Construction personnel vital sign monitoring bracelet and monitoring method thereof

Abstract

The invention discloses a constructor vital sign monitoring bracelet and a monitoring method thereof, and particularly relates to the technical field of health monitoring, the constructor vital sign monitoring bracelet comprises a vital sign sensing module, an environment monitoring module, a local data processing module, a cloud analysis module, a local early warning module and a cloud early warning module; the vital sign sensing module monitors vital sign data in real time, and the environment monitoring module monitors environment information; the local data processing module performs preliminary health analysis and determines data uploading or early warning according to a result; the cloud analysis module carries out health risk assessment through a decision tree algorithm in combination with environmental factors, and the cloud early warning module determines whether to carry out positioning early warning or not according to an analysis result; the local early warning module executes a corresponding early warning strategy; according to the method, a consideration mechanism of environmental factors is introduced, and the environmental data and the vital sign data are combined, so that the health risk of the construction personnel can be evaluated more comprehensively, the risk evaluation model is dynamically adjusted, and the accuracy of health risk early warning is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of health monitoring, and more specifically, to a construction worker's vital sign monitoring bracelet and a monitoring method thereof. Background Art

[0002] With the development of modern construction industry and engineering projects, the working environment of construction workers has become increasingly complex and harsh, especially in high temperature environments. The health risks faced by construction workers have increased significantly, such as heat stroke and heat stroke. Therefore, how to effectively monitor the vital signs of construction workers, detect health abnormalities in time and issue early warnings has become an urgent problem to be solved. Traditional health monitoring of construction workers usually relies on manual observation and simple vital sign measurement methods, such as regular temperature monitoring and heart rate monitoring. However, these methods are not only inefficient and difficult to respond to sudden health problems in a timely manner, but also lack comprehensive consideration of environmental factors in special environments such as high temperature and humidity, and cannot accurately assess the actual health risks of construction workers.

[0003] In the prior art, some smart wearable devices can monitor vital signs in real time, but their analytical capabilities are limited. They usually rely on local devices for simple health data processing and are difficult to cope with the influence of complex environmental factors. In addition, existing health monitoring systems usually only consider vital sign data and fail to combine environmental factors with health data for comprehensive analysis, thereby reducing the accuracy and timeliness of early warnings. Therefore, there is an urgent need for a method that can monitor the vital signs of construction workers in real time and, when necessary, conduct comprehensive analysis in combination with environmental factors to provide accurate and timely health early warning methods to effectively protect the health and safety of construction workers in harsh environments. Summary of the invention

[0004] To achieve the above object, the present invention provides the following technical solutions: A construction worker's vital signs monitoring bracelet, comprising a vital signs sensing module, an environmental monitoring module, a local data processing module, a cloud analysis module, a local early warning module, and a cloud early warning module; The vital sign sensing module is used to monitor the vital sign information of construction workers in real time through a variety of preset vital sign sensors and transmit it to the local data processing module; The environmental monitoring module is used to monitor the construction workers' working environment information in real time through a variety of pre-installed environmental sensors and transmit it to the cloud analysis module; The local data processing module is used to preliminarily process the vital signs information of the construction workers monitored in real time and conduct a preliminary health status analysis locally. Then, based on the results of the preliminary health status analysis, it is decided to transmit the preliminary processing results to the cloud analysis module or send local warning signals of different levels to the local warning module; The cloud analysis module determines whether environmental factors need to be considered based on the decision tree algorithm and the construction personnel's working environment information, and performs cloud analysis operations based on the judgment results, and then transmits the cloud analysis results to the cloud warning module; The local warning module is used to execute the preset local warning strategy according to the level of the local warning signal; The cloud warning module is used to decide whether to perform cloud positioning warning operations based on the cloud analysis results.

[0005] In a preferred embodiment, the initial processing of the vital signs information of the construction personnel monitored in real time refers to: The collected vital signs data are subjected to preliminary noise filtering and correction, and then the basic statistics of the preset vital signs data are calculated.

[0006] In a preferred embodiment, performing a preliminary health status analysis locally refers to: The average value, maximum value, and minimum value of the basic statistics of each vital sign data are obtained, and then the difference between the maximum value and the minimum value of the target vital sign data is calculated, and the average value is compared with the preset standard numerical range of the target vital sign data. If the difference satisfies the condition that the difference is less than or equal to the preset standard fluctuation threshold of the target vital sign data and the average value is within the preset standard numerical range of the target vital sign data, the target vital sign data is marked as normal data; if the difference does not satisfy the condition that the difference is less than or equal to the preset standard fluctuation threshold of the target vital sign data and the average value is within the preset standard numerical range of the target vital sign data, the target vital sign data is marked as abnormal data.

[0007] In a preferred embodiment, according to the result of the preliminary health status analysis, determining whether to transmit the preliminary processing result to the cloud analysis module or to send local warning signals of different levels to the local warning module refers to: When there is no vital sign data marked as abnormal data, the preliminary processing results are transmitted to the cloud analysis module. When there is vital sign data marked as abnormal data, the total number m of vital sign data types marked as abnormal data and the total number n of vital sign data types marked as normal data are counted, and the proportion p of abnormal data is calculated, p=m / (n+m). There is a preset proportion range interval-warning level form, and a local warning signal of the corresponding level is issued to the local warning module according to the proportion range interval into which the proportion p of abnormal data falls.

