Method for evaluating health condition of worker based on AI (Artificial Intelligence) identification

Abstract

The invention relates to the technical field of engineering construction, and provides an AI identification-based worker health condition assessment method, which comprises the steps of obtaining worker health data in engineering construction, and establishing a health file; importing the health archive into a pre-configured exposure factor identification network, identifying exposure factors in engineering construction, and establishing attribution risk degree functions of different exposure factors; according to the attribution risk degree function, attribution risk degree values of different exposure factors relative to the workers are determined, and a worker health condition assessment report is generated. The method not only provides powerful support for engineering safety management, but also shows a remarkable effect in solving practical problems. According to the method, the source, the influence degree and the occurrence probability of various potential risk factors in the engineering construction process are deeply analyzed, so that the risk degree is accurately and quantitatively evaluated. And corresponding risk prevention and control measures are formulated according to the evaluation result, so that the possibility of accidents is effectively reduced, and safe and stable engineering construction is ensured.

Description

Technical Field

[0001] The present invention relates to the field of risk calculation and control of engineering construction, and in particular to a method for assessing the health status of workers based on AI recognition. Background Art

[0002] With the development of science and technology and people's increasing concern about health, the health of construction personnel in the process of engineering construction has become a key issue for the country. The attributable risk calculation method, as a core area, has brought great changes and conveniences to people's lives.

[0003] Because construction workers work on construction sites, they are often far away from home and live in simple accommodation. The working hours are long, usually more than 12 hours, from 6 am to 6 pm or 8 pm, and illegal long working hours have become the norm. Because of long hours of labor, their bodies are easily affected. Because of long-term high-load work, many workers suffer from various occupational diseases. For example, asthma, bronchitis, etc. Some workers even become physically weak due to long hours of work, or even in life-threatening situations, and can only be treated with some conventional means:

[0004] For example: 1. Physical examination, 2. Occupational disease detection, 3. Common disease screening, 4. Mental health assessment, 5. Nutritional examination, 6. Personal protective equipment inspection, etc.

[0005] However, conventional methods may lead to missing the best treatment time due to untimely examinations and irregular testing, and may also cause sudden death due to the underlying diseases of the construction personnel themselves.

[0006] Therefore, the existing detection methods and early warning measures cannot comprehensively analyze the health status of construction personnel. Summary of the invention

[0007] The present invention proposes a worker health status assessment method based on AI recognition, which is used to solve the problem that existing detection methods and early warning measures cannot comprehensively analyze the health status of construction personnel well.

[0008] The present invention proposes a worker health status assessment method based on AI recognition, comprising:

[0009] Obtain health data of workers in construction projects and build health records;

[0010] Import health records into a pre-configured exposure factor identification network to identify exposure factors in engineering construction and build attributable risk functions for different exposure factors;

[0011] Based on the attributable risk function, the attributable risk values ​​of different exposure factors relative to the workers are determined, and a worker health status assessment report is generated.

[0012] Furthermore, the health records include personal records and engineering records, where:

[0013] Personal files are used to obtain workers’ physiological and biochemical data, lifestyle data, personal health history data, and family health history data;

[0014] Engineering archives are used to obtain exposure environment data during engineering construction. The exposure environment data include: noise data, air pollution data, building safety data, equipment safety risk data, water resource pollution data and temperature status data.

[0015] Furthermore, the exposure environment data acquisition process also includes:

[0016] Obtain exposure environmental data from historical engineering construction and determine the characteristics of environmental ecological factors;

[0017] According to the characteristics of environmental ecological factors, determine the time point of occurrence of environmental ecological factors in different environmental ecology;

[0018] According to the time point of occurrence of environmental and ecological factors, a statistical matrix of environmental ecology is constructed;

[0019] According to the statistical matrix of environmental ecology, the attribute intervals of different environmental ecological factors are determined; wherein the attribute intervals include time intervals and intensity intervals;

[0020] According to the attribute range, the distribution patterns of different environmental ecological factors are determined, and distribution labels of different environmental ecological factors are set; among them, the distribution patterns include regional distribution, intensity distribution and time distribution.

[0021] Furthermore, the health record is also configured with a classification filling mechanism, wherein:

[0022] Configure the data attributes of various health data in similar configuration modes, including:

[0023] In the similar configuration mode, there are multiple groups of filling positions in the health record, and the data attributes of each group of filling positions in the multiple groups of filling positions are consistent with the attributes of the target phrase; each of the multiple groups of filling positions is configured with a displacement filling rule;

[0024] When the worker health data responds to any target phrase attribute, the worker health data and the filling position of the corresponding phrase are classified and traversed, wherein,

[0025] When the filling rules of any filling position are met after classification traversal, healthy data filling is performed.

