Intelligent staff health monitoring and management system based on multi-data fusion

Abstract

The invention relates to the technical field of data monitoring and management, in particular to an intelligent staff health monitoring and management system based on multi-data fusion, which is provided with a data acquisition module, a data identification module, a time domain analysis module, a data analysis module and a data verification module. A heart rate index fluctuation curve is constructed through a data recognition module, a feature time domain monitoring section is marked through a time domain analysis module, whether the staff has health risk tendency or not is judged through the data analysis module, the feature monitoring moment is determined, and the state tendency category of the staff is judged. A data verification mode is selected through a data verification module to judge whether the staff has health risks or not, then feature time domain monitoring sections which are prone to triggering misjudgment are dynamically recognized according to the staff base line, and the health data verification mode is adaptively adjusted according to the state tendency of the staff; and the efficiency and the reliability of the staff health intelligent monitoring and management system are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data monitoring management, and particularly relates to an intelligent worker health monitoring and management system based on multi-data fusion. BACKGROUND

[0002] With the increasing attention of society to occupational health and safety, and the deepening of the concept of corporate humanistic care, worker health management has become an important part of modern enterprise management. In recent years, with the popularity of wearable devices, worker health monitoring systems based on smart bands, watches and other devices have emerged. These systems can usually continuously collect physiological and behavioral parameters such as heart rate and step count of workers, and perform early warning by setting fixed thresholds. Current monitoring systems rely on a single physiological parameter or a single dimension of data for health determination, and cannot distinguish between normal physiological fluctuations and health risks. Moreover, using fixed thresholds for comparison cannot adaptively adjust the thresholds for different users, resulting in high misjudgment rate, insufficient early warning accuracy, extensive time domain analysis, difficulty in positioning critical risk periods, and low monitoring efficiency, which affects the reliability of the worker health intelligent monitoring and management system. Therefore, improving the reliability and monitoring and management efficiency of the worker health intelligent monitoring and management system is a technical problem to be solved.

[0003] For example, Chinese patent application publication No. CN118822313A discloses an occupational health and safety monitoring and early warning system based on big data, which relates to the technical field of information management. It solves the problems of high error rate in worker health trend judgment and low accuracy in health state prediction. The occupational health and safety monitoring and early warning system includes an inspection module, a collection module, a processing module, an evaluation module, and an early warning module. It uses a health fusion monitoring model to monitor real-time various non-destructive data chains combined with working environment and physiological characteristics to obtain occupational health and safety prediction results, improving the accuracy of worker health trend judgment and the accuracy of worker health prediction.

[0004] The existing technology also has the following problems: The existing technology does not consider that workers may trigger misjudgment due to temporary and situational normal physiological fluctuations, affecting the reliability of the worker health intelligent monitoring and management system. The existing technology cannot dynamically identify feature time domain monitoring segments that are prone to trigger misjudgment based on worker baseline, and cannot adaptively adjust health data verification methods based on worker state tendency, affecting the efficiency and reliability of the worker health intelligent monitoring and management system. SUMMARY

[0005] To this end, the application provides an employee health intelligent monitoring and management system based on multi-data fusion to overcome the problem that the prior art cannot dynamically identify feature time domain monitoring segments that are prone to trigger false positives according to the employee baseline, cannot adaptively adjust the health data verification method according to the state tendency of the employee, and affects the efficiency and reliability of the employee health intelligent monitoring and management system.

[0006] To achieve the above-mentioned purpose, the application provides an employee health intelligent monitoring and management system based on multi-data fusion, comprising: A data acquisition module is configured to acquire health parameters of an employee at a plurality of monitoring time points, wherein the health parameters include heart rate parameters and displacement parameters. A data recognition module is connected to the data acquisition module and configured to construct a plurality of heart rate index fluctuation curves based on the heart rate parameters at a plurality of monitoring time points within a preset monitoring period, and divide the time domain in which each heart rate index fluctuation curve is located into a plurality of time domain monitoring segments after time domain alignment. A time domain analysis module is connected to the data recognition module and configured to compare the heart rate index fluctuation curve segments within the time domain monitoring segments to mark feature time domain monitoring segments. A data analysis module is connected to the data acquisition module and the time domain analysis module, respectively, and configured to determine whether the employee has a health risk tendency based on the real-time heart rate parameter fluctuation in the feature time domain monitoring segments, in response to the existence of a health risk tendency, fit real-time health curve segments based on a plurality of health parameters to determine feature monitoring time points, and determine the state tendency category of the employee based on the comparison between the feature monitoring time points. A data verification module is connected to the data analysis module and configured to select a data verification method according to the state tendency category to determine whether the employee has a health risk, wherein the data verification method includes acquiring a feature collection window based on the feature monitoring time points to determine a state backsliding representation parameter, or determining an abnormal tendency parameter based on the real-time health curve segments.