[0008] In a preferred embodiment, the cloud analysis module determines whether environmental factors need to be considered based on the decision tree algorithm and the construction personnel's working environment information: The environment monitoring module collects the construction workers' working environment information monitored in real time by a variety of preset environmental sensors to obtain an environmental data set, which consists of multiple environmental data parameters; In a preferred embodiment, multiple environmental data parameters are screened to obtain a main parameter and multiple subsidiary parameters; Set the root node and branch nodes of the decision tree: the root node uses the main parameters as the judgment condition, and the branch node uses the subsidiary parameters as the judgment condition. Then set the decision conditions of the root node and the branch node. The decision tree will output whether environmental factors need to be considered as part of the vital signs analysis. If the final output result is "yes", the environmental factors will be taken into consideration when analyzing the vital signs data in the cloud. If the final output result is "no", the environmental factors will not be taken into consideration when analyzing the vital signs data in the cloud.

[0009] In a preferred embodiment, screening multiple environmental data parameters to obtain a main parameter and multiple subsidiary parameters means: The correlation coefficients of all environmental data parameters in the environmental data set with the possibility of environmental heat stress are calculated respectively, and the environmental data parameters are sorted from high to low according to the absolute value of the correlation coefficient. After sorting, the parameters with the strongest correlation with the possibility of environmental heat stress are selected as the main parameters, and then the correlation coefficients of the remaining environmental data parameters are compared with the preset screening threshold. If the absolute value of the correlation coefficient of the environmental data parameter is less than the preset screening threshold, it is marked as a useless parameter and eliminated. If the absolute value of the correlation coefficient of the environmental data parameter is not less than the preset screening threshold, it is marked as an auxiliary parameter and retained.

[0010] In a preferred embodiment, the correlation coefficient uses the Pearson correlation coefficient or the Spearman rank correlation coefficient.

[0011] In a preferred embodiment, cloud analysis refers to: When environmental factors are taken into consideration, the basic statistics, main parameters, and subsidiary parameters of all preset vital sign data are summarized to obtain an input vector X, and then the input vector X is input into the pre-trained neural network model 1, and the neural network model 1 outputs a risk value 1; When environmental factors are not taken into consideration, the basic statistics of all preset vital signs data are summarized to obtain an input vector P, and then the input vector P is input into the pre-trained neural network model 2, and the neural network model 2 outputs a risk value 2.

[0012] In a preferred embodiment, the cloud warning module is used to decide whether to perform a cloud positioning warning operation based on the cloud analysis results, which means that the cloud warning module receives risk value one or risk value two, compares the risk value one or risk value two with the corresponding preset risk warning threshold, and if it exceeds the corresponding preset risk warning threshold, obtains the location information of the monitoring bracelet and issues a cloud warning.

[0013] In a preferred embodiment, a method for monitoring vital signs of construction workers comprises the following steps: The vital signs of construction workers are monitored in real time through various preset vital signs sensors, and the working environment of construction workers is monitored in real time through various preset environmental sensors; Preliminary processing of the vital signs of construction workers monitored in real time and preliminary health status analysis locally, then based on the results of the preliminary health status analysis, decide whether to transmit the preliminary processing results to the cloud for further analysis or send local warning signals of different levels, and then execute the preset local warning strategy according to the level of the local warning signal; Based on the decision tree algorithm and the construction personnel's working environment information, it is determined whether environmental factors need to be considered, and cloud analysis operations are performed based on the judgment results. Then, based on the cloud analysis results, it is decided whether to perform cloud positioning and early warning operations.

[0014] Technical effects and advantages of the present invention: The present invention uses smart wearable devices to monitor the vital signs data of construction workers in real time, including heart rate, blood oxygen saturation, body temperature, etc., and combines environmental data collected by environmental sensors, such as temperature, humidity, light intensity, etc. Through the cloud analysis module, the health status of construction workers can be evaluated in real time to ensure the accuracy and timeliness of health monitoring. The present invention specifically introduces a mechanism for considering environmental factors. By combining environmental data with vital signs data, the health risks of construction workers can be more comprehensively evaluated. According to different environmental conditions, the risk assessment model, that is, the neural network model, is dynamically adjusted to improve the accuracy of health risk warnings, especially in high temperatures or other harsh environments. The system can identify potential health threats earlier.

[0015] The dual risk assessment mechanism of the present invention uses two neural network models for risk assessment: one takes environmental factors into account and the other does not. By comparing the risk assessment results of the two models, the system can more flexibly respond to health risks under different environmental conditions, thereby providing more personalized and accurate health management services. When it is detected that the health risk of the construction workers exceeds the preset warning threshold, the cloud-based warning module can automatically obtain the real-time location of the construction workers and immediately issue a warning signal. This intelligent early warning mechanism ensures that when health risks occur, managers can obtain the location information of the construction workers in a timely manner, quickly take necessary intervention measures, and reduce the incidence and severity of health problems.