[0026] Furthermore, the step of importing the health records into a pre-configured exposure factor identification network to identify exposure factors in engineering construction includes:

[0027] According to the exposure factor identification network, the environmental ecological data in the health records are identified one by one to determine the environmental ecological characteristics;

[0028] The strong correlation between different environmental ecological characteristics and workers' health status was determined, and a strong correlation network was constructed, in which:

[0029] Different environmental ecological characteristics and workers' health status in the strong correlation network have unique connected components.

[0030] Based on the strong association network, the positive and negative effects of different environmental ecological characteristics on the workers' health status were determined;

[0031] Based on the negative impacts, the corresponding target environmental ecological characteristics are determined and marked as exposure factors.

[0032] Furthermore, the construction of the attributable risk function of different exposure factors includes:

[0033] Based on the exposure factors, the risk levels of different exposure factors are predefined, and the exposure behaviors of the corresponding exposure factors are collected. Based on the exposure behaviors, the statistical risk rates of different exposure behaviors are determined, where: the statistical risk rate is the incidence rate of different exposure behaviors among workers;

[0034] Based on the statistical risk rate, the population attributable risk percentage was calculated;

[0035] According to the population attributable percentage, the relative risk of different exposure behaviors was calculated;

[0036] Based on the relative risk, construct the attributable risk function of different exposure factors.

[0037] Furthermore, the specific formula of the attributable risk function is:

[0038] CRF=C*RT*SI*IP+C*MD*P(CI0-CI1)

[0039] in:

[0040] C = risk factor, value range [0,6]

[0041] RT = risk transfer rate, range [0,1] dimensionless

[0042] MD = rate of change, value range [-1,1] dimensionless

[0043] P = the probability of potential risk, with a value range of [0,1] dimensionless

[0044] CI0 = safety index before the change, value range [0,1] dimensionless

[0045] CI1 = safety index after change, value range [0,1] dimensionless

[0046] SI = safety index, value range [0,1] dimensionless

[0047] IP = implementation plan, value range [0,1] dimensionless

[0048] Where C=Cy+Ck+Cj+Ct+Cs+Cp

[0049] Cy is the noise pollution hazard coefficient; Ck is the air pollution hazard coefficient; Cj is the building hazard coefficient; Ct is the equipment hazard coefficient; Cs is the water resource hazard coefficient; Cp is the water resource hazard coefficient.

[0050] Furthermore, the determination of the attributable risk values ​​of different exposure factors relative to the worker according to the attributable risk function includes:

[0051] According to the attributable risk function, the first simulation evolution is performed on different exposure factors in turn; wherein the simulation calculation includes the initial exposure time and the end exposure time of the different exposure factors;

[0052] According to the simulation calculation, the original calculation time of multiple different exposure factors is calculated in turn, and the adjustment time is determined, wherein: there is a proportional coefficient between adjacent adjustment times;

[0053] According to the adjustment time, a second simulation evolution is performed for different exposure factors, and the first simulation evolution result is compared with the second simulation evolution result;

[0054] When the second simulation evolution result is greater than the first simulation evolution result, the target and attributed risk value of the second simulation evolution result are changed;

[0055] When the second simulation evolution result is greater than the first simulation evolution result, the first simulation evolution result or the first simulation evolution result is used as the target attribution risk value;

[0056] When the second simulation evolution result is less than the first simulation evolution result, the first simulation evolution result is used as the target attribution risk value;

[0057] Furthermore, the generating of the worker health status assessment report includes:

[0058] Configure assessment maps for different exposure factors;

[0059] Based on the evaluation map, the workers' health status is divided into multiple evaluation display pages according to different exposure factors, and each evaluation display page corresponds to the display page of different exposure factors;

[0060] Based on the display page, determine at least one health risk type corresponding to different exposure factors, use the health risk type to connect the different assessment contents of the display page, and generate a visual assessment report.

[0061] Furthermore, the generating of the worker health status assessment report also includes:

[0062] Displaying at least one visualization area in the visualization evaluation report and setting a trigger instruction; wherein the trigger instruction is used to trigger a corresponding evaluation process;

[0063] When the assessment process is triggered, the corresponding attribution mapping graph is configured according to the worker’s physical risk data;

[0064] According to the attribution mapping map, the exposure distribution status of users in engineering construction is determined, and the exposure factor arrangement table of engineering construction is generated based on the exposure distribution status.