[0007] Further, the time domain analysis module is configured to acquire clustering representation parameters, wherein The time domain analysis module acquires a plurality of heart rate index fluctuation curve segments within the same time domain monitoring segment, calculates the average degree of coincidence between the heart rate index fluctuation curve segments, and determines the average degree of coincidence as the clustering representation parameter of the time domain monitoring segment. The heart rate index fluctuation curve is constructed with the heart rate parameter as the vertical axis and the time as the horizontal axis.

[0008] Further, the time domain analysis module is configured to mark feature time domain monitoring segments based on the clustering representation parameters, wherein The time domain analysis module labels the time domain monitoring segment as a feature time domain monitoring segment based on a determination result that the clustering characteristic parameter of the time domain monitoring segment does not exceed a preset clustering characteristic parameter threshold.

[0009] Further, the data analysis module is configured to determine whether the employee has a health risk tendency, wherein, The data analysis module determines that the employee has a health risk tendency based on a determination result that the fluctuation of the real-time heart rate parameter in the feature time domain monitoring segment meets a health risk tendency condition. The health risk tendency condition is that the difference between the maximum value and the minimum value of the real-time heart rate parameter in the feature time domain monitoring segment exceeds a preset difference threshold.

[0010] Further, the data analysis module is configured to fit real-time health curve segments according to a plurality of health parameters to determine feature monitoring time points, wherein, The data analysis module is configured to obtain health parameters of a plurality of monitoring time points in real time within the feature time domain monitoring segment to construct real-time heart rate health curve segments and real-time displacement health curve segments, respectively. The data analysis module is configured to determine a first feature monitoring time point corresponding to a peak value in the real-time heart rate health curve segment, and determine a second feature monitoring time point corresponding to a peak value in the real-time displacement health curve segment. The real-time heart rate health curve is constructed with the heart rate parameter as the vertical axis and time as the horizontal axis, and the real-time displacement health curve is constructed with the displacement parameter as the vertical axis and time as the horizontal axis.

[0011] Further, the data analysis module is configured to determine the state tendency category of the employee in the feature time domain monitoring segment, wherein, The data analysis module determines that the state tendency category of the employee in the feature time domain monitoring segment is an active state tendency category based on a determination result that the comparison between the feature monitoring time points of the employee in the feature time domain monitoring segment meets an active state tendency condition. The data analysis module determines that the state tendency category of the employee in the feature time domain monitoring segment is a non-active state tendency category based on a determination result that the comparison between the feature monitoring time points of the employee in the feature time domain monitoring segment does not meet the active state tendency condition.

[0012] Further, the active state tendency condition is that the time interval between the first feature monitoring time point and the second feature monitoring time point does not exceed a preset time interval threshold, and the second feature monitoring time point is not later than the first feature monitoring time point in time sequence.

[0013] Further, the data verification module is configured to select a data verification method according to the state tendency category, wherein, The data verification module, based on the determination result that the state tendency category is the active state tendency category, selects a feature acquisition window based on the feature monitoring time to determine the state fallback characterization parameters, and determines whether the employee has a health risk. The data verification module, based on the determination result that the state tendency category is inactive, selects an abnormal tendency parameter based on the real-time health curve segment to determine whether the employee has a health risk.

[0014] Furthermore, the data verification module is used to determine whether the employee has a health risk based on the state fallback characterization parameters, wherein, The data verification module determines that the employee has a health risk based on the judgment result that the state fallback characterization parameter exceeds the preset state fallback characterization parameter threshold; The state fallback characterization parameter is the variance of the heart rate parameter within the feature acquisition window. The feature acquisition window starts at the first feature monitoring time and lasts for a preset duration.

[0015] Furthermore, the data verification module is used to determine whether the employee has a health risk based on the abnormal tendency parameter, wherein, The data verification module determines that the employee has a health risk based on the judgment result that the abnormal tendency parameters meet the health risk conditions; The health risk condition is that the first abnormal tendency parameter exceeds the preset first abnormal tendency parameter threshold, and the second abnormal tendency parameter exceeds the preset second abnormal tendency parameter threshold. The first abnormal tendency parameter is the number of inflection points on the real-time heart rate health curve segment, and the second abnormal tendency parameter is the variance of the absolute value of the difference in heart rate parameters between adjacent inflection points.

[0016] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention sets up a data acquisition module, a data identification module, a time-domain analysis module, a data analysis module, and a data verification module. The data acquisition module acquires employees' health parameters; the data identification module constructs several heart rate fluctuation curves; the time-domain analysis module marks characteristic time-domain monitoring segments; the data analysis module determines whether employees have health risk tendencies; real-time health curve segments are fitted to determine characteristic monitoring times and the employee's state tendency category; and the data verification module selects a data verification method to determine whether employees have health risks. Thus, it achieves dynamic identification of characteristic time-domain monitoring segments prone to misjudgment based on employee baselines and adaptive adjustment of health data verification methods according to employee state tendencies, improving the efficiency and reliability of the intelligent employee health monitoring and management system.

