Intelligent supervision system for health of life and workplace workers

Abstract

The invention relates to the technical field of intelligent health supervision, in particular to an intelligent supervision system for health of life and workplace workers. Precise health assessment is realized through multi-dimensional data fusion. Environmental monitoring, video behavior recognition and personalized physiological data are integrated, and a three-dimensional assessment system covering a physical environment, a behavior state and physiological signs is constructed; closed-loop management is formed by means of intelligent early warning and an active intervention mechanism, multi-level early warning is triggered based on a dynamic threshold value, health management is changed from post-treatment to beforehand prevention, the probability of occurrence of health risks is effectively reduced, and key crowd protection is enhanced through a management personnel linkage mechanism; according to the invention, collaborative optimization of group health and working environment is realized: the working space is taken as a unit to count environment alarm times and employee health abnormal data, the influence of the environment on health is quantitatively evaluated, and the management efficiency of the working environment and employee health is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent health monitoring, and in particular to an intelligent monitoring system for the health of people in their daily lives and workplaces. Background Art

[0002] Smart health monitoring is the product of the deep integration of technologies such as the Internet of Things, big data, and artificial intelligence. It aims to achieve health management for individuals or groups through real-time monitoring, data analysis, and intelligent intervention. Although existing technologies have made progress, they still have significant shortcomings in the following aspects: Existing smart health monitoring systems often focus on a single dimension and lack the ability to integrate multi-source data. Furthermore, some health monitoring devices only record heart rate and blood pressure, failing to integrate environmental factors. They focus solely on temperature and humidity, ignoring the impact of light stability on eye fatigue.

[0003] In addition, the existing intelligent health monitoring system is unable to make intelligent judgments based on individual historical data and environmental changes. It only pushes simple reminders after the early warning and does not link the equipment to automatic adjustment.

[0004] Finally, existing smart health monitoring systems do not count environmental alarms and employee health abnormalities by workspace, making it impossible to quantify the impact of the environment on group health. In response to the above problems, it is necessary to propose an intelligent monitoring system for the health of people in their daily lives and workplaces. Summary of the Invention

[0005] The purpose of the present invention is to solve the problems existing in the background technology and to propose an intelligent monitoring system for the health of people in their daily lives and workplaces.

[0006] The purpose of the present invention can be achieved through the following technical solutions: An intelligent monitoring system for the health of people in their daily lives and workplaces includes an environmental monitoring module, a video surveillance and behavior recognition module, a personal health monitoring module, a data fusion and health assessment module, and an early warning and guidance module.

[0007] The environmental monitoring module collects workplace environmental data, including temperature, humidity, air pressure, and light intensity. It comprehensively evaluates these data to analyze the overall comfort level of the work environment. It also conducts a targeted analysis of light intensity, including both natural light and that from lighting fixtures, to analyze the overall lighting conditions of the work environment.

[0008] Get real-time temperature, humidity, air pressure, natural light intensity, and lighting intensity for each workplace.

[0009] Feature extraction is performed on real-time temperature, humidity and natural light intensity, and the real-time wet-bulb temperature and real-time black globe temperature of each workplace are calculated using a preset formula. The real-time wet-bulb temperature and real-time black globe temperature are calculated to obtain the real-time wet-bulb temperature of each workplace; the temperature and humidity index of each workplace is calculated by calculating the real-time temperature and humidity.

[0010] For real-time air pressure, obtain the local annual average air pressure and calculate the standard deviation of historical data; calculate the time gradient of air pressure, and calculate the air pressure correction factor and air pressure fluctuation index of each workplace through preset formulas.

[0011] As a preferred embodiment of the present invention, data fusion is performed on the real-time temperature and humidity index, real-time wet-bulb globe temperature and air pressure correction factor, and air pressure fluctuation index of each workplace, and a real-time comprehensive environmental assessment index is calculated using a preset formula.

[0012] As a preferred embodiment of the present invention, an accelerometer is accessed to obtain the real-time acceleration generated by the employee in the vertical direction and the real-time electromagnetic communication signal strength. Based on the real-time acceleration and electromagnetic communication signal strength, it is determined whether the employee is sitting in an elevator. The air pressure fluctuation index is numerically corrected to prevent misjudgment of abnormal air pressure fluctuations caused by entering and exiting the elevator. Obtain the real-time acceleration and the rate of change of the electromagnetic communication signal strength. If the real-time acceleration change rate is greater than a preset threshold, and the electromagnetic communication signal strength change rate is greater than a preset threshold, it is determined that the employee is in an elevator environment, and the value of the elevator environment correction factor at the current moment is set to 1; otherwise, the value of the elevator environment correction factor at time t is set to 0.

[0013] As a preferred embodiment of the present invention, a comprehensive analysis is conducted on the natural light intensity and the lighting intensity of the lighting facilities in each workplace, the illumination conditions and lighting stability of the lighting facilities in the working environment are analyzed, and it is evaluated whether the lighting environment provided by the lighting facilities can cooperate with natural light to provide a comfortable working environment.

[0014] Calculate the proportion of natural light in the total illuminance, the preset minimum illuminance of the working environment and the preset optimal comfortable illuminance, and use them as calculation parameters to calculate the basic illuminance efficiency characteristic value of each workplace through the preset formula.

[0015] Calculate the average value of the total light intensity from the current moment to the preset moment, and use it as a calculation parameter. Enter the preset formula together with the preset moment length, natural light intensity and light intensity of the lighting facility to calculate the light stability characteristic value.