[0065] The beneficial effects of the present invention are:

[0066] The present invention uses AI technology to conduct in-depth analysis of workers' health data, which can more accurately identify individual health conditions, thereby providing workers with more personalized health management recommendations. Through continuous monitoring and analysis of health records, early signs of workers' health problems can be discovered in a timely manner, so that preventive measures can be taken in a timely manner to avoid the deterioration of health problems. By building an attributable risk function, the impact of different exposure factors on workers' health can be quantified. By identifying the attributable risk of different exposure factors, safety protection resources and health intervention measures can be allocated in a targeted manner. Automated data processing and evaluation report generation can greatly improve the efficiency of health management and reduce the workload of managers. Through systematic health assessments, it is helpful to improve the working environment, reduce the incidence of occupational diseases, and protect the physical health and life safety of workers.

[0067] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0068] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0070] Figure 1This is a method flow chart of a worker health status assessment method based on AI recognition in an embodiment of the present invention;

[0071] Figure 2 A flow chart of obtaining distribution identification of environmental ecological factors in an embodiment of the present invention;

[0072] Figure 3 This is the process of obtaining the exposure factor arrangement table in the embodiment of the present invention. DETAILED DESCRIPTION

[0073] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0074] The present invention relates to a construction project, in which a noise pollution hazard coefficient, an air pollution hazard system, a building safety risk hazard coefficient, an equipment safety risk hazard coefficient, a water resource pollution hazard coefficient, a body temperature hazard coefficient, and the like are fully considered through multiple parameters involved; based on the noise pollution hazard coefficient, the air pollution hazard system, the building safety risk hazard coefficient, the equipment safety risk hazard coefficient, and the water resource pollution hazard coefficient, a calculation formula for attributing the hazard of the construction project based on artificial intelligence identification is established.

[0075] The attributable risk calculation method has many advantages over traditional health monitoring methods. First, the attributable risk calculation method is highly real-time and can monitor and analyze individual physiological indicators anytime and anywhere, detect problems in a timely manner and take measures. Second, the attributable risk calculation method is individualized and can provide corresponding health advice based on the needs and characteristics of users, and accurately manage individual health. In addition, the attributable risk calculation method combines technologies such as cloud computing and big data analysis.

[0076] like Figure 1 As shown, this embodiment provides a method for assessing worker health status based on AI recognition, including:

[0077] Obtain health data of workers in construction projects and build health records;

[0078] Import health records into a pre-configured exposure factor identification network to identify exposure factors in engineering construction and build attributable risk functions for different exposure factors;

[0079] Based on the attributable risk function, the attributable risk values ​​of different exposure factors relative to the workers are determined, and a worker health status assessment report is generated.

[0080] The principle of the above technical solution is:

[0081] In actual implementation, the health data of workers in engineering construction is obtained and health files are established. The health data of workers is collected in various ways, such as through employee health management systems, physical examination records or medical records. After the health data of workers is collected, it is imported into a preset worker health database, and combined with the corresponding software port of the present invention, the data is stored and managed.

[0082] During the establishment of health records, disease screening and prevention will be carried out to detect many diseases early, such as hypertension, diabetes and other chronic diseases. Through health records, personal physiological and biochemical data, lifestyle, personal or family health history and other information are collected. This information is used to assess the individual's attributable risk. Attributable risk (AR) refers to the difference between the incidence rate of the exposed group and the incidence rate of the non-exposed group, reflecting the degree to which the disease is attributed to the exposure factor.

[0083] Import health records into the pre-configured exposure factor identification network. The exposure factor identification network identifies factors that may affect workers' health according to different environments and conditions, and classifies them into different exposure factors. These exposure factors include but are not limited to: noise, vibration, radiation, high / low temperature, chemical pollution, etc. In the network, each exposure factor has a corresponding weight and parameter to characterize the potential impact of the corresponding factor on workers' health.

[0084] Based on the output of the exposure factor identification network, the attributable risk values ​​of different exposure factors relative to the workers are determined. The attributable risk value reflects the impact of the corresponding factor on the health of the workers and assesses the health risks of the workers in a specific environment. The calculation process of the attributable risk value involves a comprehensive evaluation of multiple factors and a weighted average calculation.

[0085] The attributable risk function includes:

[0086] The attributable risk percentage (AR%) refers to the percentage of all diseases caused by exposure factors in the exposed population. Personal health information recorded in health records can be used to calculate the individual's AR%, thereby assessing the impact of specific health behaviors or conditions on individual health.

[0087] The population attributable risk percentage (PAR%) refers to the proportion of disease caused by exposure in the entire population. By analyzing the health records of a large number of people, the distribution of specific exposure factors in the population can be estimated and PAR% can be calculated, which helps public health decision makers understand the health benefits that may be brought about by controlling specific exposure factors.