[0016] Through accurate risk assessment and multi-level warning threshold setting, the present invention effectively reduces the incidence of false alarms, sets different warning thresholds according to different scenarios, ensures that warning signals are triggered only when there is a real health risk, avoids overreaction, and improves the credibility and practicality of warnings. The present invention provides a comprehensive and intelligent health monitoring and early warning solution in the construction environment, effectively improving the safety level of construction workers. Especially in extreme weather or harsh working environments, it can play a key role in helping to prevent and respond to health emergencies in a timely manner and reduce health risks on construction sites. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings; Figure 1 The figure is a schematic diagram of a construction worker's vital signs monitoring bracelet in the present invention.

[0018] Figure 2 The schematic diagram is a construction worker vital sign monitoring method according to the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Reference Figure 1 - Figure 2 The following examples are obtained: Embodiment 1: A construction worker's vital signs monitoring bracelet, comprising a vital signs sensing module, an environmental monitoring module, a local data processing module, a cloud analysis module, a local early warning module, and a cloud early warning module; The vital signs sensing module is used to monitor the vital signs of construction workers in real time through a variety of preset vital signs sensors and transmit them to the local data processing module; sensors: heart rate sensor, blood oxygen saturation sensor, body temperature sensor, etc., are used to monitor the vital signs of construction workers in real time. Data collection: collect heart rate, blood oxygen, body temperature and other information in real time and transmit them to the local data processing module.

[0021] The environment monitoring module is used to monitor the construction workers' working environment information in real time through a variety of pre-set environmental sensors and transmit it to the cloud analysis module; light sensor: used to detect the ambient light intensity. Temperature sensor: detects the temperature of the construction environment to determine whether it is in a high temperature environment. Humidity sensor: detects air humidity to determine the degree of air dryness.

[0022] The local data processing module is used to preliminarily process the vital signs information of the construction workers monitored in real time and conduct a preliminary health status analysis locally. Then, based on the results of the preliminary health status analysis, it is decided to transmit the preliminary processing results to the cloud analysis module or send local warning signals of different levels to the local warning module; The cloud analysis module determines whether environmental factors need to be considered based on the decision tree algorithm and the construction personnel's working environment information, and performs cloud analysis operations based on the judgment results, and then transmits the cloud analysis results to the cloud warning module; The local warning module is used to execute the preset local warning strategy according to the level of the local warning signal; The cloud warning module is used to decide whether to perform cloud positioning warning operations based on the cloud analysis results.

[0023] The initial processing of the vital signs of construction workers monitored in real time refers to: The collected vital signs data are subjected to preliminary noise filtering and correction, and then the basic statistics of the preset vital signs data are calculated.

[0024] Noise filtering: When collecting data, the sensor may be affected by factors such as the external environment, equipment jitter, and electromagnetic interference, resulting in noise in the data. For example, in a construction environment, the heart rate sensor may record abnormal heart rate values ​​(such as sudden abnormally high or low values) due to worker movement or equipment vibration. For example: If the heart rate recorded by the heart rate sensor at a certain point in time is 200bpm, and the surrounding values ​​are all in the range of 80-100bpm, it may be noise. Eliminating noisy data can improve the accuracy of the data and ensure the reliability of subsequent analysis and decision-making.

[0025] Data correction: Correct the deviation of data to ensure that the data collected by the sensor accurately reflects the actual situation. The sensor may cause inaccurate measurements due to temperature changes, equipment aging or other factors, which need to be corrected through the correction model. For example: If the temperature sensor shows a measured value of 32°C when the actual temperature is 30°C, it can be corrected to an actual value close to 30°C through historical calibration data or standard calibration curves. Significance: Data correction can eliminate systematic errors, ensure that the data matches the actual situation more closely, and improve the accuracy of the overall monitoring system.

[0026] Basic statistics: Statistics include the mean, maximum, minimum, and standard deviation of vital sign data. These statistics are used to assess the health status of construction workers and identify abnormalities. Mean: For example, the mean of heart rate is calculated to assess whether the daily heart rate of construction workers is within the normal range. If a worker's average heart rate during work is 90bpm, and the normal range is 60-100bpm, it is considered normal; but if the average heart rate is above 120bpm, it may indicate that the worker is in a state of excessive tension or physical exertion. Maximum and minimum values: Recording the maximum and minimum values ​​of body temperature can determine whether a high fever or hypothermia occurs. Standard deviation: The standard deviation reflects the fluctuation of the data. If the standard deviation of the heart rate is very large, it may mean that the heart rate is unstable and there is a potential health risk. Calculating these statistics can provide a preliminary quantitative assessment of the health status of workers, help identify health risks in a timely manner, and provide early warnings or further analysis.

[0027] Conducting a preliminary local health status analysis means: The average value, maximum value, and minimum value of the basic statistics of each vital sign data are obtained, and then the difference between the maximum value and the minimum value of the target vital sign data is calculated, and the average value is compared with the preset standard numerical range of the target vital sign data. If the difference satisfies the condition that the difference is less than or equal to the preset standard fluctuation threshold of the target vital sign data and the average value is within the preset standard numerical range of the target vital sign data, the target vital sign data is marked as normal data; if the difference does not satisfy the condition that the difference is less than or equal to the preset standard fluctuation threshold of the target vital sign data and the average value is within the preset standard numerical range of the target vital sign data, the target vital sign data is marked as abnormal data.