[0016] The calculated real-time comprehensive environmental assessment index, basic illumination efficiency characteristic value and lighting stability characteristic value of each workplace are sent to the data fusion and health assessment module.

[0017] The video surveillance and behavior recognition module uses video surveillance equipment to identify staff members' mental fatigue, work concentration, and sedentary conditions based on facial recognition and motion analysis, and issues timely warnings.

[0018] The convolutional neural network model is used to identify and locate the staff, distinguish the staff in the video from the background environment, and then number the staff and recognize their faces.

[0019] The limb movement frequency, limb movement amplitude, and movement frequency of the staff identified by the convolutional neural network model are analyzed through a recurrent neural network to extract the movement frequency of the employees in the past hour.

[0020] The employee's facial recognition images are used for expression recognition and eye movement recognition through convolutional neural networks to extract the employee's eye movement frequency in the past hour.

[0021] Access the employee's computer to obtain behavioral data, including the employee's typing and click count per unit time, record the typing count per unit time as the first behavioral characteristic parameter, and record the click count per unit time as the second behavioral characteristic parameter; The movement frequency, eye movement frequency, first behavior characteristic parameter and second behavior characteristic parameter of each employee are sent to the data fusion and health assessment module.

[0022] The personal health monitoring module uses customized wearable devices to collect each employee's physiological data, including heart rate, blood oxygen levels, and body temperature. Based on this data, statistical features are extracted from historical data, a timeline is created, and real-time employee physiological data is visualized and anomalies are identified. Health examination information is obtained and, using a natural language processing model, characteristic phrases are captured to identify each employee's health status. Personalized physiological monitoring plans are then developed based on their health status.

[0023] The acquired physiological sign data include heart rate data, blood oxygen data and body temperature data.

[0024] As a preferred embodiment of the present invention, a statistical feature analysis is performed on the heart rate data, and the average values ​​and standard deviations of the heart rate data, blood oxygen data, and body temperature data on the day are calculated. The values ​​are input into a preset formula, and the absolute value of a preset multiple of the difference between the heart rate data, blood oxygen data, and body temperature data and their own average values ​​is calculated to obtain a first result of the heart rate data, blood oxygen data, and body temperature data. The first result is divided by the standard deviation of the heart rate data, blood oxygen data, and body temperature data to obtain a characteristic value of abnormal heart rate fluctuation, a characteristic value of blood oxygen fluctuation, and a characteristic value of body temperature fluctuation.

[0025] Data fusion is performed based on the abnormal heart rate fluctuation characteristic values, blood oxygen fluctuation characteristic values, and body temperature fluctuation characteristic values. These values ​​are weighted and summed to obtain real-time comprehensive vital sign characteristic values. The real-time heart rate data, blood oxygen data, body temperature data, and real-time comprehensive vital sign characteristic values ​​are sent to the data fusion and health assessment module.

[0026] As a preferred method of the present invention, health checkup information is obtained, and characteristic phrases therein are captured based on a natural language processing model to identify the health status of each employee. A personalized physiological sign monitoring plan is formulated based on their health status. The specific process is as follows: Access the health examination information system, perform word segmentation on the physical examination report, perform natural language processing, entity recognition and classification through the TF-IDF algorithm, and extract specific physical examination data items, including employees' height, weight, blood pressure, blood pressure, over-limit indicators in the physical examination data table, doctor's order notes, past medical history and family genetic disease history data.

[0027] All extracted physical examination data items are structured to generate a structured data table.

[0028] For employees whose keywords "hypertension, hyperglycemia, history of past illness, and family history of genetic diseases" are extracted from the excessive indicators, medical notes, past medical history, and family genetic disease history in the physical examination data sheet, they will be marked as centrally monitored employees, and targeted physiological sign monitoring plans will be formulated for all centrally monitored employees.

[0029] As a preferred embodiment of the present invention, for all centrally monitored employees, a monitoring schedule for physiological sign data is prepared based on the physiological sign data and their abnormal heart rate fluctuation characteristic values, blood oxygen fluctuation characteristic values ​​and body temperature fluctuation characteristic values. The heart rate data, blood oxygen data, body temperature data and real-time comprehensive vital sign characteristic values ​​at each moment are input into a two-dimensional rectangular coordinate system with time as the independent variable, and the preset upper and lower limits corresponding to the heart rate data, blood oxygen data, body temperature data and real-time comprehensive vital sign characteristic values ​​are input respectively, and the part greater than the preset upper limit and the part lower than the preset lower limit are highlighted.

[0030] The Data Fusion and Health Assessment Module collects data from the Environmental Monitoring Module, the Video Surveillance and Behavior Recognition Module, and the Personal Health Monitoring Module. It integrates all this data to assess the work environment, employee fatigue, and overall work health, matching it to pre-set signals. It sends rest reminders when fatigue is detected and alerts when health risks are identified. It conducts a holistic assessment of the health characteristics of employees within each workspace.