[0088] Provide relevant health advice and intervention measures based on the generated worker health status assessment report. The worker health status assessment report should include a description of the exposure factors, the worker's health status assessment results, and the corresponding health advice and intervention measures. Generate corresponding improvement strategies or measures based on the advice obtained, including improving the working environment, strengthening personal protective equipment, increasing rest time, etc. Advice and measures can help workers reduce health risks and prevent the occurrence and deterioration of diseases.

[0089] In actual implementation, by establishing health records, popularizing health knowledge, and changing bad behaviors and living habits, the attributable risk can be reduced. Health records help users make comprehensive self-health assessments based on records, effectively control health status, and prevent the occurrence and development of diseases.

[0090] Health records can quickly understand the changing trends of the patient's health indicators and possible risk factors, so as to make a comprehensive evaluation of the condition and take accurate and effective treatment measures, which is closely related to the assessment and management of attributable risk.

[0091] The technical effect of the above scheme is that through in-depth analysis of workers' health data through AI technology, individual health conditions can be more accurately identified, thereby providing workers with more personalized health management suggestions. Through continuous monitoring and analysis of health records, early signs of workers' health problems can be discovered in a timely manner, so that preventive measures can be taken in time to avoid the deterioration of health problems. By building an attributable risk function, the impact of different exposure factors on workers' health can be quantified. By identifying the attributable risk of different exposure factors, safety protection resources and health intervention measures can be allocated in a targeted manner. Automated data processing and evaluation report generation can greatly improve the efficiency of health management and reduce the workload of managers. Through systematic health assessment, it is helpful to improve the working environment, reduce the incidence of occupational diseases, and protect the physical health and life safety of workers.

[0092] As an embodiment of the present invention: the health record includes a personal record and an engineering record, wherein:

[0093] Personal files are used to obtain workers’ physiological and biochemical data, lifestyle data, personal health history data, and family health history data;

[0094] Engineering archives are used to obtain exposure environment data during engineering construction. The exposure environment data include: noise data, air pollution data, building safety data, equipment safety risk data, water resource pollution data and temperature status data.

[0095] The principle of the above technical solution is:

[0096] In actual implementation, workers' health-related information is collected by establishing personal files and project files. Personal files include physiological and biochemical data (such as blood pressure, blood sugar, blood lipids, etc.), lifestyle data (such as smoking, drinking, exercise habits, etc.), personal health history data and family health history data. Project files focus on exposure environment data in engineering construction, such as noise, air pollution, building safety, equipment safety risks, water resource pollution and body temperature status.

[0097] This data is processed and analyzed using artificial intelligence techniques, such as machine learning algorithms. AI can identify patterns and associations in the data to assess workers’ health status and potential health risks.

[0098] Based on the analysis results of AI, the health status of workers is assessed, including risk prediction for specific diseases, assessment of the attributable hazard of exposure to the work environment, etc.

[0099] As an embodiment of the present invention: the exposure environment data acquisition process further includes:

[0100] Obtain exposure environmental data from historical engineering construction and determine the characteristics of environmental ecological factors;

[0101] According to the characteristics of environmental ecological factors, determine the time point of occurrence of environmental ecological factors in different environmental ecology;

[0102] According to the time point of occurrence of environmental and ecological factors, a statistical matrix of environmental ecology is constructed;

[0103] According to the statistical matrix of environmental ecology, the attribute intervals of different environmental ecological factors are determined; wherein the attribute intervals include time intervals and intensity intervals;

[0104] According to the attribute range, the distribution patterns of different environmental ecological factors are determined, and distribution labels of different environmental ecological factors are set; among them, the distribution patterns include regional distribution, intensity distribution and time distribution.

[0105] The principle of the above technical solution is:

[0106] like Figure 2 As shown, in actual implementation, the historical engineering construction data is first analyzed to determine the historical data that may pose a threat to employee health. While obtaining the exposure environment data in the current engineering construction, the exposure environment data in the historical engineering construction is also collected to determine the changing trends and characteristics of environmental ecological factors over time.

[0107] The determination of environmental ecological factors is to determine the characteristics of environmental ecological factors that affect workers' health by analyzing historical and current data, such as the nature and characteristics of noise, air pollution, etc. The time point determination is to identify the time point of the generation of different environmental ecological factors, that is, when these factors begin to affect workers' health.

[0108] Then, based on the time points of the generation of different environmental and ecological factors, a statistical matrix is ​​constructed to record and analyze the data of these factors at different time points. Finally, according to the statistical results of the statistical matrix, the attribute parameters of different environmental and ecological factors after assignment are divided, and then the distribution law and distribution identification of specific environmental and ecological factors are determined, and different environmental and ecological factors are quantified.