[0028] The significance of conducting preliminary health status analysis locally is to evaluate whether the vital signs data of construction workers are within the safe range through simple and fast calculations, so as to identify potential health problems in a timely manner. This step can help the system respond quickly, for example, when there are abnormal fluctuations in vital sign data, further cloud analysis can be triggered or an alarm can be issued directly. This local analysis method can reduce the cloud computing burden, improve the overall response speed of the system, and ensure timely warning.

[0029] Specific examples: Example 1: Heart rate data analysis: Assume that the basic statistics of a construction worker's heart rate data over a period of time are as follows: average value 85bpm; maximum value 100bpm; minimum value 70bpm; the preset standard value range is 60-100bpm, and the standard fluctuation threshold is 40bpm.

[0030] Difference calculation: Maximum value minus minimum value, the difference is 100-70=30bpm. Comparative analysis: The difference of 30bpm is less than the standard fluctuation threshold of 40bpm. The average value of 85bpm is within the preset standard value range of 60-100bpm. Result: The heart rate data is marked as normal data.

[0031] Example 2: Body temperature data analysis: Assume that the body temperature data statistics of another construction worker are as follows: average value 38.5℃; maximum value 39.2℃; minimum value 37.8℃; the preset standard value range is: 36.5-37.5℃, and the standard fluctuation threshold is 1℃. Difference calculation: maximum value minus minimum value, the difference is 39.2-37.8=1.4℃. The difference of 1.4℃ is greater than the standard fluctuation threshold of 1℃. The average value of 38.5℃ is not within the standard value range of 36.5-37.5℃. Result: The body temperature data is marked as abnormal data.

[0032] Further expansion: In the preliminary health status analysis, the difference between the maximum value and the minimum value can also be replaced by the standard deviation in the basic statistics. The following is an explanation of the applicable scenarios of the two: Difference between maximum and minimum: Suitable for quickly assessing the magnitude of data fluctuations, especially when data collection frequency is low or observation time is short. This method is simple and direct and can quickly identify obvious fluctuation anomalies. Applicable parameters: When the data has clear upper and lower limits (such as heart rate, body temperature) and the fluctuation range is relatively obvious, the difference between the maximum and minimum values ​​can effectively reflect the degree of dispersion of the data.

[0033] Standard Deviation: Applicable to assessing the volatility and dispersion of data, especially when data is collected frequently or observed over a long period of time. Standard Deviation provides a more accurate measure of volatility. Applicable Parameters: When data fluctuates frequently and has many subtle changes (such as continuously monitored heart rate or blood oxygen saturation), Standard Deviation can better capture these changes and provide a more reliable assessment of volatility.

[0034] When conducting a preliminary health status analysis, the choice of the difference between the maximum value and the minimum value or the standard deviation depends on the characteristics of the data and the actual application scenario. For simple fluctuations in a short period of time, the difference between the maximum value and the minimum value is a fast and effective indicator. For scenarios that require more detailed analysis, the standard deviation provides a more comprehensive measure of volatility. Since the general monitoring time is short, the present invention provides an indicator of the difference between the maximum value and the minimum value. If the monitoring time is extended in actual use, the standard deviation, a more comprehensive volatility measurement indicator, is used.

[0035] According to the results of the preliminary health status analysis, it is decided to transmit the preliminary processing results to the cloud analysis module or send different levels of local warning signals to the local warning module: When there is no vital sign data marked as abnormal data, the preliminary processing results are transmitted to the cloud analysis module. When there is vital sign data marked as abnormal data, the total number m of vital sign data types marked as abnormal data and the total number n of vital sign data types marked as normal data are counted, and the proportion p of abnormal data is calculated, p=m / (n+m). There is a preset proportion range interval-warning level form, and a local warning signal of the corresponding level is issued to the local warning module according to the proportion range interval into which the proportion p of abnormal data falls.

[0036] The health status of construction workers can be assessed in time through local pre-processing, and corresponding early warning signals can be quickly issued when potential health risks occur. By calculating the proportion p of abnormal data, the level of early warning can be dynamically adjusted to ensure that health risks receive appropriate attention and treatment. This hierarchical early warning mechanism can reduce false alarms and improve the accuracy of early warnings, ensuring that construction workers do not overreact to relatively minor health abnormalities, and can respond in a timely manner in more serious cases.

[0037] Specific example: Construction worker vital signs monitoring: Assume that a construction worker wears a bracelet that monitors the following three vital signs data: heart rate, blood oxygen saturation, and body temperature. Heart rate: marked as normal data; blood oxygen saturation: marked as abnormal data; body temperature: marked as abnormal data. In this case: the total number of vital sign data types marked as abnormal data is m=2 (blood oxygen saturation and body temperature). The total number of vital sign data types marked as normal data is n=1 (heart rate). The proportion of abnormal data p is calculated to be 0.67. Assume that the warning level form is as follows: 0.0≤p<0.3: low-level warning; 0.3≤p<0.6: medium-level warning; 0.6≤p≤1.0: high-level warning; in this example, the proportion p falls into the interval of 0.6≤p≤1.0, and a high-level warning signal is sent to the local warning module.

