Fatigue risk level evaluation system and device and storage medium

Abstract

The invention provides a fatigue risk level evaluation system and device and a storage medium, and relates to the technical field of data processing, and the evaluation system comprises an acquisition module which is used for obtaining basic data of a to-be-tested person, the processing module is used for performing feature extraction on the sleep feature data and the health behavior data to obtain first feature data and second feature data; fusing the five-dimensional time-frequency feature data, the first feature data and the second feature data to obtain fused feature data; the prediction module is used for inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data; and the evaluation module is used for determining the fatigue risk level of the to-be-tested person according to the individual basic data and the fatigue data based on preset risk level data. According to the method, evaluation is carried out by integrating multiple data sources such as basic data, sleep features, five-dimensional time-frequency features and health behaviors, the fatigue state of an individual can be comprehensively reflected, and information loss possibly caused by a single data source is overcome.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a fatigue risk level assessment system, device and storage medium. Background Art

[0002] In recent years, due to various factors, workers in many industries (such as public security, medical health, transportation and industrial production) have suffered rapid deterioration in health due to long working hours, and even sudden death and other malignant health events. Therefore, it is particularly important to assess the fatigue risk level of personnel.

[0003] However, there are currently limitations to the calibration rules for fatigue assessment for different groups, such as the significantly different nature of work and stress levels in different occupations. For example, police officers and doctors may show different fatigue manifestations after high-intensity, long-term work. Current assessment standards are often developed based on a specific group (such as drivers) and lack applicability to other industries (such as medical care, firefighting, etc.), which limits the accuracy and effectiveness of these standards in different occupational contexts.

[0004] Differences in physiological characteristics of the human body (such as age, gender, and health status) can affect the manifestation of fatigue and how it is dealt with. Most existing fatigue assessment tools fail to fully consider these individual differences. For example, young and old people may have very different physiological responses and psychological effects of fatigue when facing the same workload, and existing assessment criteria are generally poor in distinguishing these differences. Summary of the invention

[0005] The problem solved by the present disclosure is one or more of the above-mentioned prior art problems.

[0006] In order to solve the above problems, the present disclosure provides a fatigue risk level assessment system, device and storage medium.

[0007] In a first aspect, the present disclosure provides a fatigue risk level assessment system, comprising: The acquisition module is used to obtain basic data of the person to be tested, wherein the basic data includes individual basic data, sleep characteristic data at preset time intervals, five-dimensional time-frequency characteristic data and health behavior data; A processing module, configured to extract features from the sleep feature data and the health behavior data respectively to obtain first feature data and second feature data; and to fuse the five-dimensional time-frequency feature data, the first feature data and the second feature data to obtain fused feature data; A prediction module, used for inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data of the person to be tested; the preset fatigue prediction model is constructed based on a fixed effect regression model; An evaluation module is used to determine the fatigue risk level of the person to be tested based on preset risk level data, the individual basic data and the fatigue data.

[0008] Optionally, the individual basic data includes occupation information and basic information; the evaluation module is specifically used for: Filtering the preset risk level data by the occupation information to obtain a corresponding risk level data set; The risk level data set is filtered through the basic information to obtain temporary risk level data, and based on the fatigue data, the temporary risk level data is screened to obtain the fatigue risk level of the person to be tested.

[0009] Optionally, the processing module further includes an analysis unit; The analysis unit is used to perform correlation analysis on the first feature data and the second feature data respectively to obtain corresponding temporary feature data.

[0010] Optionally, the processing module further includes a fusion unit; The fusion unit is used to fuse all the temporary feature data with the five-dimensional time-frequency feature data to obtain the fused feature data.

[0011] Optionally, the five-dimensional time-frequency feature data includes time-domain type features and frequency-domain type features; and fusing all the temporary feature data with the five-dimensional time-frequency feature data to obtain the fused feature data includes: Evaluate the time domain type features, the frequency domain type features and each of the temporary feature data respectively to obtain a feature weight matrix; Based on the feature weight matrix, all the temporary feature data are fused with the five-dimensional time-frequency feature data to obtain the fused feature data.