[0109] As an embodiment of the present invention: the health record is also configured with a classification filling mechanism, wherein:

[0110] Configure the data attributes of various health data in similar configuration modes, including:

[0111] In the similar configuration mode, there are multiple groups of filling positions in the health record, and the data attributes of each group of filling positions in the multiple groups of filling positions are consistent with the attributes of the target phrase; each of the multiple groups of filling positions is configured with a displacement filling rule;

[0112] When the worker health data responds to any target phrase attribute, the worker health data and the filling position of the corresponding phrase are classified and traversed, wherein,

[0113] When the filling rules of any filling position are met after classification traversal, healthy data filling is performed.

[0114] The principle of the above technical solution is:

[0115] In actual implementation, the classification filling mechanism of the present invention is to configure a plurality of data attribute similarity configuration modes of health data so that the health records can be classified and filled according to different data attributes. In actual operation, the data attribute similarity configuration mode fills multiple groups of filling positions set in the health records with specific data attributes, and the data attribute filling matches the attributes of the target phrase (i.e., the keyword or phrase related to the health data).

[0116] Then, through the displacement filling rule, each filling position can be configured with a specific displacement filling rule, and the displacement filling rule defines how the data is filled into the corresponding position.

[0117] Then, by setting the corresponding target phrase attribute, the worker's health data is matched with a target phrase attribute, and the worker's health data and the filling position of the target phrase are classified and traversed to check whether the data meets the filling rules of any filling position. If it meets the requirements, the health data filling will be performed, that is, the data will be placed in the corresponding position.

[0118] As an embodiment of the present invention, the step of importing the health records into a pre-configured exposure factor identification network to identify exposure factors in engineering construction includes:

[0119] According to the exposure factor identification network, the environmental ecological data in the health records are identified one by one to determine the environmental ecological characteristics;

[0120] The strong correlation between different environmental ecological characteristics and workers' health status was determined, and a strong correlation network was constructed, in which:

[0121] Different environmental ecological characteristics and worker resistance status in the strongly connected network have a unique connected component.

[0122] Based on the strong association network, the positive and negative effects of different environmental ecological characteristics on the workers' health status were determined;

[0123] Based on the negative impacts, the corresponding target environmental ecological characteristics are determined and marked as exposure factors.

[0124] The principle of the above technical solution is:

[0125] In actual implementation, the exposure factor identification network is a pre-configured AI network of the present invention, which is specifically used to identify and evaluate environmental ecological data in engineering construction and determine exposure factors that may affect the health of workers.

[0126] Then, based on the identification of environmental ecological data, the environmental ecological data in the health records are input into the exposure factor identification network. The network identifies the data one by one and determines the environmental ecological characteristics. Through the strong correlation between different environmental ecological characteristics and the health status of workers, a strong correlation network is constructed. In this process, the nature of the impact is determined by the unique connected component between each environmental ecological characteristic and the health status of the workers, that is, the relationship between them is clear and direct. Then, the strong correlation network is used to evaluate the positive and negative effects of different environmental ecological characteristics on the health status of workers.

[0127] Then, by labeling exposure factors, that is, according to negative impacts, the environmental ecological characteristics that have a negative impact on workers' health are identified and labeled as exposure factors.

[0128] As an embodiment of the present invention: the construction of the attributable risk function of different exposure factors includes:

[0129] Based on the exposure factors, the risk levels of different exposure factors are predefined, and the exposure behaviors of the corresponding exposure factors are collected.

[0130] According to the exposure behaviors, the statistical risk rates of different exposure behaviors are determined, where: the statistical risk rate is the incidence rate of different exposure behaviors among workers;

[0131] Based on the statistical risk rate, the population attributable risk percentage was calculated;

[0132] According to the population attributable percentage, the relative risk of different exposure behaviors was calculated;

[0133] Based on the relative risk, construct the attributable risk function of different exposure factors.

[0134] The principle of the above technical solution is:

[0135] In the actual implementation of the present invention, the risk levels of different exposure factors are predefined. According to different exposure factors, their respective risk levels are predefined. The risk levels are determined based on existing epidemiological studies and data.

[0136] Then, exposure behaviors are collected, followed by data on workers' exposure behaviors related to these exposure factors, including workers' activities in a specific environment and exposure time.

