, it indicates a negative correlation; when it is close to 0, it means there is almost no correlation.

[0103] For example, during health monitoring: there may be a strong negative correlation between heart rate and blood oxygen, that is, an increase in heart rate is usually accompanied by a decrease in blood oxygen. At this time, when the heart rate fluctuates greatly, the changes in blood oxygen should also be given more attention. There may be a certain positive correlation between blood pressure and body temperature. When the body temperature rises, the blood pressure may show an upward trend. In a fuzzy inference system, the role of cross-correlation feedback is reflected in the mutual influence between multiple vital sign parameters. The system can intelligently adjust health assessment and risk prediction based on the correlation of these parameters.

[0104] After the input variables are calculated, they enter the fuzzy logic system. In the fuzzy inference system, the input variables are fuzzified, converting specific numerical values into fuzzy sets. This is because vital sign parameters and their fluctuation states often do not have clear boundaries but show a gradual change property.

[0105] For example: the fluctuation intensity can be fuzzified into several levels, such as "low fluctuation", "medium fluctuation", "high fluctuation". The cross-correlation can also be fuzzified into levels such as "weak correlation", "medium correlation", "strong correlation", etc. This fuzzification method enables the system to more flexibly handle uncertainties or incomplete data, avoiding the limitations brought by overly precise classification. Once the input variables are fuzzified, they will serve as the input for fuzzy logic reasoning. Next, reasoning is performed on different combinations of input variables according to the set fuzzy rules to obtain the output result. These fuzzy rules are formulated based on historical data and empirical rules, describing the fluctuation state levels under different combinations of vital sign parameters.

[0106] For example, suppose there are the following fuzzy rules: If the fluctuation intensity is "high fluctuation" and the cross-correlation is "strong correlation", then the fluctuation state level should be "high risk". If the fluctuation intensity is "low fluctuation" and the cross-correlation is "weak correlation", then the fluctuation state level is "low risk". These rules determine the level of the fluctuation state based on the actual fluctuations of vital sign parameters and their mutual relationships.

[0107] After reasoning through the fuzzy rules, the system will obtain a fuzzy output result, which is usually the fluctuation state level of the vital sign parameters. The fluctuation state level will indicate the current health risk level of the parameter.

[0108] For example, after reasoning, if the system obtains a fluctuation state of "high risk", this means that there may be significant abnormal fluctuations in the current health state and further attention and intervention are required. The fuzzification process of the output variable will convert the inferred level (such as "low risk" or "high risk") into specific numerical values or levels and finally feedback to the user or the health monitoring system.

[0109] In the process of fuzzy reasoning, the Max-Min Method is usually adopted for reasoning. This method is a commonly used reasoning method in fuzzy logic systems, which infers the output result based on the fuzzy sets of input variables and preset fuzzy rules.

[0110] Max Method: For the input of fuzzy rules, in each rule, the minimum input membership degree value is selected. Specifically, under multiple input conditions, taking the minimum value represents the minimum activation degree of the current rule. For example, if the rule is "If the heart rate is high and the blood oxygen is low, then the fluctuation state is a high risk", then for this rule, the fluctuation degree of the heart rate is "high" and the blood oxygen is "low", and the system will take the minimum value of the two as the index of the activation degree.

[0111] Min Method: For the output of the reasoning result, the Min Method is used to determine the membership degree of the output variable. That is to say, among the output results of multiple rules, the minimum value is selected as the final reasoning output. This can ensure that under the influence of multiple rules, the output result of the system will not be too drastic, but is restricted according to the minimum limit of all rules.

[0112] For example, in fuzzy rule 1, the membership degree of the heart rate is 0.8 and the membership degree of the blood pressure is 0.7, and the activation value of rule 1 is 0.7 (taking the minimum value). Similarly, in fuzzy rule 2, the membership degree of the heart rate is 0.8 and the membership degree of the blood oxygen is 0.6, and the activation value of rule 2 is 0.6. Through the Max-Min Method, the final output of the health risk is the minimum value of the two, that is, 0.6.

[0113] Suppose the system is monitoring four vital sign parameters of a patient: heart rate, blood pressure, blood oxygen saturation, and body temperature. These data have been evaluated for short-term fluctuations and long-term trends, and the system classifies them into low-risk and high-risk categories. The risk type of each parameter can be represented by 0 for low risk and 1 for high risk. Initial classification data group: Heart rate: 1 (high risk); Blood pressure: 0 (low risk); Blood oxygen saturation: 1 (high risk); Body temperature: 0 (low risk); This classification data group indicates that the fluctuations of the heart rate and blood oxygen saturation are relatively large, belonging to a situation with a higher health risk, while the blood pressure and body temperature are relatively stable and the risk is lower.