[0012] Optionally, the preset fatigue prediction model includes a fatigue non-sleep module and a fatigue sleep module; and both the fatigue non-sleep module and the fatigue sleep module are constructed based on a fixed effect regression model.

[0013] Optionally, the five-dimensional time-frequency feature data includes time-domain type features and frequency-domain type features; and the step of obtaining the five-dimensional time-frequency feature data of the person to be tested includes: Collecting volume pulse wave data of the person to be tested through a smart wearable device, and preprocessing the volume pulse wave data to obtain temporary pulse wave data; Extracting the temporary pulse wave data to obtain the time domain type feature; The temporary pulse wave data is subjected to waveform conversion to obtain the frequency domain type feature.

[0014] Optionally, the fatigue risk level assessment system further includes an early warning module; The early warning module is used to issue an early warning prompt based on the fatigue risk level of the person to be tested.

[0015] In a second aspect, the present disclosure provides a fatigue risk level assessment device, including a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the following steps when executing the computer program: Obtaining basic data of the person to be tested, the basic data including individual basic data, sleep characteristic data at preset time intervals, five-dimensional time-frequency characteristic data and health behavior data; Extracting features from the sleep feature data and the health behavior data respectively to obtain first feature data and second feature data; Fusing the five-dimensional time-frequency feature data, the first feature data, and the second feature data to obtain fused feature data; Inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data of the person to be tested; Based on the preset risk level data, the fatigue risk level of the person to be tested is determined according to the individual basic data and the fatigue data.

[0016] In a third aspect, the present disclosure provides a computer-readable storage medium, characterized in that a computer program is stored on the storage medium, and when the computer program is executed by a processor, the following steps are implemented: Obtaining basic data of the person to be tested, the basic data including individual basic data, sleep characteristic data at preset time intervals, five-dimensional time-frequency characteristic data and health behavior data; Extracting features from the sleep feature data and the health behavior data respectively to obtain first feature data and second feature data; Fusing the five-dimensional time-frequency feature data, the first feature data, and the second feature data to obtain fused feature data; Inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data of the person to be tested; Based on the preset risk level data, the fatigue risk level of the person to be tested is determined according to the individual basic data and the fatigue data.

[0017] The beneficial effects of the fatigue risk level assessment system disclosed in the present invention are: The acquisition module is used to obtain the multi-dimensional basic data of the person to be tested in order to provide the necessary information for subsequent feature extraction and analysis. Among them, the basic data types are: Individual basic data: including basic information such as age, gender, weight, height, etc., which are helpful to understand the individual's physical condition and fatigue performance. Sleep characteristic data: data such as sleep quality, duration, depth, etc. within a preset time interval (for example, the past week). These data can be obtained through devices such as smart bracelets and sleep monitors. Five-dimensional time-frequency feature data: five-dimensional features extracted by analyzing physiological signals (such as brain waves, heart rate, etc.) to reflect the individual's physiological state in different frequency bands. Health behavior data: including data such as body mass index and average heart rate to evaluate the impact of individual health behaviors on fatigue.

[0018] The processing module extracts the sleep feature data. For example, it can extract key indicators in the sleep feature data through algorithms (such as time domain analysis and frequency domain analysis) to form the first feature data. For the health behavior data processing process, statistical analysis or machine learning methods can be used to extract important features in the health behavior data to form the second feature data. The extracted first feature data, second feature data and five-dimensional time-frequency feature data are then fused to form a comprehensive fused feature data set. This process may use techniques such as feature splicing and weighted fusion to ensure maximum integration of information between different features.

[0019] The prediction module inputs the fused feature data into the preset fatigue prediction model to generate fatigue data of the person to be tested. The preset fatigue prediction model built based on the fixed effect regression model can control the influence of individual invariant features, emphasize the changes of individuals caused by the environment or time at different time points, and thus more accurately capture the relationship between fatigue and input features. That is, based on the selection of the fixed effect regression model, the system can more accurately reflect the various factors affecting fatigue, while reducing the interference of individual inherent differences on the results, and improving the accuracy of fatigue level prediction.

[0020] The assessment module determines the fatigue risk level of the person being tested based on individual basic data and fatigue data, combined with preset risk level data (such as mild, moderate, severe fatigue, etc.).