[0038] The cloud analysis module determines whether environmental factors need to be considered based on the decision tree algorithm and the construction personnel's working environment information: The environment monitoring module collects the construction workers' working environment information monitored in real time by a variety of preset environmental sensors to obtain an environmental data set, which consists of multiple environmental data parameters; Screening multiple environmental data parameters to obtain a main parameter and multiple subsidiary parameters; Set the root node and branch nodes of the decision tree: the root node uses the main parameters as the judgment condition, and the branch node uses the subsidiary parameters as the judgment condition. Then set the decision conditions of the root node and the branch node. The decision tree will output whether environmental factors need to be considered as part of the vital signs analysis. If the final output result is "yes", the environmental factors will be taken into consideration when analyzing the vital signs data in the cloud. If the final output result is "no", the environmental factors will not be taken into consideration when analyzing the vital signs data in the cloud.

[0039] Through reasonable environmental parameter screening and the application of decision trees, it is dynamically determined whether environmental factors need to be considered in the health analysis of construction workers. Various factors in the construction environment, such as temperature, humidity, light intensity, etc., will affect the vital signs of construction workers. Through this process, the system can intelligently decide whether to take these environmental factors into consideration during health analysis, thereby improving the accuracy of the analysis and the effectiveness of the early warning. This helps to more accurately identify potential health risks in complex environments and make corresponding early warnings or interventions in a timely manner.

[0040] For example, suppose the environmental monitoring module at the construction site monitors the following environmental data parameters: temperature, humidity, and light intensity. Data screening and parameter division: Based on the significance and relevance of each parameter on the health of construction workers, temperature is first selected as the main parameter because temperature has the greatest impact on health. Humidity and light intensity are divided into subsidiary parameters.

[0041] Decision tree settings: Root node: Set the temperature as the judgment condition. For example, if the temperature exceeds 35 degrees, it is considered that the ambient temperature has a significant impact on health and other environmental factors need to be further considered.

[0042] Branch node: Set humidity and light intensity as judgment conditions. For example, if the humidity exceeds 75% and the light intensity is also high, the risk of heat stress may be aggravated, so these factors need to be considered comprehensively.

[0043] Decision tree output: If the temperature is over 35 degrees and the humidity or light intensity also reaches a certain level, the decision tree outputs "yes", indicating that environmental factors need to be considered when analyzing vital signs data. On the contrary, if the temperature is not high or other auxiliary parameters are within the normal range, the decision tree outputs "no", indicating that environmental factors can be ignored and vital signs data can be directly analyzed.

[0044] Example explanation: Example 1: On a certain day, the construction environment temperature is 36 degrees, the humidity is 80%, and the light intensity is also high. According to the set decision tree, the temperature has exceeded the root node threshold of 35 degrees, and the humidity and light intensity are also high. The decision tree output result is "yes", indicating that environmental factors must be considered when analyzing the vital signs of construction workers.

[0045] Example 2: On another construction day, the temperature was 32 degrees, the humidity was 65 percent, and the light intensity was moderate. Although the temperature was at a high level, it did not reach the 35-degree threshold of the root node, and the decision tree output was "no", indicating that the impact of environmental factors was small and could be ignored. Environmental factors did not need to be considered specifically when analyzing vital signs data.

[0046] Screening multiple environmental data parameters to obtain a main parameter and multiple subsidiary parameters means: The correlation coefficients of all environmental data parameters in the environmental data set with the possibility of environmental heat stress are calculated respectively, and the environmental data parameters are sorted from high to low according to the absolute value of the correlation coefficient. After sorting, the parameters with the strongest correlation with the possibility of environmental heat stress are selected as the main parameters, and then the correlation coefficients of the remaining environmental data parameters are compared with the preset screening threshold. If the absolute value of the correlation coefficient of the environmental data parameter is less than the preset screening threshold, it is marked as a useless parameter and eliminated. If the absolute value of the correlation coefficient of the environmental data parameter is not less than the preset screening threshold, it is marked as an auxiliary parameter and retained.

[0047] By analyzing the correlation of environmental data, the decision tree structure of health risk assessment is optimized. By screening and classifying environmental data parameters, the system can focus on those factors that have the greatest impact on environmental heat stress risks, thereby improving the accuracy and efficiency of decision-making. After eliminating useless parameters, the calculation complexity of the system is reduced, making health analysis more streamlined and efficient, while retaining the key factors that have a significant impact on health, ensuring the effectiveness of early warning.

[0048] The benefits of setting branch nodes in sequence: Improve decision-making efficiency: By setting the most relevant parameters as the root node, you can quickly decide whether to further consider other factors, thereby reducing unnecessary calculations and judgments. Enhance the accuracy of early warnings: After sorting, set branch nodes in sequence so that each step of the judgment is based on the most influential factors, ensuring the accuracy of the early warning signal. Simplify the decision tree structure: After filtering out useless parameters, the decision tree structure is more concise, reducing unnecessary complexity, making the system more flexible and efficient in practical applications.

[0049] For example: Assume that at a construction site, the following environmental data are obtained through the environmental monitoring module: temperature, humidity, wind speed, and light intensity. First, calculate the correlation coefficient between each environmental parameter and the possibility of environmental heat stress, and obtain the strength of the relationship between each parameter and heat stress. Sorting: After sorting according to the absolute value of the correlation coefficient, temperature has the highest correlation, followed by humidity and light intensity, and wind speed has the lowest. Main parameter determination: Since temperature has the strongest correlation with the possibility of environmental heat stress, temperature is selected as the main parameter. Filter useless parameters: Filter the remaining humidity, light intensity, and wind speed. Assuming that the screening threshold is set to a certain value such as 0.2, it is found that the absolute value of the correlation coefficient of wind speed is less than the threshold, and the wind speed is marked as a useless parameter and removed. Subsidiary parameter determination: If the absolute value of the correlation coefficient of humidity and light intensity is not less than the screening threshold, the system will retain these two parameters as subsidiary parameters.