[0137] The statistical risk rate is then determined by analyzing the exposure behavior data to determine the statistical risk rate of different exposure behaviors, that is, the incidence of these behaviors among workers. The population attributable risk percentage is calculated using the statistical risk rate to calculate the population attributable risk percentage (Population Attributable Fraction, PAF), which represents the expected reduction in the incidence of a disease in a specific population if a certain exposure factor is eliminated. The relative risk is calculated based on the population attributable risk percentage to calculate the relative risk (Relative Risk, RR) of different exposure behaviors, which is the ratio of the risk of disease in the exposed group to the non-exposed group.

[0138] Finally, the attributable risk function is constructed. In this process, the attributable risk functions of different exposure factors are constructed based on the relative risks, and the specific impact of specific exposure factors on the health status of workers is evaluated through the attributable risk function.

[0139] As an embodiment of the present invention: the specific formula of the attributable risk function is:

[0140] CRF=C*RT*SI*IP+C*MD*P(CI0-CI1)

[0141] in:

[0142] C = risk factor, value range [0,6]

[0143] RT = risk transfer rate, range [0,1] dimensionless

[0144] MD = rate of change, value range [-1,1] dimensionless

[0145] P = the probability of potential risk, with a value range of [0,1] dimensionless

[0146] CI0 = safety index before the change, value range [0,1] dimensionless

[0147] CI1 = safety index after change, value range [0,1] dimensionless

[0148] SI = safety index, value range [0,1] dimensionless

[0149] IP = implementation plan, value range [0,1] dimensionless

[0150] Where C=Cy+Ck+Cj+Ct+Cs+Cp

[0151] Cy is the noise pollution hazard coefficient; Ck is the air pollution hazard coefficient; Cj is the building hazard coefficient; Ct is the equipment hazard coefficient; Cs is the water resource hazard coefficient; Cp is the water resource hazard coefficient.

[0152] In actual implementation, a second attributable risk function is also included to determine the attribution probability of different exposure factors:

[0153]

[0154] Where x represents the comprehensive coefficient of exposure factor, risk intensity and level; y represents the duration of exposure factor in the exposure environment; D(X) represents the relative risk under exposure factor x; P x It represents the proportion of people exposed to the exposure factor x; I x represents the risk factor under exposure factor x; N is the total number of workers.

[0155] As an embodiment of the present invention, determining the attributable risk values ​​of different exposure factors relative to the worker according to the attributable risk function includes:

[0156] According to the attributable risk function, the first simulation evolution is performed on different exposure factors in turn; wherein the simulation calculation includes the initial exposure time and the end exposure time of the different exposure factors;

[0157] According to the simulation calculation, the original calculation time of multiple different exposure factors is calculated in turn, and the adjustment time is determined, wherein: there is a proportional coefficient between adjacent adjustment times;

[0158] According to the adjustment time, a second simulation evolution is performed for different exposure factors, and the first simulation evolution result is compared with the second simulation evolution result;

[0159] When the second simulation evolution result is greater than the first simulation evolution result, the target and attributed risk value of the second simulation evolution result are changed;

[0160] When the second simulation evolution result is greater than the first simulation evolution result, the first simulation evolution result or the first simulation evolution result is used as the target attribution risk value;

[0161] When the second simulation evolution result is less than the first simulation evolution result, the first simulation evolution result is used as the target attribution risk value.

[0162] The principle of the above technical solution is:

[0163] In the actual implementation process, during the first simulation evolution, different exposure factors are simulated and evolved for the first time according to the attributable risk function. In this process, the simulation calculation takes into account the initial exposure time and the end time of each exposure factor, simulating the impact of the exposure factor on the health of workers.

[0164] Then, the original calculation time of multiple different exposure factors is adjusted based on the results of the first simulation evolution through the determined adjustment time. The determination of the adjustment time takes into account the proportional coefficient between adjacent adjustment times and simulates the dynamic changes of exposure time more accurately.

[0165] Then the second simulation evolution is carried out, and the second simulation evolution is carried out using the adjusted time. The second simulation evolution is intended to verify the accuracy and stability of the first simulation results. Finally, the results are compared, which is to compare the results of the first and second simulation evolutions to determine which result more accurately reflects the attributable risk of the exposure factor to the health of the workers. The target attributable risk value is determined by selecting a more reliable simulation evolution result as the target attributable risk value based on the comparison results. If the second simulation evolution result is better than the first, the second result is adopted; otherwise, the first result is retained.

[0166] As an embodiment of the present invention: generating a worker health status assessment report includes:

[0167] Configure assessment maps for different exposure factors;

[0168] Based on the evaluation map, the workers' health status is divided into multiple evaluation display pages according to different exposure factors, and each evaluation display page corresponds to the display page of different exposure factors;

[0169] Based on the display page, determine at least one health risk type corresponding to different exposure factors, use the health risk type to connect the different assessment contents of the display page, and generate a visual assessment report.