[0017] In particular, this invention uses a time-domain analysis module to compare the fluctuation curves of various heart rate indicators within a time-domain monitoring segment to mark characteristic time-domain monitoring segments. It is understood that, under the same working environment, employees' heart rate fluctuations usually show high similarity. When the clustering representation parameter of each heart rate curve within a certain time-domain monitoring segment is low, it indicates an abnormal situation that deviates from the normal fluctuation pattern of the group. These abnormalities may be caused by individualized normal physiological responses such as temporary activities, or they may be early manifestations of real health risks such as autonomic nervous system dysfunction. The single threshold method cannot distinguish between these two essentially different situations. By identifying characteristic time-domain monitoring segments and further analyzing them, accurate target intervals are provided for subsequent analysis, reducing invalid data analysis and improving system operating efficiency. Thus, it realizes the dynamic identification of characteristic time-domain monitoring segments that are prone to misjudgment based on employee baselines, thereby improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0018] In particular, this invention determines the employee's state tendency category based on the comparison between feature monitoring times through a data analysis module. This means that by using the causal temporal relationship between behavior and physiological heart rate to distinguish active states, the system categorizes employee state tendencies into active and inactive states. This allows the system to adopt the most suitable verification strategy, reducing the risk of misjudgment, optimizing resource allocation, avoiding resource waste caused by using the same discrimination standard for different states, improving computational efficiency, providing a scientific basis for personalized health management, reducing the risk of misjudgment, capturing health risks in different states, avoiding key missed judgments, adapting to multiple job scenarios, and improving universality. Ultimately, this invention achieves the determination of state tendency categories based on the causal temporal relationship between employee behavior and physiological heart rate, improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0019] In particular, this invention, through a data verification module, determines the state decline characterization parameters by acquiring feature acquisition windows based on feature monitoring times under the condition of an active state tendency category, thereby determining whether employees have health risks. It is understood that under the active state tendency category condition, the focus is on analyzing the heart rate recovery characteristics after exercise, reducing the misjudgment rate in the active state, and reducing misjudgments caused by the inability of traditional methods to distinguish between normal exercise responses and real risks. By quantifying the stability of the recovery period rather than simply focusing on the absolute value of heart rate, it effectively identifies potential risks that appear to have normal heart rate recovery but are actually abnormal in the recovery process. Judging active state risks by the magnitude of heart rate recovery can easily miss hidden risks that meet the magnitude but have large fluctuations in the process. Under the active state tendency, there may be slight fluctuations in the heart rate recovery process. If fluctuation is judged as abnormal, it is easy to make misjudgments. Thus, it realizes the adaptive adjustment of health data verification methods according to the employee's state tendency, improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0020] In particular, this invention, through a data verification module, determines abnormal tendency parameters based on real-time health curve segments to assess whether employees face health risks when the state tendency category is inactive. Under the inactive state tendency category, by analyzing two key abnormal tendency parameters—the first and second abnormal tendency parameters—of the real-time heart rate health curve segment, it is understood that in a healthy inactive state, normal heart rate fluctuations are regulated by the autonomic nervous system, resulting in smooth, rhythmic curve changes. The first abnormal tendency parameter (the number of inflection points) is moderate, and the second abnormal tendency parameter (the fluctuation amplitude) is relatively stable. However, when autonomic nervous system dysfunction or potential cardiovascular risk exists, the heart rate regulation mechanism becomes uncontrolled, leading to frequent, irregular, and violent fluctuations in the curve. The first and second abnormal tendency parameters increase simultaneously. By setting dual threshold conditions for the first and second abnormal tendency parameters, health risks in the inactive state are accurately captured. This allows for adaptive adjustment of health data verification methods based on the employee's state tendency, improving the efficiency and reliability of the employee health intelligent monitoring and management system. Attached Figure Description

[0021] Figure 1 This is a functional block diagram of the employee health intelligent monitoring and management system based on multi-data fusion, as described in an embodiment of the present invention. Figure 2 This is a flowchart illustrating the logic of the time-domain monitoring segment for marking features in the time-domain analysis module according to an embodiment of the present invention. Figure 3 This is a flowchart illustrating the logic of the data analysis module in this embodiment of the invention for determining whether an employee has a health risk tendency. Figure 4 This is a flowchart illustrating the logic of the data analysis module in this embodiment of the invention for determining the state tendency category of employees within a specific time domain monitoring period. Detailed Implementation

[0022] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0023] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0024] It should be noted that in the description of this invention, the terms "upper," "lower," "inner," "outer," etc., which indicate the direction or positional relationship, are based on the direction or positional relationship shown in the drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0025] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0026] Please see Figure 1 The diagram shown is a functional block diagram of the employee health intelligent monitoring and management system based on multi-data fusion according to an embodiment of the present invention. The employee health intelligent monitoring and management system based on multi-data fusion of the present invention includes: The data acquisition module is used to acquire the health parameters of employees at several monitoring times, including heart rate parameters and displacement parameters. Specifically, the embodiments of the present invention do not impose specific limitations on the structure of the data acquisition module. Preferably, it can be a smart wearable device such as a smart bracelet or watch to acquire health parameters, with the heart rate parameter being the heart rate value and the displacement parameter being the acceleration value. Of course, other methods can also be used, which will not be elaborated here.