[0050] Decision tree settings: Root node: Temperature is set as the root node of the decision tree to first determine whether environmental factors need to be considered. Branch nodes: Humidity and light intensity are used as branch nodes in turn to further determine whether early warning needs to be strengthened.

[0051] For example, at a construction site, if the temperature is high, the humidity is high, and the light intensity is strong, the decision tree determines that all parameters are at high risk, and it is decided to include environmental factors in the health analysis. On the contrary, if the temperature is high but the humidity and light intensity are within the normal range, environmental factors are not considered.

[0052] The correlation coefficient uses the Pearson correlation coefficient or the Spearman rank correlation coefficient. When the relationship between environmental data and the possibility of environmental heat stress is approximately linear, select the Pearson correlation coefficient. It is suitable for scenarios where the data conforms to the normal distribution. The Pearson correlation coefficient works best when the data is a continuous variable and does not contain significant outliers. Significance: The Pearson correlation coefficient measures the linear correlation between two variables, that is, the degree to which when one variable changes, the other variable changes synchronously in a linear manner. Its value is between -1 and 1. The closer the value is to 1 or -1, the stronger the linear correlation is. A value of 0 indicates no linear relationship. Scenarios for choosing the Spearman rank correlation coefficient: When the relationship between environmental data and the possibility of environmental heat stress is nonlinear or the data does not conform to the normal distribution, select the Spearman rank correlation coefficient. It is suitable for scenarios where the data is a sequence variable or contains outliers. When the data shows a monotonic relationship (that is, when one variable increases or decreases, the other variable also increases or decreases monotonically), the Spearman rank correlation coefficient is a better choice. Significance: The Spearman rank correlation coefficient measures the monotonic relationship between two variables. Whether the relationship is linear or nonlinear, as long as it is monotonic, it can be measured by the Spearman rank correlation coefficient. Its value is also between -1 and 1. The closer it is to 1 or -1, the stronger the monotonic correlation.

[0053] The Pearson correlation coefficient is applicable to situations where the linear relationship and data distribution are ideal (such as normal distribution), and can provide a more accurate linear correlation measurement. The Spearman rank correlation coefficient is applicable to situations where there are nonlinear relationships or data with outliers, non-normal distribution, etc., and can provide a more robust monotonicity measurement. In construction environment monitoring, if it is determined that the relationship between the data is linear and the data quality is good, the Pearson correlation coefficient is a suitable choice. If the data has nonlinear relationships or other complexities, the Spearman rank correlation coefficient is more suitable.

[0054] Cloud analytics refers to: When environmental factors are taken into consideration, the basic statistics, main parameters, and subsidiary parameters of all preset vital sign data are summarized to obtain an input vector X, and then the input vector X is input into the pre-trained neural network model 1, and the neural network model 1 outputs a risk value 1; When environmental factors are not taken into consideration, the basic statistics of all preset vital signs data are summarized to obtain an input vector P, and then the input vector P is input into the pre-trained neural network model 2, and the neural network model 2 outputs a risk value 2.

[0055] By incorporating or excluding environmental factors into the analysis, it can better adapt to the health risk assessment of construction workers under different environmental conditions. Both neural network models 1 and 2 use convolutional neural networks. Neural network models 1 and 2 calculate risk values ​​with and without considering environmental factors, respectively. This dual assessment mechanism helps to accurately predict the health risks of construction workers, especially in high temperature or other adverse environments. In addition, the introduction of neural networks enables the system to handle complex nonlinear relationships and improves the accuracy of risk assessment.

[0056] Example: Scenario 1: Consider environmental factors: Construction workers work in a hot environment, and the temperature, humidity, light intensity and other parameters recorded by the environmental monitoring system are high. At the same time, vital signs data such as heart rate and body temperature are also fluctuating. The basic statistics of these environmental data (such as temperature, humidity, light intensity) and vital signs data (such as heart rate, blood oxygen saturation, body temperature) are aggregated to generate input vector X. Input vector X is input into the pre-trained neural network model 1, and the model outputs a higher risk value 1, indicating that the construction worker has a higher health risk in the current environment.

[0057] Scenario 2: Environmental factors are not considered: The construction workers work in a relatively suitable environment, and environmental parameters such as temperature and humidity are within the normal range. Only the basic statistics of the vital signs data are used to generate the input vector P. The input vector P is input into the pre-trained neural network model 2, and the model outputs a lower risk value 2, indicating that the health risk of the construction worker is low.

[0058] Training process: Data collection: Collect a large amount of historical data of construction workers under different environmental conditions, including vital signs data (such as heart rate, blood oxygen saturation, and basic statistics of body temperature) and environmental data (such as temperature, humidity, and light intensity). Label these data and generate corresponding health risk labels (such as low risk, medium risk, and high risk).