[0031] According to the comprehensive environmental characteristic values, behavioral state characteristic values ​​and comprehensive physical sign characteristic values, a logical judgment based on numerical values ​​is performed to match the preset signal. The specific process is as follows: If it is identified that the real-time comprehensive environmental assessment index of a workplace is greater than a preset threshold value for a continuous preset time period, a first environmental warning signal of the workplace is output; If it is identified that there is a workplace where the basic illumination efficiency characteristic value and the lighting stability characteristic value are both greater than the preset threshold value for a continuous preset time, a second environmental warning signal for the workplace is output; If there is a workplace that outputs the first environmental warning signal and the second environmental warning signal at the same time, the environmental warning signal of the workplace is triggered; If the employee's blood oxygen data is detected to be lower than the preset threshold, a first-level fatigue warning signal will be output; If it is identified that the employee's movement frequency f1 and eye movement frequency f2 are both less than the preset threshold, a secondary fatigue warning signal is output; If the employee's first behavioral characteristic parameter and second behavioral characteristic parameter are both less than the preset threshold, a third-level fatigue warning signal is output; If the employee's first-level fatigue warning signal and second-level fatigue warning signal are recognized at the same time, or if the employee's first-level fatigue warning signal and third-level fatigue warning signal are recognized at the same time, the employee's fatigue warning signal is output.

[0032] If the employee's real-time heart rate data, blood oxygen data, and body temperature data are greater than their own maximum preset thresholds, or lower than their own minimum preset thresholds, a first-level vital sign data warning signal for the employee will be output; As a preferred embodiment of the present invention, the maximum and minimum thresholds of the employee's real-time heart rate data, blood oxygen data, and body temperature data are personalized. The specific process is as follows: For employees undergoing centralized monitoring, the maximum threshold for heart rate data is set at 120 beats / minute, and the minimum threshold is set at 50 beats / minute; the maximum threshold for blood oxygen data is set at 100%, and the minimum threshold is set at 90%; the maximum threshold for body temperature data is set at 38.5°C, and the minimum threshold is set at 35.5°C; For other employees, the maximum threshold for heart rate data is limited to 100 beats / minute, and the minimum threshold is 60 beats / minute; the maximum threshold for blood oxygen data is 100%, and the minimum threshold is 920%; the maximum threshold for body temperature data is 38°C, and the minimum threshold is 36°C.

[0033] If the employee's real-time heart rate data, blood oxygen data and body temperature data are greater than or lower than the preset threshold, the system will output a first-level vital sign data warning signal for the employee; If the employee's real-time comprehensive vital sign characteristic value is greater than the preset threshold, the secondary vital sign warning signal of the employee will be output; If it is identified that an employee outputs a first-level vital sign data warning signal and a second-level vital sign data warning signal at the same time, the employee's vital sign warning signal will be triggered.

[0034] As a preferred method of the present invention, a holistic assessment of the health characteristics of the group of employees in each workspace is performed, with the workspace as the unit. The specific process is as follows: Count the number of times environmental alarm signals are triggered in each workplace in each month; Count the number of times fatigue warning signals and physical sign warning signals are triggered by all employees in each workplace in each month.

[0035] If the number of times a workplace sends out environmental alarm signals is greater than a preset threshold, and the weighted sum of the number of times all employees in the workplace trigger fatigue warning signals and the number of physical sign alarm signals is greater than the preset threshold, then the working environment of the workplace is judged to be poor and has greatly affected the work quality and work experience of employees.

[0036] The early warning and guidance module performs early warning and guidance operations according to the matching signals.

[0037] A reminder message is sent to all personnel in the workplace where the environmental alarm signal is triggered, reminding them to adjust the ventilation, temperature, and lighting environment to improve the working environment. The ventilation fans and air conditioners in the workplace where the environmental alarm signal is triggered are turned on.

[0038] Send reminder messages to employees who have output fatigue warning signals to remind them to take a break in time; Send a reminder message to the employee who triggered the vital sign alarm signal, reminding the employee to pay attention to abnormal vital sign data; For employees who have output fatigue warning signals and triggered physical sign alarm signals, the employee's physical sign data will be sent to the administrator, and the employee's number will be highlighted in the background management table.

[0039] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves accurate health assessment through multi-dimensional data fusion: integrating environmental monitoring, video behavior recognition, and personalized physiological data to build a three-dimensional assessment system covering physical environment, behavioral status, and physiological signs. This breaks through the limitations of single-dimensional monitoring and significantly improves the comprehensiveness and accuracy of health risk prediction. 2. This invention forms a closed-loop management system based on intelligent early warning and proactive intervention mechanisms: multi-level early warnings are triggered based on dynamic thresholds, shifting health management from "post-treatment" to "pre-emptive prevention," effectively reducing the probability of health risks and strengthening the protection of key groups through an administrator linkage mechanism; 3. This invention achieves the coordinated optimization of group health and work environment: it uses the workspace as a unit to count the number of environmental alarms and abnormal employee health data, quantitatively evaluates the impact of the environment on health, and guides targeted transformation and process optimization (such as sedentary reminders), thereby improving the management efficiency of the work environment and employee health. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings: Figure 1 This is a system block diagram of the present invention. DETAILED DESCRIPTION

[0041] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0042] See also Figure 1 As shown, an intelligent monitoring system for the health of living and working personnel includes an environmental monitoring module, a video monitoring and behavior recognition module, a personal health monitoring module, a data fusion and health assessment module, and an early warning and guidance module.

[0043] The environmental monitoring module collects workplace environmental data, including temperature, humidity, air pressure, and light intensity. It comprehensively evaluates these data to analyze the overall comfort level of the work environment. It also conducts a targeted analysis of light intensity, including both natural light and that from lighting fixtures, to analyze the overall lighting conditions of the work environment.