[0114] To further understand the cross-correlation between these parameters, the system calculates the Pearson correlation coefficient between each pair of parameters. The Pearson correlation coefficient is a prior art and its formula will not be elaborated here. To evaluate their linear relationship: Pearson correlation coefficient between heart rate and blood pressure: The assumed value is 0.2, indicating a weak correlation between them. Pearson correlation coefficient between heart rate and blood oxygen saturation: The assumed value is -0.7, indicating a strong negative correlation between them, that is, an increase in heart rate is usually accompanied by a decrease in blood oxygen. Pearson correlation coefficient between blood pressure and body temperature: The assumed value is 0.3, indicating a slight positive correlation between them. Pearson correlation coefficient between blood oxygen saturation and body temperature: The assumed value is -0.5, indicating a negative correlation between them as well.

[0115] Based on the initial classification data set and the Pearson correlation coefficient, the system classifies these parameters into different fluctuation levels and assigns different sampling frequencies to each level. Assume that the level classification of the system is based on the following criteria: High fluctuation state (high-risk level): If the fluctuation of a parameter is evaluated as high risk and has a strong correlation with other parameters, then this parameter is classified into the high fluctuation state. Medium fluctuation state (medium-risk level): If the fluctuation of a parameter is evaluated as medium risk and has a weak correlation with other parameters, then it is classified into the medium fluctuation state. Low fluctuation state (low-risk level): If the fluctuation of a parameter is evaluated as low risk and has a low correlation, then it is classified into the low fluctuation state. Specific level classification: Heart rate: Since the heart rate is evaluated as high risk and has a strong negative correlation (-0.7) with blood oxygen saturation, the system classifies it into the high-risk level. Blood pressure: The blood pressure is evaluated as low risk and has a weak correlation (0.2 and 0.3) with other parameters, so it is classified into the low-risk level. Blood oxygen saturation: There is a strong negative correlation (-0.7) between blood oxygen saturation and heart rate, and it is evaluated as high risk, so it is classified into the high-risk level. Body temperature: The body temperature is evaluated as low risk, has a weak positive correlation (0.3) with blood pressure and a negative correlation (-0.5) with blood oxygen saturation, so it is classified into the low-risk level.

[0116] According to the hierarchical division, the system presets different sampling frequencies for each level. The following are the sampling frequencies corresponding to each level: High-risk level (high-fluctuation state): For example, heart rate and blood oxygen saturation are classified as high-risk levels. The fluctuations of these parameters are relatively large, and more frequent sampling is required to monitor their changes in real time. Therefore, high-frequency sampling (multiple samplings per minute) is set for these parameters. Medium-risk level (medium-fluctuation state): For example, when the fluctuations of certain parameters are relatively stable but still have a certain degree of volatility, the system will allocate medium-frequency sampling (one sampling per minute) for them. Low-risk level (low-fluctuation state): Blood pressure and body temperature are evaluated as low-risk, and their fluctuations are small. Therefore, a lower sampling frequency can be selected to ensure that the system saves electrical energy without affecting the validity of the data. For these parameters, the system can set low-frequency sampling (one sampling per hour).

[0117] Example 2:

[0118] A vital sign monitoring wristband, comprising:

[0119] An in-built sensor module for real-time monitoring of multiple preset vital sign parameters;

[0120] A data processing unit for receiving and processing the vital sign parameter data collected by the in-built sensor module, and uploading the data to a cloud data processing platform to generate a set of vital sign parameter data;

[0121] A short-term fluctuation evaluation module for respectively performing short-term fluctuation evaluation on each preset vital sign parameter based on the set of vital sign parameter data to obtain the short-term fluctuation evaluation results of each parameter;

[0122] A long-term trend evaluation module for respectively performing long-term trend evaluation on each preset vital sign parameter based on the set of vital sign parameter data to obtain the long-term trend evaluation results of each parameter;

[0123] A machine learning decision module for inputting the short-term fluctuation evaluation results and the long-term trend evaluation results into a pre-trained machine learning model, outputting the fluctuation risk types of each preset vital sign parameter, and generating a preliminary classification data set;

[0124] A cross-correlation feedback module for analyzing the cross-correlation between the preset vital sign parameters;

[0125] A sampling frequency optimization module for classifying the fluctuation states of each preset vital sign parameter according to the preliminary classification data set and the cross-correlation feedback information, and adjusting the sampling frequencies of the vital sign parameters according to the preset sampling optimization rules.