[0059] Data preprocessing: Clean and denoise the collected raw data, handle missing values, and normalize the data so that they have the same scale before being input into the neural network. For data sets that take environmental factors into account, generate an input vector X containing vital signs data and environmental data; for data sets that do not take environmental factors into account, generate only an input vector P containing vital signs data.

[0060] Model training: Neural network model 1: Use the input vector X containing environmental factors and the corresponding health risk labels to train neural network model 1. Through multiple iterations of optimization, the model gradually learns how to combine vital sign data with environmental factors to predict health risks. Neural network model 2: Use the input vector P without environmental factors and the corresponding health risk labels to train neural network model 2. This model focuses on the impact of vital sign data on health risks.

[0061] Model validation and adjustment: Model 1 and Model 2 were validated using independent data sets to evaluate their accuracy and robustness. Based on the validation results, model parameters were adjusted to optimize their performance and ensure that they can accurately predict health risks in practical applications.

[0062] Model deployment: The verified and optimized Model 1 and Model 2 are deployed in the cloud respectively. When the cloud receives real-time data, it can quickly generate risk values ​​to assist in the health management of construction workers.

[0063] The cloud warning module is used to decide whether to perform a cloud positioning warning operation based on the cloud analysis results. It means that the cloud warning module receives risk value one or risk value two, and compares risk value one or risk value two with the corresponding preset risk warning threshold. If it exceeds the corresponding preset risk warning threshold, the location information of the monitoring bracelet is obtained and a cloud warning is performed to ensure that when the health risk of the construction personnel reaches a certain level, an early warning can be issued in time and the construction personnel can be located for rapid intervention. By comparing the risk value with the preset risk warning threshold, the present invention can automatically determine whether the current health risk requires immediate action. Different warning threshold settings can be distinguished according to whether environmental factors are taken into account to ensure that the warning is more accurate and effective, thereby reducing false alarms or missed alarms.

[0064] Preset risk warning thresholds: Setting different thresholds: Since risk value 1 (taking environmental factors into account) and risk value 2 (not taking environmental factors into account) come from different analysis models, their risk warning thresholds should also be different. Threshold for risk value 1: Usually a lower threshold is set. Because when environmental factors are taken into account, the risk assessment will be more comprehensive, so even if the risk value is slightly higher, it may mean a higher actual risk and require a more timely warning. Threshold for risk value 2: Set a slightly higher threshold. Because environmental factors are not taken into account, the model's risk assessment may be more conservative, and a warning will only be triggered when the vital signs data clearly show abnormalities.

[0065] Example: Scenario 1: Considering environmental factors: The calculated risk value is 70, which exceeds the preset risk warning threshold of 50. The cloud warning module recognizes this situation, immediately obtains the location of the construction worker's wristband, and sends a cloud warning signal to notify the management personnel, while providing the real-time location of the construction worker so that emergency measures can be taken.

[0066] Scenario 2: Without considering environmental factors: The calculated risk value 2 is 85, but the preset risk warning threshold is 90. After comparison, the cloud warning module finds that the risk value 2 does not exceed the threshold, so it will not issue a warning signal, and believes that the current risk level is within an acceptable range.

[0067] Embodiment 2: A method for monitoring vital signs of construction workers, comprising the following steps: The vital signs of construction workers are monitored in real time through various preset vital signs sensors, and the working environment of construction workers is monitored in real time through various preset environmental sensors; Preliminary processing of the vital signs of construction workers monitored in real time and preliminary health status analysis locally, then based on the results of the preliminary health status analysis, decide whether to transmit the preliminary processing results to the cloud for further analysis or send local warning signals of different levels, and then execute the preset local warning strategy according to the level of the local warning signal; Based on the decision tree algorithm and the construction personnel's working environment information, it is determined whether environmental factors need to be considered, and cloud analysis operations are performed based on the judgment results. Then, based on the cloud analysis results, it is decided whether to perform cloud positioning and early warning operations.

[0068] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0069] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0070] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0071] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0072] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A construction worker vital signs monitoring bracelet, characterized in that: It includes vital signs sensing module, environmental monitoring module, local data processing module, cloud analysis module, local warning module and cloud warning module; The vital sign sensing module is used to monitor the vital sign information of construction workers in real time through a variety of preset vital sign sensors and transmit it to the local data processing module; The environmental monitoring module is used to monitor the construction workers' working environment information in real time through a variety of pre-installed environmental sensors and transmit it to the cloud analysis module; The local data processing module is used to preliminarily process the vital signs information of the construction workers monitored in real time and conduct a preliminary health status analysis locally. Then, based on the results of the preliminary health status analysis, it is decided to transmit the preliminary processing results to the cloud analysis module or send local warning signals of different levels to the local warning module; The cloud analysis module determines whether environmental factors need to be considered based on the decision tree algorithm and the construction personnel's working environment information, and performs cloud analysis operations based on the judgment results, and then transmits the cloud analysis results to the cloud warning module; The local warning module is used to execute the preset local warning strategy according to the level of the local warning signal; The cloud warning module is used to decide whether to perform cloud positioning warning operations based on the cloud analysis results.

2. A construction worker vital signs monitoring bracelet according to claim 1, characterized in that: The initial processing of the vital signs of construction workers monitored in real time refers to: The collected vital signs data are subjected to preliminary noise filtering and correction, and then the basic statistics of the preset vital signs data are calculated.