[0170] The principle of the above scheme is:

[0171] In the actual implementation process, in order to achieve accurate assessment and issue an assessment report, it is necessary to configure an assessment map. First, configure the assessment map according to different exposure factors. The assessment map is based on a pre-set framework or template to organize and display the health risk assessment information related to each exposure factor.

[0172] By dividing the evaluation display page, the worker's health status is divided into multiple evaluation display pages according to different exposure factors based on the evaluation map. Each display page is dedicated to a specific exposure factor and displays the health risk assessment results related to it.

[0173] Finally, the health risk type. In this process, based on each display page, the health risk type corresponding to the exposure factor is determined, such as respiratory diseases, skin diseases, hearing loss, etc.

[0174] Based on the docking assessment content, the different assessment contents of the display pages are docked using the determined health risk types to ensure that the information in each part is associated with a specific health risk. A visual assessment report is generated. In the process of generating a visual assessment report, the information of all display pages is integrated to generate a visual assessment report, making the assessment results intuitive and easy to understand.

[0175] As an embodiment of the present invention, the generating of the worker health status assessment report further includes:

[0176] Displaying at least one visualization area in the visualization evaluation report and setting a trigger instruction; wherein the trigger instruction is used to trigger a corresponding evaluation process;

[0177] When the assessment process is triggered, the corresponding attribution mapping graph is configured according to the worker’s physical risk data;

[0178] According to the attribution mapping map, the exposure distribution status of users in engineering construction is determined, and the exposure factor arrangement table of engineering construction is generated based on the exposure distribution status.

[0179] The principle of the above technical solution is:

[0180] like Figure 3As shown, during the specific implementation process, at least one visualization area is set in the visual assessment report and equipped with a trigger instruction. The trigger instruction can be a click of a button, a selection of a drop-down menu or other interactive elements to activate a specific assessment process. Configuring the attribution mapping map means that when the user activates the assessment process through the trigger instruction, the corresponding attribution mapping map is configured according to the worker's physical risk data. The attribution mapping map is a chart or model that associates workers' health risks with specific exposure factors. Then determine the exposure distribution status. This process is based on the attribution mapping map to determine the user's exposure distribution status in engineering construction, that is, which workers or work areas are affected by which exposure factors, and the extent and scope of these impacts.

[0181] Finally, an exposure factor arrangement table is generated. In this process, an exposure factor arrangement table for engineering construction is generated based on the exposure distribution status. The exposure factor arrangement table records in detail the distribution and impact of each exposure factor in the project.

[0182] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A worker health status assessment method based on AI recognition, characterized in that: include: Obtain health data of workers in construction projects and build health records; Import health records into a pre-configured exposure factor identification network to identify exposure factors in engineering construction and build attributable risk functions for different exposure factors; Based on the attributable risk function, the attributable risk values ​​of different exposure factors relative to the workers are determined, and a worker health status assessment report is generated.

2. A worker health status assessment method based on AI recognition as claimed in claim 1, characterized in that: The health records include personal records and engineering records, including: Personal files are used to obtain workers’ physiological and biochemical data, lifestyle data, personal health history data, and family health history data; Engineering archives are used to obtain exposure environment data during engineering construction. The exposure environment data include: noise data, air pollution data, building safety data, equipment safety risk data, water resource pollution data and temperature status data.

3. A worker health status assessment method based on AI recognition as claimed in claim 2, characterized in that: The exposure environment data also includes the following steps during the acquisition process: Obtain exposure environmental data from historical engineering construction and determine the characteristics of environmental ecological factors; According to the characteristics of environmental ecological factors, determine the time point of occurrence of environmental ecological factors in different environmental ecology; According to the time point of occurrence of environmental and ecological factors, a statistical matrix of environmental ecology is constructed; According to the statistical matrix of environmental ecology, the attribute intervals of different environmental ecological factors are determined; wherein the attribute intervals include time intervals and intensity intervals; According to the attribute range, the distribution patterns of different environmental ecological factors are determined, and distribution labels of different environmental ecological factors are set; among them, the distribution patterns include regional distribution, intensity distribution and time distribution.

4. A worker health status assessment method based on AI recognition as claimed in claim 2, characterized in that: The health record is also configured with a classification filling mechanism, where: Configure the data attributes of various health data in similar configuration modes, including: In the similar configuration mode, there are multiple groups of filling positions in the health record, and the data attributes of each group of filling positions in the multiple groups of filling positions are consistent with the attributes of the target phrase; each of the multiple groups of filling positions is configured with a displacement filling rule; When the worker health data responds to any target phrase attribute, the worker health data and the filling position of the corresponding phrase are classified and traversed, wherein, When the filling rules of any filling position are met after classification traversal, healthy data filling is performed.