[0044] Obtain the real-time temperature T(t), humidity H(t), air pressure P(t), natural light intensity I1(t) and lighting intensity I2(t) of each workplace.

[0045] For the real-time temperature T(t), humidity H(t), natural light intensity I1(t) and illumination intensity I2(t) of lighting facilities, the preset formula is: Calculate the real-time temperature and humidity index THI (t) and real-time wet bulb globe temperature WBGT (t) of each workplace.

[0046] in is the rate of change of temperature T(t) with time, where It is the preset temperature sensitivity coefficient, and its value range is 0.2-0.5.

[0047] in and They are real-time wet-bulb temperature and real-time black globe temperature respectively.

[0048] It's important to note that the real-time temperature and humidity index is a dynamically calculated indicator of perceived temperature based on current temperature and humidity. It aims to quantify the subjective stress response of the human body to a hot and humid environment through a nonlinear function. The real-time wet-bulb temperature represents the evaporative cooling limit, while the real-time black-globe temperature indicates the attenuation effect of humidity on radiant heat. Together, the real-time wet-bulb and black-globe temperatures reflect the evaporation efficiency of sweat and the overall comfort level with temperature and humidity in the current environment.

[0049] Furthermore, the accelerometer is accessed to obtain the real-time acceleration az(t) generated by the employee in the vertical direction. The real-time electromagnetic communication signal strength S(t) is obtained, and whether the employee is sitting in an elevator is determined based on the real-time acceleration and electromagnetic communication signal strength. The air pressure fluctuation index BPI(t) is numerically corrected to prevent misjudgment of abnormal air pressure fluctuations caused by entering and exiting the elevator.

[0050] Obtain the real-time acceleration az(t) and the rate of change of electromagnetic communication signal strength S(t) △az and △S, If the real-time acceleration change rate △az is greater than the preset threshold, and the electromagnetic communication signal strength change rate △S is greater than the preset threshold, it is determined that the employee is in the elevator environment, and the value of the elevator environment correction factor PAF(t) at time t is set to 1.

[0051] Otherwise, let the value of the elevator environment correction factor PAF(t) at time t be 0.

[0052] For the real-time air pressure P(t), the preset formula is:

[0053] Calculate the air pressure correction factor for each workplace And the air pressure fluctuation index BPI (t). is the preset pressure sensitivity characteristic value, P0 is the local annual average pressure, where is the standard deviation of historical data of air pressure P(t), where is the pressure gradient, and , is the air pressure data collected at the last moment, is the time difference of collecting air pressure data. Where T is the preset time window, sgn is the sign function, For the positive time, The value is 1; When it is negative, The value is -1.

[0054] It should be noted that the air pressure correction factor is an environmental adjustment coefficient constructed based on personnel environmental parameters, real-time air pressure values ​​and their change rates. It aims to quantify the combined effects of static air pressure deviation and dynamic air pressure gradient on human physiology, while shielding the misjudgment of abnormal air pressure fluctuations caused by air pressure fluctuations caused by entering and exiting the elevator; the air pressure fluctuation index is an environmental parameter constructed based on the fluctuation characteristics of real-time air pressure values. It aims to quantify the adverse effects of abnormal air pressure fluctuations on the working environment.

[0055] Furthermore, the real-time temperature and humidity index of each workplace, the elevator environment correction factor, the real-time wet-bulb globe temperature and pressure correction factor, and the pressure fluctuation index are fused and calculated by the preset formula. Calculate the real-time integrated environmental assessment index IEA(t), where α1, α2, and α3 are preset weight factors.

[0056] Furthermore, a comprehensive analysis is conducted on the natural light intensity I1(t) and the lighting intensity I2(t) of the lighting facilities in each workplace, the illumination conditions and lighting stability of the lighting facilities in the working environment are analyzed, and it is evaluated whether the lighting environment provided by the lighting facilities can provide a comfortable working environment in combination with natural light.

[0057] By preset formula

[0058] Calculate the basic illumination efficiency characteristic value A(t) and lighting stability characteristic value of each workplace ω is the natural light weight factor, which represents the proportion of natural light in the total illumination. ;in is the preset minimum illumination of the working environment; is the preset optimal comfortable illumination; It is the average value of the total light intensity [I1(t)+I2(t)] from the current time t to time T.

[0059] It should be noted that the basic illuminance efficiency characteristic value is intended to evaluate whether the combination of natural light and lighting facilities can meet the lighting needs of the working environment; the lighting stability characteristic value is intended to evaluate the fluctuations in the lighting environment, such as the flickering of natural light caused by cloud movement, and to capture characteristic data related to light intensity that causes eye fatigue.

[0060] The calculated real-time comprehensive environmental assessment index IEA(t), basic illumination efficiency characteristic value A(t) and lighting stability characteristic value of each workplace are Sent to the data fusion and health assessment module.

[0061] The video surveillance and behavior recognition module uses video surveillance equipment to identify staff members' mental fatigue, work concentration, and sedentary conditions based on facial recognition and motion analysis, and issues timely warnings.

[0062] The convolutional neural network model is used to identify and locate the staff, distinguish the staff in the video from the background environment, and then number the staff and recognize their faces.

[0063] The limb movement frequency, limb movement amplitude, and movement frequency of the staff identified by the convolutional neural network model are analyzed through a recurrent neural network to extract the employee's movement frequency f1 in the past hour.

[0064] The employee’s facial recognition images are used for expression recognition and eye movement recognition through convolutional neural networks to extract the employee’s eye movement frequency f2 in the past hour.