[0027] The data identification module, which is connected to the data acquisition module, is used to construct several heart rate index fluctuation curves based on heart rate parameters at several monitoring times within a preset monitoring period. After time domain alignment, the time domain of each heart rate index fluctuation curve is divided into several time domain monitoring segments. Specifically, the embodiments of the present invention do not specifically limit the structure of the data recognition module. Preferably, it can be a microprocessor to construct several heart rate index fluctuation curves and divide the time domain monitoring segments. Of course, other methods can also be used, which will not be elaborated here.

[0028] Specifically, there are no restrictions on the method for constructing the heart rate index fluctuation curve. For example, the heart rate index fluctuation curve can be fitted using MATLAB correlation fitting software, which will not be elaborated further.

[0029] Specifically, the preset monitoring period, the interval between adjacent monitoring times, and the division duration of the time domain monitoring segment can be set by those skilled in the art according to the accuracy requirements of the employee health monitoring and management system. The higher the accuracy requirement, the shorter the setting. The preset monitoring period can be in the range of [12, 36], with the interval unit being h. Preferably, it can be 24h. The interval duration can be in the range of [10, 30], with the interval unit being s. Preferably, it can be 20s. The division duration of the time domain monitoring segment can be in the range of [20, 40], with the interval unit being min. Preferably, it can be 30min. Ten heart rate index fluctuation curves can be constructed to mark the characteristic time domain monitoring segments.

[0030] The time-domain analysis module, which is connected to the data recognition module, is used to compare the fluctuation curve segments of each heart rate index within the time-domain monitoring segment to mark the characteristic time-domain monitoring segment. Specifically, the embodiments of the present invention do not impose specific limitations on the structure of the time-domain analysis module. Preferably, it can be a microprocessor used to mark the feature time-domain monitoring segment. Of course, other methods can also be used, which will not be elaborated here.

[0031] The data analysis module is connected to the data acquisition module and the time domain analysis module respectively. It is used to determine whether the employee has a health risk tendency based on the real-time heart rate parameter fluctuation within the characteristic time domain monitoring segment. In response to the existence of a health risk tendency, it fits real-time health curve segments according to several health parameters to determine the characteristic monitoring time. It determines the employee's state tendency category based on the comparison between the characteristic monitoring times. Specifically, the embodiments of the present invention do not specifically limit the structure of the data analysis module. Preferably, it can be a processor used in a computer to determine whether an employee has a health risk tendency and to determine the employee's state tendency category. Of course, other methods can also be used, which will not be elaborated here.

[0032] The data verification module, which is connected to the data analysis module, is used to select a data verification method based on the state tendency category to determine whether the employee has a health risk. The data verification method includes obtaining a feature acquisition window based on the feature monitoring time to determine the state decline characterization parameter, or determining the abnormal tendency parameter based on the real-time health curve segment.

[0033] Specifically, the embodiments of the present invention do not impose specific limitations on the structure of the data verification module. Preferably, it can be a microprocessor used to determine whether an employee has a health risk. Of course, other methods can also be used, which will not be elaborated here.

[0034] Specifically, the time-domain analysis module is used to obtain clustering characterization parameters, wherein, The time-domain analysis module acquires several heart rate index fluctuation curve segments within the same time-domain monitoring segment, calculates the average overlap between each heart rate index fluctuation curve segment, and determines the average overlap as the clustering characterization parameter of the time-domain monitoring segment. The heart rate fluctuation curve is constructed with heart rate parameter as the vertical axis and time as the horizontal axis.

[0035] Specifically, there is no limitation on the method for determining the degree of overlap. For example, the cosine similarity method can be used to determine it. After vectorizing the heart rate index fluctuation curve segment, the cosine similarity is calculated, and the obtained cosine similarity is determined as the degree of overlap. Of course, other methods can also be used, which will not be elaborated here.

[0036] Please see Figure 2 As shown, this is a logical flowchart of the time-domain analysis module marking feature time-domain monitoring segments according to an embodiment of the present invention. The time-domain analysis module is used to mark feature time-domain monitoring segments based on the clustering representation parameters, wherein... The time-domain analysis module marks the time-domain monitoring segment as a feature time-domain monitoring segment based on the determination result that the clustering representation parameters of the time-domain monitoring segment do not exceed the preset clustering representation parameter threshold. The time-domain analysis module does not mark the time-domain monitoring segment based on the determination result that the clustering representation parameters of the time-domain monitoring segment exceed the preset clustering representation parameter threshold.

[0037] Specifically, the preset clustering representation parameter threshold is the product of the clustering representation parameter reference value and the clustering factor. The clustering representation parameter reference value is the mean value of the clustering representation parameters of medical guidelines that meet the relevant requirements in historical data. The clustering factor can be set by those skilled in the art according to the accuracy requirements of the employee health monitoring and management system. The higher the accuracy requirement, the larger the value should be. The value range can be [1.1, 1.2], and preferably, it can be 1.15.