5. The method for assessing worker health status based on AI recognition according to claim 1, characterized in that: The method of importing health records into a pre-configured exposure factor identification network to identify exposure factors in engineering construction includes: According to the exposure factor identification network, the environmental ecological data in the health records are identified one by one to determine the environmental ecological characteristics; The strong correlation between different environmental ecological characteristics and workers' health status was determined, and a strong correlation network was constructed, in which: Different environmental ecological characteristics and workers' health status in the strong correlation network have unique connected components. Based on the strong association network, the positive and negative effects of different environmental ecological characteristics on the workers' health status were determined; Based on the negative impacts, the corresponding target environmental ecological characteristics are determined and marked as exposure factors.

6. A worker health status assessment method based on AI recognition as claimed in claim 1, characterized in that: The attributable risk function of different exposure factors is constructed, including: Based on the exposure factors, the risk levels of different exposure factors are predefined, and the exposure behaviors of the corresponding exposure factors are collected. According to the exposure behaviors, the statistical risk rates of different exposure behaviors are determined, where: the statistical risk rate is the incidence rate of different exposure behaviors among workers; Based on the statistical risk rate, the population attributable risk percentage was calculated; According to the population attributable percentage, the relative risk of different exposure behaviors was calculated; Based on the relative risk, construct the attributable risk function of different exposure factors.

7. A worker health status assessment method based on AI recognition as claimed in claim 6, characterized in that: The specific formula of the attributable risk function is: CRF=C*RT*SI*IP+C*MD*P(CI0-CI1) in: C = risk factor, value range [0,6] RT = risk transfer rate, range [0,1] dimensionless MD = rate of change, value range [-1,1] dimensionless P = the probability of potential risk, with a value range of [0,1] dimensionless CI0 = safety index before the change, value range [0,1] dimensionless CI1 = safety index after change, value range [0,1] dimensionless SI = safety index, value range [0,1] dimensionless IP = implementation plan, value range [0,1] dimensionless Where C=Cy+Ck+Cj+Ct+Cs+Cp Cy is the noise pollution hazard coefficient; Ck is the air pollution hazard coefficient; Cj is the building hazard coefficient; Ct is the equipment hazard coefficient; Cs is the water resource hazard coefficient; Cp is the water resource hazard coefficient.

8. The method for assessing worker health status based on AI recognition according to claim 1, characterized in that: Determining the attributable risk values ​​of different exposure factors relative to the workers according to the attributable risk function includes: According to the attributable risk function, the first simulation evolution is performed on different exposure factors in turn; wherein the simulation calculation includes the initial exposure time and the end exposure time of the different exposure factors; According to the simulation calculation, the original calculation time of multiple different exposure factors is calculated in turn, and the adjustment time is determined, wherein: there is a proportional coefficient between adjacent adjustment times; According to the adjustment time, a second simulation evolution is performed for different exposure factors, and the first simulation evolution result is compared with the second simulation evolution result; When the second simulation evolution result is greater than the first simulation evolution result, the target and attributed risk value of the second simulation evolution result are changed; When the second simulation evolution result is greater than the first simulation evolution result, the first simulation evolution result or the first simulation evolution result is used as the target attribution risk value; When the second simulation evolution result is less than the first simulation evolution result, the first simulation evolution result is used as the target attribution risk value.

9. A worker health status assessment method based on AI recognition as claimed in claim 1, characterized in that: The worker health status assessment report is generated, including: Configure assessment maps for different exposure factors; Based on the evaluation map, the workers' health status is divided into multiple evaluation display pages according to different exposure factors, and each evaluation display page corresponds to the display page of different exposure factors; Based on the display page, determine at least one health risk type corresponding to different exposure factors, use the health risk type to connect the different assessment contents of the display page, and generate a visual assessment report.

10. The method for assessing worker health status based on AI recognition according to claim 1, characterized in that: The generating of the worker health status assessment report also includes: Displaying at least one visualization area in the visualization evaluation report and setting a trigger instruction; wherein the trigger instruction is used to trigger a corresponding evaluation process; When the assessment process is triggered, the corresponding attribution mapping graph is configured according to the worker’s physical risk data; According to the attribution mapping map, the exposure distribution status of users in engineering construction is determined, and the exposure factor arrangement table of engineering construction is generated based on the exposure distribution status.

Images