[0065] Access the employee's computer and obtain behavioral data, including the employee's typing quantity U1 and click quantity U2 per unit time. Record the typing quantity per unit time as the first behavioral characteristic parameter, and record the click quantity per unit time as the second behavioral characteristic parameter.

[0066] The movement frequency f1, eye movement frequency f2, first behavior characteristic parameter and second behavior characteristic parameter of each employee are sent to the data fusion and health assessment module.

[0067] The personal health monitoring module uses customized wearable devices to collect each employee's physiological data, including heart rate, blood oxygen levels, and body temperature. Based on this data, statistical features are extracted from historical data, a timeline is created, and real-time employee physiological data is visualized and anomalies are identified. Health examination information is obtained and, using a natural language processing model, characteristic phrases are captured to identify each employee's health status. Personalized physiological monitoring plans are then developed based on their health status.

[0068] The acquired physiological sign data include heart rate data r1(t), blood oxygen data r2(t) and body temperature data r3(t).

[0069] Furthermore, the statistical characteristics of the heart rate data were analyzed, and the mean and standard deviation of the heart rate data, blood oxygen data and body temperature data on the day were calculated, including the mean value μ(r1) of the heart rate data, the mean value μ(r2) of the blood oxygen data and the mean value μ(r3) of the body temperature data, the standard deviation σ(r1) of the heart rate data, the standard deviation σ(r2) of the blood oxygen data and the standard deviation σ(r3) of the body temperature data.

[0070] By preset formula

[0071] Calculate the abnormal heart rate fluctuation characteristic value E(r1, t), blood oxygen fluctuation characteristic value E(r2, t), and body temperature fluctuation characteristic value E(r3, t).

[0072] Data fusion is performed based on the abnormal heart rate fluctuation characteristic values, blood oxygen fluctuation characteristic values, and body temperature fluctuation characteristic values. The abnormal heart rate fluctuation characteristic values, blood oxygen fluctuation characteristic values, and body temperature fluctuation characteristic values ​​are weighted and summed to obtain the real-time comprehensive vital sign characteristic value E(t).

[0073]

[0074] Among them, r1, r2 and r3 are all preset weight coefficients.

[0075] The real-time heart rate data r1(t), blood oxygen data r2(t), body temperature data r3(t) and real-time comprehensive vital sign characteristic value E(t) are sent to the data fusion and health assessment module.

[0076] Furthermore, we obtain health checkup information, extract characteristic phrases from it based on the natural language processing model, and identify the health status of each employee. We then develop a personalized physiological sign monitoring plan based on their health status. The specific process is as follows: Access the health examination information system, perform word segmentation on the physical examination report, perform natural language processing, entity recognition and classification through the TF-IDF algorithm, and extract specific physical examination data items, including employees' height, weight, blood pressure, blood pressure, over-limit indicators in the physical examination data table, doctor's order notes, past medical history and family genetic disease history data.

[0077] All extracted physical examination data items are structured to generate a structured data table.

[0078] For employees whose keywords "hypertension, hyperglycemia, history of past illness, and family history of genetic diseases" are extracted from the excessive indicators, medical notes, past medical history, and family genetic disease history in the physical examination data sheet, they will be marked as centrally monitored employees, and targeted physiological sign monitoring plans will be formulated for all centrally monitored employees.

[0079] Furthermore, for all centrally monitored employees, a monitoring schedule for physiological sign data is prepared based on the physiological sign data and their abnormal heart rate fluctuation characteristic values, blood oxygen fluctuation characteristic values, and body temperature fluctuation characteristic values. The heart rate data r1(t), blood oxygen data r2(t), body temperature data r3(t), and real-time comprehensive vital sign characteristic value E(t) at each moment are input into a two-dimensional rectangular coordinate system with time as the independent variable, and the preset upper and lower limits corresponding to the heart rate data, blood oxygen data, body temperature data, and real-time comprehensive vital sign characteristic value are input, and the parts greater than the preset upper limit and the parts lower than the preset lower limit are highlighted.

[0080] The Data Fusion and Health Assessment Module collects data from the Environmental Monitoring Module, the Video Surveillance and Behavior Recognition Module, and the Personal Health Monitoring Module. It integrates all this data to assess the work environment, employee fatigue, and overall work health, matching it to pre-set signals. It sends rest reminders when fatigue is detected and alerts when health risks are identified. It conducts a holistic assessment of the health characteristics of employees within each workspace.

[0081] According to the comprehensive environmental characteristic values, behavioral state characteristic values ​​and comprehensive physical sign characteristic values, a logical judgment based on numerical values ​​is performed to match the preset signal. The specific process is as follows: Obtain the real-time comprehensive environmental assessment index IEA(t), basic illumination efficiency characteristic value A(t) and lighting stability characteristic value of each work area sent by the environmental monitoring module ; Obtain the movement frequency f1 and eye movement frequency f2 of each employee sent by the video surveillance and behavior recognition module; Obtain the real-time heart rate data r1(t), blood oxygen data r2(t) and body temperature data r3(t) and real-time comprehensive vital sign characteristic value E(t) sent by the personal health monitoring module.