[0038] Specifically, this embodiment of the invention uses a time-domain analysis module to compare the fluctuation curves of various heart rate indicators within a time-domain monitoring segment to mark characteristic time-domain monitoring segments. It is understood that, under the same working environment, the heart rate fluctuations of employees usually show a high degree of similarity. When the clustering representation parameter of each heart rate curve within a certain time-domain monitoring segment is low, it indicates that there is an abnormal situation that deviates from the normal fluctuation pattern of the group. These abnormalities may be caused by individualized normal physiological responses such as temporary activities, or they may be real health risks such as early manifestations of autonomic nervous system dysfunction. The single threshold method cannot distinguish between these two essentially different situations. By identifying characteristic time-domain monitoring segments and further analyzing them, accurate target intervals are provided for subsequent analysis, reducing invalid data analysis and improving system operating efficiency. Thus, it realizes the dynamic identification of characteristic time-domain monitoring segments that are prone to misjudgment based on the employee baseline, thereby improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0039] Specifically, the clustering representation parameter, which is the average overlap of several heart rate fluctuation curve segments within the same time domain monitoring period, can represent the stability and consistency of heart rate fluctuations during that period. A higher clustering representation parameter indicates a more consistent heart rate trend and a lower possibility of misjudgment. Conversely, a lower clustering representation parameter indicates a more significant difference in heart rate trends and an unstable state. In this case, heart rate fluctuations lose their regularity and may be caused by short-term normal factors or health risks, making direct judgment difficult and prone to misjudgment. This provides a precise analysis target for subsequent multi-data fusion verification. Coarse analysis across the entire time period can easily waste computing power in stable time domain segments or lead to missed / misjudged cases in ambiguous time domain segments due to a lack of focus. After marking, these time domain segments can be precisely focused on, and multi-dimensional data can be called for verification analysis to distinguish between normal and risky conditions, reducing misjudgment. Thus, it enables the dynamic identification of characteristic time domain monitoring segments that are prone to misjudgment based on employee baselines, improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0040] Please see Figure 3 The diagram shown is a flowchart illustrating the logic of the data analysis module in this embodiment of the invention for determining whether an employee has a predisposition to health risks. The data analysis module is used to determine whether the employee has a predisposition to health risks. The data analysis module determines that the employee has a health risk tendency based on the judgment result that the real-time heart rate parameter fluctuation within the characteristic time domain monitoring segment meets the health risk tendency conditions. Based on the determination result that the real-time heart rate parameter fluctuations within the characteristic time domain monitoring segment do not meet the conditions for health risk tendency, it is determined that the employee does not have a health risk tendency. The health risk propensity condition is that the difference between the maximum and minimum heart rate parameters in real time within a specific time domain monitoring segment exceeds a preset difference threshold.

[0041] Specifically, the preset difference threshold is the product of the difference reference value and the regulation factor. The difference reference value is the average difference of medical guidelines that meet the relevant requirements in historical data. The regulation factor can be set by those skilled in the art according to the accuracy requirements of the employee health monitoring and management system. The higher the accuracy requirement, the smaller the value should be. The value range can be [1.05, 1.15], preferably 1.1.

[0042] Specifically, the data analysis module is used to fit real-time health curve segments based on several health parameters to determine the feature monitoring time, wherein, The data analysis module is used to acquire health parameters at several monitoring times in real time within a characteristic time domain monitoring segment, so as to construct real-time heart rate health curve segment and real-time displacement health curve segment respectively. The time corresponding to the peak value within the real-time heart rate health curve segment is used to determine the first feature monitoring time, and the time corresponding to the peak value within the real-time displacement health curve segment is used to determine the second feature monitoring time. The real-time heart rate health curve is constructed with heart rate parameters as the vertical axis and time as the horizontal axis, and the real-time displacement health curve is constructed with displacement parameters as the vertical axis and time as the horizontal axis.

[0043] Please see Figure 4 As shown, this is a flowchart illustrating the logic of the data analysis module in this embodiment of the invention for determining the state tendency category of an employee within a specific time domain monitoring period. The data analysis module is used to determine the state tendency category of an employee within the specific time domain monitoring period. The data analysis module determines that the employee's state tendency category within the characteristic time domain monitoring period is an active state tendency category based on the comparison between characteristic monitoring times within the characteristic time domain monitoring period. The data analysis module determines that the employee's state tendency category within the characteristic time domain monitoring period is an inactive state tendency category based on the judgment result that the comparison between characteristic monitoring times within the characteristic time domain monitoring period does not meet the active state tendency condition.

[0044] Specifically, the active state tendency condition is that the time interval between the first feature monitoring time and the second feature monitoring time does not exceed a preset time interval threshold, and the second feature monitoring time does not exceed the first feature monitoring time in terms of timing.