3. A construction worker vital signs monitoring bracelet according to claim 2, characterized in that: Conducting a preliminary local health status analysis means: The average value, maximum value, and minimum value of the basic statistics of each vital sign data are obtained, and then the difference between the maximum value and the minimum value of the target vital sign data is calculated, and the average value is compared with the preset standard numerical range of the target vital sign data. If the difference satisfies the condition that the difference is less than or equal to the preset standard fluctuation threshold of the target vital sign data and the average value is within the preset standard numerical range of the target vital sign data, the target vital sign data is marked as normal data; if the difference does not satisfy the condition that the difference is less than or equal to the preset standard fluctuation threshold of the target vital sign data and the average value is within the preset standard numerical range of the target vital sign data, the target vital sign data is marked as abnormal data.

4. A construction worker vital signs monitoring bracelet according to claim 3, characterized in that: According to the results of the preliminary health status analysis, it is decided to transmit the preliminary processing results to the cloud analysis module or send different levels of local warning signals to the local warning module: When there is no vital sign data marked as abnormal data, the preliminary processing results are transmitted to the cloud analysis module. When there is vital sign data marked as abnormal data, the total number m of vital sign data types marked as abnormal data and the total number n of vital sign data types marked as normal data are counted, and the proportion p of abnormal data is calculated, p=m / (n+m). There is a preset proportion range interval-warning level form, and a local warning signal of the corresponding level is issued to the local warning module according to the proportion range interval into which the proportion p of abnormal data falls.

5. A construction worker vital signs monitoring bracelet according to claim 4, characterized in that: The cloud analysis module determines whether environmental factors need to be considered based on the decision tree algorithm and the construction personnel's working environment information: The environment monitoring module collects the construction workers' working environment information monitored in real time by a variety of preset environmental sensors to obtain an environmental data set, which consists of multiple environmental data parameters; Screening multiple environmental data parameters to obtain a main parameter and multiple subsidiary parameters; Set the root node and branch nodes of the decision tree: the root node uses the main parameters as the judgment condition, and the branch node uses the subsidiary parameters as the judgment condition. Then set the decision conditions of the root node and the branch node. The decision tree will output whether environmental factors need to be considered as part of the vital signs analysis. If the final output result is "yes", the environmental factors will be taken into consideration when analyzing the vital signs data in the cloud. If the final output result is "no", the environmental factors will not be taken into consideration when analyzing the vital signs data in the cloud.

6. A construction worker vital signs monitoring bracelet according to claim 5, characterized in that: Screening multiple environmental data parameters to obtain a main parameter and multiple subsidiary parameters means: The correlation coefficients of all environmental data parameters in the environmental data set with the possibility of environmental heat stress are calculated respectively, and the environmental data parameters are sorted from high to low according to the absolute value of the correlation coefficient. After sorting, the parameters with the strongest correlation with the possibility of environmental heat stress are selected as the main parameters, and then the correlation coefficients of the remaining environmental data parameters are compared with the preset screening threshold. If the absolute value of the correlation coefficient of the environmental data parameter is less than the preset screening threshold, it is marked as a useless parameter and eliminated. If the absolute value of the correlation coefficient of the environmental data parameter is not less than the preset screening threshold, it is marked as an auxiliary parameter and retained.

7. A construction worker vital signs monitoring bracelet according to claim 6, characterized in that: The correlation coefficient used was the Pearson correlation coefficient or the Spearman rank correlation coefficient.

8. A construction worker vital signs monitoring bracelet according to claim 7, characterized in that: Cloud analytics refers to: When environmental factors are taken into consideration, the basic statistics, main parameters, and subsidiary parameters of all preset vital sign data are summarized to obtain an input vector X, and then the input vector X is input into the pre-trained neural network model 1, and the neural network model 1 outputs a risk value 1; When environmental factors are not taken into consideration, the basic statistics of all preset vital signs data are summarized to obtain an input vector P, and then the input vector P is input into the pre-trained neural network model 2, and the neural network model 2 outputs a risk value 2.

9. A construction worker vital signs monitoring bracelet according to claim 8, characterized in that: The cloud warning module is used to decide whether to perform a cloud positioning warning operation based on the cloud analysis results. The cloud warning module receives risk value one or risk value two, compares risk value one or risk value two with the corresponding preset risk warning threshold. If it exceeds the corresponding preset risk warning threshold, the location information of the monitoring bracelet is obtained and a cloud warning is issued.

10. A method for monitoring vital signs of construction workers, based on a construction worker vital signs monitoring bracelet according to any one of claims 1 to 9, characterized in that: The following steps are involved: The vital signs of construction workers are monitored in real time through various preset vital signs sensors, and the working environment of construction workers is monitored in real time through various preset environmental sensors; Preliminary processing of the vital signs of construction workers monitored in real time and preliminary health status analysis locally, then based on the results of the preliminary health status analysis, decide whether to transmit the preliminary processing results to the cloud for further analysis or send local warning signals of different levels, and then execute the preset local warning strategy according to the level of the local warning signal; Based on the decision tree algorithm and the construction personnel's working environment information, it is determined whether environmental factors need to be considered, and cloud analysis operations are performed based on the judgment results. Then, based on the cloud analysis results, it is decided whether to perform cloud positioning and early warning operations.

Images