[0126] Example 3:

[0127] A vital sign monitoring system, comprising:

[0128] A vital sign monitoring terminal, which is built-in with a sensor module for real-time monitoring of multiple preset vital sign parameters, and a data processing unit for receiving and processing the vital sign parameter data collected by the built-in sensor module, and uploading the data to a cloud data processing platform to generate a set of vital sign parameter data;

[0129] A cloud data processing platform, which is used to receive the set of vital sign data uploaded from the vital sign monitoring wristband, and perform the following steps:

[0130] Conduct a short-term fluctuation assessment on each vital sign parameter to obtain a short-term fluctuation assessment result;

[0131] Conduct a long-term trend assessment on each vital sign parameter to obtain a long-term trend assessment result;

[0132] Input the assessment results into a pre-trained machine learning model, output the fluctuation risk type of each vital sign parameter, and hierarchically classify the fluctuation states of each parameter according to the cross-correlation between different parameters;

[0133] Optimize the sampling frequency of each preset vital sign parameter according to the result of the hierarchical classification.

[0134] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0135] It should be understood that in various embodiments of the present application, the magnitudes of the serial numbers of the above processes do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0136] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0137] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0138] As described above, it is only the specific implementation manner of the present application. However, the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

Claims

1. A method for monitoring vital signs, characterized in that: The following steps are involved: The built-in sensor of the monitoring wristband monitors multiple preset vital sign parameters in real time and uploads them to the cloud to obtain a collection of vital sign parameter data; Based on the vital sign parameter data set, short-term fluctuation assessment and long-term trend assessment are performed on each preset vital sign parameter to obtain short-term fluctuation assessment results and long-term trend assessment results; Substitute the short-term fluctuation assessment results and the long-term trend assessment results into the pre-trained machine learning model to obtain the fluctuation risk type of each preset vital sign parameter, and summarize the fluctuation risk types to obtain an initial classification data group; Based on the initial classification data group and the cross-correlation feedback results of each preset vital sign parameter, the fluctuation state of each preset vital sign parameter is hierarchically divided, and the sampling frequency of each preset vital sign parameter is adjusted according to the preset sampling optimization rule.

2. A vital sign monitoring method according to claim 1, characterized in that: Short-term volatility assessment refers to: Based on the changes in preset vital sign parameters within a preset time window, a dynamic fluctuation index is calculated to measure its instantaneous fluctuation.

3. A vital sign monitoring method according to claim 2, characterized in that: Long-term trend assessment refers to: Based on historical data of preset vital sign parameters, a health trend index is calculated to assess their long-term health status.

4. A vital sign monitoring method according to claim 3, characterized in that: The machine learning model is a convolutional neural network. The dynamic fluctuation index and the health trend index are substituted into the convolutional neural network model together. The convolutional neural network outputs a result of 0 or 1, where 0 indicates that the fluctuation risk type corresponding to the vital sign parameter is a low-risk type, and 1 indicates that the fluctuation risk type corresponding to the vital sign parameter is a high-risk type. The vital sign parameters are sorted according to the corresponding numbers of the preset vital sign parameters, and all the convolutional neural network output results are summarized to obtain an initial classification data group.

5. A vital sign monitoring method according to claim 4, characterized in that: The logic for obtaining the dynamic volatility index is: The data of the preset vital sign parameter X in the preset time window is obtained from the sensor in real time. Assuming that the data collected in the preset time window is a time series, the standard deviation is used as a measure of fluctuation, and the weighted standard deviation is considered to increase the sensitivity to recent changes: wi represents the weight coefficient corresponding to the i-th numbered data in the time series, N represents the total number of data in the time series, and σ(X) represents the volatility measurement value of the time series; Calculate the change in volatility as a complementary indicator of dynamic volatility: Calculate the difference between the maximum and minimum values ​​of the data in the current time window to obtain the fluctuation amplitude value A(X); compare the fluctuation amplitude of the current time window with the change in the fluctuation amplitude of the previous time window to obtain the amplitude change value. The calculation formula is: ΔA(X)=A(X)-A(Xprevious); A(Xprevious) represents the fluctuation amplitude value corresponding to the previous time window, and ΔA(X) represents the amplitude change value; The calculation formula of dynamic volatility index is: ∈ represents a preset non-zero constant, and DVI represents the dynamic fluctuation index.

6. A vital sign monitoring method according to claim 5, characterized in that: The logic for obtaining the weight coefficient corresponding to the i-th numbered data in the time series is: Based on the exponential decay factor, dynamic adjustment is performed according to the fluctuation rate of the current vital sign parameters. The formula is: represents the rate of change of the current vital sign parameter sampling data at time ti, β is the non-zero influence coefficient of the preset control fluctuation rate on attenuation, t0 represents the time starting point of the time series X, ti represents the time point of the current vital sign parameter sampling data in the time series X, and λ is the preset attenuation rate constant.