[0045] Specifically, the second feature monitoring time is no more than the first feature monitoring time in terms of timing, that is, the second feature monitoring time is earlier than or equal to the first feature monitoring time in terms of time.

[0046] Specifically, the preset time interval threshold is the product of the time interval reference value and the time factor. The time interval reference value is the average time interval of medical guidelines that meet the relevant requirements in historical data. The time factor can be set by those skilled in the art based on the average of several historical experimental data, and the value range can be [1.1, 1.2]. Preferably, it can be 1.15.

[0047] Specifically, this embodiment of the invention determines the employee's state tendency category based on the comparison between feature monitoring times through a data analysis module. This means that by utilizing the causal temporal relationship between behavior and physiological heart rate to distinguish active states, the employee's state tendency is categorized into active and inactive states. This allows the system to adopt the most suitable verification strategy, reducing the risk of misjudgment, optimizing resource allocation, avoiding resource waste caused by using the same discrimination standard for different states, improving computational efficiency, providing a scientific basis for personalized health management, reducing the risk of misjudgment, capturing health risks in different states, avoiding key missed judgments, adapting to multiple job scenarios, and improving universality. Thus, it achieves the determination of state tendency category based on the causal temporal relationship between employee behavior and physiological heart rate, improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0048] Specifically, it can be understood that when physical activity causes changes in heart rate, normal physiological responses exhibit a clear causal temporal relationship. The displacement peak, i.e., the second characteristic monitoring moment, can characterize the moment of maximum physical activity intensity, while the heart rate peak, i.e., the first characteristic monitoring moment, can characterize the moment of maximum cardiovascular system response. Under normal physiological conditions, when employees are in an active state, such as carrying materials, walking briskly, or operating on the production line, their actions directly lead to increased oxygen consumption, which in turn causes an increase in heart rate. The activity intensity peak should precede or be synchronous with the heart rate response peak, and the time interval between the two should be within a reasonable range. By understanding the basic temporal patterns of human physiological responses, different state tendencies can be distinguished, providing data support for subsequent targeted health verification. Thus, it is possible to determine the state tendency category based on the causal temporal relationship between employee behavior and physiological heart rate, thereby improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0049] Specifically, the data verification module is used to select a data verification method based on the state tendency category, wherein, The data verification module, based on the determination result that the state tendency category is the active state tendency category, selects a feature acquisition window based on the feature monitoring time to determine the state fallback characterization parameters, and determines whether the employee has a health risk. The data verification module, based on the determination result that the state tendency category is inactive, selects an abnormal tendency parameter based on the real-time health curve segment to determine whether the employee has a health risk.

[0050] Specifically, the data verification module is used to determine whether the employee has a health risk based on the state fallback characterization parameters, wherein, The data verification module determines that the employee has a health risk based on the judgment result that the state fallback characterization parameter exceeds the preset state fallback characterization parameter threshold; Based on the determination result that the state fallback characterization parameter does not exceed the preset state fallback characterization parameter threshold, it is determined that the employee does not have a health risk; The state fallback characterization parameter is the variance of the heart rate parameter within the feature acquisition window. The feature acquisition window starts at the first feature monitoring time and lasts for a preset duration.

[0051] Specifically, the preset threshold for the state fallback characterization parameter is the product of the state fallback characterization parameter reference value and the fallback factor. The state fallback characterization parameter reference value is the average state fallback characterization parameter of medical guidelines that meet the relevant requirements in historical data. The fallback factor can be set by those skilled in the art according to the accuracy requirements of the employee health monitoring and management system. The higher the accuracy requirement, the smaller the value should be. The value range can be [1.05, 1.15], preferably 1.1. The duration of the feature acquisition window can be set by those skilled in the art according to the average of several historical experimental data. The value range can be [1, 3], and the interval unit is min. Preferably, it can be 2 min.