[0082] If the real-time integrated environmental assessment index IEA(t) of a workplace is identified to be greater than a preset threshold value for a continuous preset time, a first environmental warning signal of the workplace is output; If the basic illuminance efficiency characteristic value A(t) and the light stability characteristic value of the workplace are identified If the value is greater than the preset threshold value within a continuous preset time, a second environmental warning signal for the workplace is output; If there is a workplace that outputs the first environmental warning signal and the second environmental warning signal at the same time, the environmental warning signal of the workplace is triggered; If the employee's blood oxygen data is detected to be lower than the preset threshold, a first-level fatigue warning signal will be output; If it is identified that the employee's movement frequency f1 and eye movement frequency f2 are both less than the preset threshold, a secondary fatigue warning signal is output; If the employee's first behavioral characteristic parameter and second behavioral characteristic parameter are both less than the preset threshold, a third-level fatigue warning signal is output; If the employee's first-level fatigue warning signal and second-level fatigue warning signal are recognized at the same time, or if the employee's first-level fatigue warning signal and third-level fatigue warning signal are recognized at the same time, the employee's fatigue warning signal is output.

[0083] If the employee's real-time heart rate data r1(t), blood oxygen data r2(t) and body temperature data r3(t) are greater than their own maximum preset thresholds, or lower than their own minimum preset thresholds, the first-level vital sign data warning signal of the employee will be output; Furthermore, the maximum and minimum thresholds of employees' real-time heart rate data, blood oxygen data, and body temperature data are customized. The specific process is as follows: For employees undergoing centralized monitoring, the maximum threshold for heart rate data is set at 120 beats / minute, and the minimum threshold is set at 50 beats / minute; the maximum threshold for blood oxygen data is set at 100%, and the minimum threshold is set at 90%; the maximum threshold for body temperature data is set at 38.5°C, and the minimum threshold is set at 35.5°C; For other employees, the maximum threshold for heart rate data is limited to 100 beats / minute, and the minimum threshold is 60 beats / minute; the maximum threshold for blood oxygen data is 100%, and the minimum threshold is 920%; the maximum threshold for body temperature data is 38°C, and the minimum threshold is 36°C.

[0084] If the employee's real-time comprehensive vital sign characteristic value E(t) is greater than the preset threshold, the secondary vital sign warning signal of the employee will be output; If it is identified that an employee outputs a first-level vital sign data warning signal and a second-level vital sign data warning signal at the same time, the employee's vital sign warning signal will be triggered.

[0085] Furthermore, taking the workspace as a unit, a holistic assessment of the group health characteristics of employees in each space is conducted. The specific process is as follows: Count the number of times N1 environmental alarm signals are triggered in each workplace in each month; The number of times N2 fatigue warning signals are triggered and the number of times N3 physical sign warning signals are triggered by all employees in each workplace in each month are counted.

[0086] If the number N1 of workplace environmental alarm signals is greater than the preset threshold N1max, and the weighted sum of the number N2 of fatigue warning signals and the number N3 of physical sign alarm signals triggered by all employees in the workplace is greater than the preset threshold, then it is determined that the working environment of the workplace is poor and has greatly affected the work quality and work experience of employees.

[0087] The early warning and guidance module performs early warning and guidance operations according to the matching signals.

[0088] A reminder message is sent to all personnel in the workplace where the environmental alarm signal is triggered, reminding them to adjust the ventilation, temperature, and lighting environment to improve the working environment. The ventilation fans and air conditioners in the workplace where the environmental alarm signal is triggered are turned on.

[0089] Send reminder messages to employees who have output fatigue warning signals to remind them to take a break in time; Send a reminder message to the employee who triggered the vital sign alarm signal, reminding the employee to pay attention to abnormal vital sign data; For employees who have output fatigue warning signals and triggered physical sign alarm signals, the employee's physical sign data will be sent to the administrator, and the employee's number will be highlighted in the background management table.

[0090] It should be understood that the terms “include” and “comprising” used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0091] It should also be understood that the terms used in this disclosure are for the purpose of describing particular embodiments only and are not intended to limit the disclosure. As used in this disclosure and the claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and the claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations; The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An intelligent monitoring system for the health of people in their daily lives and workplaces, comprising an environmental monitoring module, a video surveillance and behavior recognition module, and a personal health monitoring module, characterized by: The environmental monitoring module collects environmental data of the workplace and analyzes the overall comfort of the working environment; Conduct targeted analysis of light intensity, including natural light intensity and light intensity from lighting facilities, to analyze the overall lighting conditions of the work environment; The video surveillance and behavior recognition module uses video surveillance equipment to identify workers' mental fatigue, work concentration, and sedentary behavior based on facial recognition and motion analysis; The personal health monitoring module collects each staff member's physiological sign data through a customized wearable device; Based on physiological data, we can extract statistical features from historical data of physiological signs data, create a timetable of physiological signs data, and visualize and identify abnormalities of employees' real-time physiological signs data; obtain health examination information, capture feature phrases based on natural language processing models, and identify the health status of each employee; and formulate personalized physiological signs monitoring plans based on their health status.

2. The intelligent monitoring system for the health of living and working personnel according to claim 1 is characterized in that: It also includes data fusion and health assessment modules and early warning and guidance modules: The data fusion and health assessment module obtains data from the environmental monitoring module, video surveillance and behavior recognition module, and personal health monitoring module, integrates all data to assess the work environment, employee fatigue, and overall health status at work, and matches it to the preset signal; Send a rest reminder when identifying an employee's fatigue, and send a reminder signal when identifying an employee's health risk; Taking the workspace as a unit, conduct a holistic assessment of the group health characteristics of employees in each space; The early warning and guidance module performs early warning and guidance operations according to the matching signals.