7. A vital sign monitoring method according to claim 6, characterized in that: The logic for obtaining the health trend index is: The historical data of the preset vital sign parameter X in the preset historical window is obtained from the sensor in real time, and the trend of long-term changes is extracted through regression analysis: X(j)=a+b·Tj+ε; Tj represents the sampling data of the vital sign parameter X at time point j in the historical window, a and b are regression coefficients, representing the starting value and change rate of the vital sign parameter X in the historical window, ε is the regression residual, X(Tj) represents the regression prediction value of the vital sign parameter X at time point j in the historical window, and the trend intensity coefficient TSI is defined to measure the changing trend of the health status, TSI=|b|, and at the same time, the root mean square error between the sampling value and the regression prediction value of the vital sign parameter X in the historical window is introduced to obtain the long-term stability coefficient LTSI reflecting the long-term stability of the health parameter; The calculation formula of the health trend index is: LTSImax is the preset standard value of the long-term stability coefficient, which is used to normalize the long-term stability coefficient LTSI. The normalized long-term stability coefficient LTSI is in the range of [0, 1]. f1 and f2 are both preset non-zero adjustment coefficients, and the sum of f1 and f2 is one. HTI represents the health trend index.

8. A vital sign monitoring method according to claim 7, characterized in that: Based on the initial classification data group and the cross-correlation feedback results of each preset vital sign parameter, the fluctuation state of each preset vital sign parameter is divided into levels: Fuzzy reasoning is used to obtain the result of the preset vital sign parameter X in the initial classification data group. The cross-correlation value of the preset vital sign parameter X and other vital sign parameters calculated by the Pearson correlation coefficient calculation formula, as well as the preset correlation threshold, are used as the input variables of the fuzzy logic. The level corresponding to the fluctuation state of the vital sign parameter X is used as the output variable of the fuzzy logic. The input variable is fuzzified and the value of the input variable is converted into a fuzzy set. The output variable is fuzzified and the output variable is converted into a fuzzy set. Fuzzy rules are formulated to describe the adaptation degree of each level under different data type combinations. The fuzzified input variable is inferred through fuzzy rules to obtain the level corresponding to the fluctuation state of the vital sign parameter X.

9. A vital signs monitoring wristband, used to implement a vital signs monitoring method as claimed in claims 1-8, characterized in that: include: Built-in sensor module for real-time monitoring of multiple preset vital sign parameters; A data processing unit, used to receive and process vital sign parameter data collected by the built-in sensor module, and upload the data to a cloud data processing platform to generate a vital sign parameter data set; A short-term fluctuation assessment module, based on the vital sign parameter data set, performs short-term fluctuation assessment on each preset vital sign parameter to obtain a short-term fluctuation assessment result for each parameter; A long-term trend evaluation module, based on the vital sign parameter data set, performs a long-term trend evaluation on each preset vital sign parameter to obtain a long-term trend evaluation result for each parameter; The machine learning decision module inputs the short-term fluctuation assessment results and the long-term trend assessment results into a pre-trained machine learning model, outputs the fluctuation risk type of each preset vital sign parameter, and generates a preliminary classification data group; A cross-correlation feedback module analyzes the cross-correlation between various preset vital sign parameters; The sampling frequency optimization module hierarchically divides the fluctuation state of each preset vital sign parameter according to the preliminary classification data group and the cross-correlation feedback information, and adjusts the sampling frequency of each vital sign parameter according to the preset sampling optimization rules.

10. A vital sign monitoring system, used to implement a vital sign monitoring method according to any one of claims 1 to 8, characterized in that: include: A vital sign monitoring terminal, having a built-in sensor module for real-time monitoring of a plurality of preset vital sign parameters, and a data processing unit for receiving and processing vital sign parameter data collected by the built-in sensor module, and uploading the data to a cloud data processing platform to generate a vital sign parameter data set; The cloud data processing platform is used to receive the vital sign data set uploaded from the vital sign monitoring wristband and perform the following steps: Perform short-term fluctuation assessment on each vital sign parameter to obtain a short-term fluctuation assessment result; Conduct a long-term trend assessment on each vital sign parameter to obtain a long-term trend assessment result; The evaluation results are input into the pre-trained machine learning model to output the fluctuation risk type of each vital sign parameter, and the fluctuation status of each parameter is hierarchically divided according to the cross-correlation between different parameters; According to the results of the hierarchical division, the sampling frequency of each preset vital sign parameter is optimized.

Images