[0052] Specifically, this embodiment of the invention uses a data verification module to determine the state decline characterization parameters based on the feature acquisition window obtained at the feature monitoring time when the state tendency category is active, thereby determining whether an employee has a health risk. Under the active state tendency category, the focus is on analyzing the heart rate recovery characteristics after exercise, reducing the misjudgment rate in active states and minimizing misjudgments caused by the inability of traditional methods to distinguish between normal exercise responses and actual risks. By quantifying the stability of the recovery period rather than simply focusing on the absolute heart rate value, it effectively identifies potential risks that appear to have normal heart rate recovery but are actually abnormal in the recovery process. Judging active state risks based on the magnitude of heart rate recovery can easily miss hidden risks where the magnitude meets the standard but the process fluctuates greatly. Under the active state tendency, the heart rate recovery process may have slight fluctuations; judging abnormality based on fluctuations can easily lead to misjudgments. This embodiment of the invention uses a data verification module to determine the state decline characterization parameters based on the feature acquisition window obtained at the feature monitoring time when the state tendency category is active, thereby determining whether an employee has a health risk. Furthermore, it achieves adaptive adjustment of the health data verification method according to the employee's state tendency, improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0053] Specifically, it can be understood that after determining whether an employee is in an active state, a feature acquisition window is set with the peak heart rate time, i.e., the first feature monitoring time, as the starting point. The variance of the heart rate parameter within this feature acquisition window is calculated as a state recovery characterization parameter. Based on the core mechanism of exercise physiology, in healthy individuals, after the peak exercise time, the autonomic nervous system is rapidly activated, causing the heart rate to recover in a stable and monotonous manner, usually manifested as a low heart rate variance. However, when there are abnormalities in cardiovascular function or impaired autonomic nervous regulation, the heart rate recovery process will be disordered, fluctuating, or unstable, usually manifested as a high heart rate variance. This verification method can effectively distinguish between normal exercise-induced heart rate fluctuations and abnormal pathophysiological responses. Thus, it enables the adaptive adjustment of health data verification methods according to the employee's state tendency, improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0054] Specifically, the data verification module is used to determine whether the employee has a health risk based on the abnormal tendency parameter, wherein, The data verification module determines that the employee has a health risk based on the judgment result that the abnormal tendency parameters meet the health risk conditions; Based on the determination result that the abnormal tendency parameters do not meet the health risk conditions, it is determined that the employee does not have a health risk. The health risk condition is that the first abnormal tendency parameter exceeds the preset first abnormal tendency parameter threshold, and the second abnormal tendency parameter exceeds the preset second abnormal tendency parameter threshold. The first abnormal tendency parameter is the number of inflection points on the real-time heart rate health curve segment, and the second abnormal tendency parameter is the variance of the absolute value of the difference in heart rate parameters between adjacent inflection points.

[0055] Specifically, the preset threshold for the first abnormal tendency parameter is the product of the reference value of the first abnormal tendency parameter and the first factor; the preset threshold for the second abnormal tendency parameter is the product of the reference value of the second abnormal tendency parameter and the second factor; the reference value for the first abnormal tendency parameter is the average value of the first abnormal tendency parameter of medical guidelines that meet the relevant requirements in historical data; the reference value for the second abnormal tendency parameter is the average value of the second abnormal tendency parameter of medical guidelines that meet the relevant requirements in historical data; the first factor and the second factor can be set by those skilled in the art according to the accuracy requirements of the employee health monitoring and management system. The higher the accuracy requirement, the smaller the value should be. The value range of the first factor can be [1.05, 1.15], preferably 1.1; the value range of the second factor can be [1.05, 1.15], preferably 1.1.

[0056] Specifically, in this embodiment of the invention, the data verification module determines abnormal tendency parameters based on real-time health curve segments under the condition of an inactive state tendency category to determine whether an employee has health risks. Under the inactive state tendency category, by analyzing two key abnormal tendency parameters of the real-time heart rate health curve segment—the first abnormal tendency parameter and the second abnormal tendency parameter—normal heart rate fluctuations in a healthy inactive state are balanced and regulated by the autonomic nervous system, resulting in smooth, rhythmic curve changes. The first abnormal tendency parameter (i.e., the number of inflection points) is moderate, and the second abnormal tendency parameter (i.e., the fluctuation amplitude) is relatively stable. However, when autonomic nervous system function is disordered or there is a potential cardiovascular risk, the heart rate regulation mechanism becomes uncontrolled, leading to frequent, irregular, and violent fluctuations in the curve. The first abnormal tendency parameter increases, and the second abnormal tendency parameter also increases. By setting dual threshold conditions for the first and second abnormal tendency parameters, health risks in the inactive state are accurately captured. This allows for adaptive adjustment of the health data verification method based on the employee's state tendency, improving the efficiency and reliability of the employee health intelligent monitoring and management system.

[0057] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0058] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A smart monitoring and management system for employee health based on multi-data fusion, characterized in that, include: The data acquisition module is used to acquire the health parameters of employees at several monitoring times, including heart rate parameters and displacement parameters. The data identification module, which is connected to the data acquisition module, is used to construct several heart rate index fluctuation curves based on heart rate parameters at several monitoring times within a preset monitoring period. After time domain alignment, the time domain of each heart rate index fluctuation curve is divided into several time domain monitoring segments. The time-domain analysis module, which is connected to the data recognition module, is used to compare the fluctuation curve segments of each heart rate index within the time-domain monitoring segment to mark the characteristic time-domain monitoring segment. The data analysis module is connected to the data acquisition module and the time domain analysis module respectively. It is used to determine whether the employee has a health risk tendency based on the real-time heart rate parameter fluctuation within the characteristic time domain monitoring segment. In response to the existence of a health risk tendency, it fits real-time health curve segments according to several health parameters to determine the characteristic monitoring time. It determines the employee's state tendency category based on the comparison between the characteristic monitoring times. The data verification module, which is connected to the data analysis module, is used to select a data verification method based on the state tendency category to determine whether the employee has a health risk. The data verification method includes obtaining a feature acquisition window based on the feature monitoring time to determine the state decline characterization parameter, or determining the abnormal tendency parameter based on the real-time health curve segment.