3. The intelligent monitoring system for the health of living and working personnel according to claim 1 is characterized in that: The specific process of overall evaluation of temperature, humidity and air pressure data is as follows: Obtain real-time temperature, humidity, air pressure, natural light intensity, and lighting intensity of each workplace; Feature extraction is performed on real-time temperature, humidity, and natural light intensity. The real-time wet-bulb temperature and real-time black-globe temperature of each workplace are calculated using a preset formula. The real-time wet-bulb temperature and real-time black-globe temperature are calculated to obtain the real-time wet-bulb black-globe temperature of each workplace; the temperature and humidity index of each workplace is calculated by calculating the real-time temperature and humidity. Access the accelerometer to obtain the employee's real-time vertical acceleration and electromagnetic communication signal strength. Based on the real-time acceleration and electromagnetic communication signal strength, determine whether the employee is in an elevator. Perform numerical correction on the air pressure fluctuation index to prevent misjudgments of abnormal air pressure fluctuations caused by entering and exiting the elevator. Obtain the real-time acceleration and the rate of change of the electromagnetic communication signal strength. If the real-time acceleration rate of change is greater than a preset threshold, and the electromagnetic communication signal strength rate of change is greater than a preset threshold, it is determined that the employee is in an elevator environment, and the current elevator environment correction factor is set to 1. Otherwise, let the value of the elevator environment correction factor at time t be 0; For real-time air pressure, obtain the local annual average air pressure and calculate the standard deviation of historical data; Calculate the time gradient of air pressure and calculate the air pressure correction factor and air pressure fluctuation index of each workplace through preset formulas; The real-time temperature and humidity index, real-time wet-bulb globe temperature and pressure correction factor, and pressure fluctuation index of each workplace are integrated to calculate the real-time comprehensive environmental assessment index using a preset formula. Conduct a comprehensive analysis of the natural light intensity and lighting intensity of each workplace, analyze the illumination conditions and lighting stability of the lighting facilities in the work environment, and evaluate whether the lighting environment provided by the lighting facilities can provide a comfortable working environment in conjunction with natural light; Calculate the proportion of natural light in the total illuminance, the preset minimum illuminance of the working environment, and the preset optimal comfortable illuminance, and use them as calculation parameters to calculate the basic illuminance efficiency characteristic value of each workplace using the preset formula; Calculate the average value of the total light intensity from the current moment to the preset moment, and use it as a calculation parameter. Enter it into the preset formula together with the preset time length, natural light intensity and light intensity of lighting facilities to calculate the light stability characteristic value. The calculated real-time comprehensive environmental assessment index, elevator environment correction factor, basic illumination efficiency characteristic value and lighting stability characteristic value of each workplace are sent to the data fusion and health assessment module.

4. The intelligent monitoring system for the health of living and working personnel according to claim 1 is characterized in that: The specific process of identifying the mental fatigue status of staff based on face recognition and motion analysis is as follows: The convolutional neural network model is used to identify and locate workers, distinguishing them from the background environment in the video, and then numbering and recognizing their faces. The recurrent neural network analyzes the body movement frequency, body movement amplitude, and movement frequency of the staff identified by the convolutional neural network model, and extracts the employee's movement frequency in the past hour; Use convolutional neural networks to perform facial recognition and eye movement recognition on employee face recognition images and extract the employee's eye movement frequency in the past hour; Access the employee's computer to obtain behavioral data, including the employee's typing and click count per unit time, record the typing count per unit time as the first behavioral characteristic parameter, and record the click count per unit time as the second behavioral characteristic parameter; The movement frequency, eye movement frequency, first behavior characteristic parameter and second behavior characteristic parameter of each employee are sent to the data fusion and health assessment module.

5. The intelligent monitoring system for the health of living and working personnel according to claim 1 is characterized in that: The specific process of extracting statistical features from historical data of physiological signs data is as follows: Performing statistical feature analysis on the heart rate data, calculating the average and standard deviation of the heart rate data, blood oxygen data, and body temperature data for the day, and inputting the average and standard deviation into a preset formula to calculate the absolute value of a preset multiple of the difference between the heart rate data, blood oxygen data, and body temperature data and their own average values, obtaining a first result of the heart rate data, blood oxygen data, and body temperature data, and dividing the first result by the standard deviation of the heart rate data, blood oxygen data, and body temperature data to obtain a heart rate abnormal fluctuation characteristic value, a blood oxygen fluctuation characteristic value, and a body temperature fluctuation characteristic value; Based on the abnormal heart rate fluctuation characteristic value, blood oxygen fluctuation characteristic value and body temperature fluctuation characteristic value, data fusion is performed, and the abnormal heart rate fluctuation characteristic value, blood oxygen fluctuation characteristic value and body temperature fluctuation characteristic value are weighted and summed to obtain the real-time comprehensive vital sign characteristic value; The real-time heart rate data, blood oxygen data, body temperature data and real-time comprehensive vital sign feature values ​​are sent to the data fusion and health assessment module.

6. The intelligent monitoring system for the health of living and working personnel according to claim 1 is characterized in that: The specific process of obtaining health examination information is as follows: Access the health examination information system, perform word segmentation on the physical examination report, perform natural language processing, entity recognition and classification using the TF-IDF algorithm, and extract specific physical examination data items, including employee height, weight, blood pressure, blood pressure, over-limit indicators in the physical examination data table, doctor's order notes, past medical history and family genetic disease history data; All extracted physical examination data items are structured to generate a structured data table.

7. The intelligent monitoring system for the health of living and working personnel according to claim 1 is characterized in that: The specific process of developing a personalized physiological sign monitoring plan based on health conditions is as follows: Employees whose physical examination data contains keywords such as "hypertension, hyperglycemia, pre-existing medical history, or family genetic disease history" in their excessive indicators, doctor's order notes, and past medical history and family genetic disease history will be marked as centralized monitoring employees. Targeted physiological sign monitoring plans will be developed for all centralized monitoring employees. For all employees under central monitoring, a monitoring schedule for physiological signs data is prepared based on the physiological signs data and their abnormal heart rate fluctuation characteristic values, blood oxygen fluctuation characteristic values ​​and body temperature fluctuation characteristic values. The heart rate data, blood oxygen data, body temperature data and real-time comprehensive vital signs characteristic values ​​at each moment are input into a two-dimensional rectangular coordinate system with time as the independent variable, and the preset upper and lower limits corresponding to the heart rate data, blood oxygen data, body temperature data and real-time comprehensive vital signs characteristic values ​​are input respectively. The parts that are greater than their own preset upper limits and parts that are lower than their own preset lower limits are highlighted.

8. The intelligent monitoring system for the health of living and working personnel according to claim 2 is characterized in that: The specific process of integrating all data to evaluate the work environment, employee fatigue and overall work health is as follows: According to the comprehensive environmental characteristic values, behavioral state characteristic values ​​and comprehensive physical sign characteristic values, a logical judgment based on numerical values ​​is performed to match the preset signal. The specific process is as follows: If it is identified that the real-time comprehensive environmental assessment index of a workplace is greater than a preset threshold value for a continuous preset time period, a first environmental warning signal of the workplace is output; If it is identified that there is a workplace where the basic illumination efficiency characteristic value and the lighting stability characteristic value are both greater than the preset threshold value for a continuous preset time, a second environmental warning signal for the workplace is output; If there is a workplace that outputs the first environmental warning signal and the second environmental warning signal at the same time, the environmental warning signal of the workplace is triggered; If the employee's blood oxygen data is detected to be lower than the preset threshold, a first-level fatigue warning signal will be output; If it is identified that the employee's movement frequency f1 and eye movement frequency f2 are both less than the preset threshold, a secondary fatigue warning signal is output; If the employee's first behavioral characteristic parameter and second behavioral characteristic parameter are both less than the preset threshold, a third-level fatigue warning signal is output; If both the first-level and second-level fatigue warning signals of an employee are identified, or both the first-level and third-level fatigue warning signals of an employee are identified, then the employee's fatigue warning signal is output; If the employee's real-time heart rate data, blood oxygen data, and body temperature data are greater than their own maximum preset thresholds, or lower than their own minimum preset thresholds, a first-level vital sign data warning signal for the employee will be output; Furthermore, the maximum and minimum thresholds of employees' real-time heart rate data, blood oxygen data, and body temperature data are customized. The specific process is as follows: For employees undergoing centralized monitoring, the maximum threshold for heart rate data is set at 120 beats / minute, and the minimum threshold is set at 50 beats / minute; the maximum threshold for blood oxygen data is set at 100%, and the minimum threshold is set at 90%; the maximum threshold for body temperature data is set at 38.5°C, and the minimum threshold is set at 35.5°C; For other employees, the maximum threshold for heart rate data is set at 100 beats / minute, and the minimum threshold is set at 60 beats / minute; the maximum threshold for blood oxygen data is set at 100%, and the minimum threshold is set at 920%; the maximum threshold for body temperature data is set at 38°C, and the minimum threshold is set at 36°C; If the employee's real-time comprehensive vital sign characteristic value is greater than the preset threshold, the secondary vital sign warning signal of the employee will be output; If it is identified that an employee outputs a first-level vital sign data warning signal and a second-level vital sign data warning signal at the same time, the employee's vital sign warning signal will be triggered.

9. The intelligent monitoring system for the health of living and working personnel according to claim 2 is characterized in that: The specific process of conducting a holistic assessment of the group health characteristics of employees in each space is as follows: Count the number of times environmental alarm signals are triggered in each workplace in each month; Count the number of fatigue warning signals and physical sign warning signals triggered by all employees in each workplace in each month; If the number of times a workplace sends out environmental alarm signals is greater than a preset threshold, and the weighted sum of the number of times all employees in the workplace trigger fatigue warning signals and the number of physical sign alarm signals is greater than the preset threshold, then the working environment of the workplace is judged to be poor and has greatly affected the work quality and work experience of employees.

10. The intelligent monitoring system for the health of living and working personnel according to claim 2 is characterized in that: The specific process of executing early warning and guidance operations is as follows: Send reminder messages to all personnel in the workplace where the environmental alarm signal is triggered, reminding them to adjust the ventilation, temperature and lighting environment to improve the working environment; Turn on ventilation fans and air conditioners in the workplace that triggers the environmental alarm signal; Send reminder messages to employees who have output fatigue warning signals to remind them to take a break in time; Send a reminder message to the employee who triggered the vital sign alarm signal, reminding the employee to pay attention to abnormal vital sign data; For employees who have output fatigue warning signals and triggered physical sign warning signals; Send the employee's vital sign data to the administrator and highlight the employee's number in the background management form.

Images