2. The employee health intelligent monitoring and management system based on multi-data fusion according to claim 1, characterized in that, The time-domain analysis module is used to obtain clustering characterization parameters, wherein, The time-domain analysis module acquires several heart rate index fluctuation curve segments within the same time-domain monitoring segment, calculates the average overlap between each heart rate index fluctuation curve segment, and determines the average overlap as the clustering characterization parameter of the time-domain monitoring segment. The heart rate fluctuation curve is constructed with heart rate parameter as the vertical axis and time as the horizontal axis.

3. The employee health intelligent monitoring and management system based on multi-data fusion according to claim 2, characterized in that, The time-domain analysis module is used to label characteristic time-domain monitoring segments based on the clustering characterization parameters, wherein, The time-domain analysis module marks the time-domain monitoring segment as a feature time-domain monitoring segment based on the determination result that the clustering representation parameters of the time-domain monitoring segment do not exceed the preset clustering representation parameter threshold.

4. The intelligent employee health monitoring and management system based on multi-data fusion according to claim 3, characterized in that, The data analysis module is used to determine whether the employee has a predisposition to health risks. The data analysis module determines that the employee has a health risk tendency based on the judgment result that the real-time heart rate parameter fluctuation within the characteristic time domain monitoring segment meets the health risk tendency conditions. The health risk propensity condition is that the difference between the maximum and minimum heart rate parameters in real time within a specific time domain monitoring segment exceeds a preset difference threshold.

5. The intelligent employee health monitoring and management system based on multi-data fusion according to claim 4, characterized in that, The data analysis module is used to fit real-time health curve segments based on several health parameters to determine the feature monitoring time, wherein, The data analysis module is used to acquire health parameters at several monitoring times in real time within a characteristic time domain monitoring segment, so as to construct real-time heart rate health curve segment and real-time displacement health curve segment respectively. The time corresponding to the peak value within the real-time heart rate health curve segment is used to determine the first feature monitoring time, and the time corresponding to the peak value within the real-time displacement health curve segment is used to determine the second feature monitoring time. The real-time heart rate health curve is constructed with heart rate parameters as the vertical axis and time as the horizontal axis, and the real-time displacement health curve is constructed with displacement parameters as the vertical axis and time as the horizontal axis.

6. The employee health intelligent monitoring and management system based on multi-data fusion according to claim 5, characterized in that, The data analysis module is used to determine the employee's state tendency category within the specified time domain monitoring period, wherein... The data analysis module determines that the employee's state tendency category within the characteristic time domain monitoring period is an active state tendency category based on the comparison between characteristic monitoring times within the characteristic time domain monitoring period. The data analysis module determines that the employee's state tendency category within the characteristic time domain monitoring period is an inactive state tendency category based on the judgment result that the comparison between characteristic monitoring times within the characteristic time domain monitoring period does not meet the active state tendency condition.

7. The intelligent monitoring and management system for employee health based on multi-data fusion according to claim 6, characterized in that, The active state tendency condition is that the time interval between the first feature monitoring time and the second feature monitoring time does not exceed a preset time interval threshold, and the second feature monitoring time does not exceed the first feature monitoring time in terms of time sequence.

8. The intelligent employee health monitoring and management system based on multi-data fusion according to claim 7, characterized in that, The data verification module is used to select a data verification method based on the state tendency category, wherein, The data verification module, based on the determination result that the state tendency category is the active state tendency category, selects a feature acquisition window based on the feature monitoring time to determine the state fallback characterization parameters, and determines whether the employee has a health risk. The data verification module, based on the determination result that the state tendency category is inactive, selects an abnormal tendency parameter based on the real-time health curve segment to determine whether the employee has a health risk.

9. The intelligent monitoring and management system for employee health based on multi-data fusion according to claim 8, characterized in that, The data verification module is used to determine whether the employee has a health risk based on the state fallback characterization parameters. The data verification module determines that the employee has a health risk based on the judgment result that the state fallback characterization parameter exceeds the preset state fallback characterization parameter threshold; The state fallback characterization parameter is the variance of the heart rate parameter within the feature acquisition window. The feature acquisition window starts at the first feature monitoring time and lasts for a preset duration.

10. The intelligent employee health monitoring and management system based on multi-data fusion according to claim 9, characterized in that, The data verification module is used to determine whether the employee has a health risk based on the abnormal tendency parameter. The data verification module determines that the employee has a health risk based on the judgment result that the abnormal tendency parameters meet the health risk conditions; The health risk condition is that the first abnormal tendency parameter exceeds the preset first abnormal tendency parameter threshold, and the second abnormal tendency parameter exceeds the preset second abnormal tendency parameter threshold. The first abnormal tendency parameter is the number of inflection points on the real-time heart rate health curve segment, and the second abnormal tendency parameter is the variance of the absolute value of the difference in heart rate parameters between adjacent inflection points.

Images