[0021] Finally, the system will output the individual's fatigue risk level and may provide corresponding suggestions or intervention measures to help the individual improve his or her fatigue status.

[0022] In summary, the system of the present invention can comprehensively reflect the fatigue state of an individual by evaluating through a combination of multiple data sources (basic data, sleep characteristics, five-dimensional time-frequency characteristics, and health behaviors), overcoming the information loss that may be caused by a single data source. And through feature extraction and fusion, the system can capture the complex characteristics of the fatigue state, so that the accuracy of the fatigue prediction model is significantly improved, thereby more effectively identifying the fatigue risk of an individual. Finally, based on the individual basic data and fatigue data, the evaluation can provide personalized fatigue risk levels for different individuals, that is, improve the accuracy and effectiveness in different occupational backgrounds, and help formulate more targeted health management strategies. And the system can continuously monitor the fatigue state of an individual within a preset time interval, and timely feedback the risk level, so as to promote the individual's attention to and management of their own health status. And by identifying the individual's fatigue risk level, the system can provide users with corresponding suggestions, such as improving sleep quality, adjusting daily activities, etc., so as to intervene before the fatigue problem worsens and reduce the potential harm of fatigue to health.

[0023] Therefore, the individual fatigue risk level assessment system provides a comprehensive, accurate and personalized fatigue status assessment solution through multi-dimensional data collection, feature extraction and fusion, fatigue prediction and risk assessment. Its beneficial effects are significant, and it can provide important support for individual health management, promote the improvement of healthy behavior and effective management of fatigue. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a structural schematic diagram of a fatigue risk level assessment system according to an embodiment of the present disclosure; Figure 2 It is a structural schematic diagram of a fatigue risk level assessment system according to an embodiment of the present disclosure; Figure 3 is a schematic diagram of temporary pulse wave data according to an embodiment of the present disclosure; Figure 4 It is a schematic diagram of the structure of a fatigue risk level assessment device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] In order to make the above-mentioned purposes, features and advantages of the present disclosure more obvious and easy to understand, the specific embodiments of the present disclosure are described in detail below in conjunction with the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be interpreted as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0026] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0027] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0028] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0029] This embodiment provides a fatigue risk level assessment system, device, and storage medium.

[0030] like Figure 1 As shown, a fatigue risk level assessment system provided by an embodiment of the present disclosure includes: The acquisition module is used to obtain basic data of the person to be tested, wherein the basic data includes individual basic data, sleep characteristic data at preset time intervals, five-dimensional time-frequency characteristic data and health behavior data; Specifically, the acquisition module is the core component of the fatigue risk level assessment system, which aims to obtain the basic data of the person to be tested in order to conduct a comprehensive fatigue status analysis. The module will collect basic data including: 1. Individual basic data includes basic information of the person to be tested, such as age, gender, weight, height and occupation, etc. This information helps to consider individual differences in subsequent analysis and ensure the accuracy and pertinence of the evaluation results.

[0031] 2. Sleep characteristic data at preset time intervals: Wearable devices or smartphone applications can be used to regularly record the sleep patterns of the test subjects, including the time of falling asleep, waking up, sleep duration, the proportion of deep sleep and light sleep, etc. These data can reflect the relationship between the individual's sleep quality and fatigue status, and help identify fatigue caused by insufficient or poor sleep quality.

[0032] 3. Five-dimensional time-frequency feature data: This data extracts five-dimensional time-frequency features by analyzing the individual's physiological signals (such as heart rate, breathing rate, brain waves, etc.) in a specific time period. These features can reveal the individual's physiological response under different workloads and provide a more detailed basis for fatigue assessment.

[0033] 4. Health behavior data: This data includes the daily living habits and health behaviors of the person being tested, such as diet, exercise, smoking and drinking. These factors have an important impact on the individual's fatigue state, so considering these behavioral data during the assessment process can provide a more comprehensive understanding of the causes of fatigue.

[0034] By collecting individual basic data and health behavior data, the system can conduct personalized fatigue assessments based on each user's unique situation and provide more targeted intervention recommendations. The system also regularly collects sleep characteristic data and five-dimensional time-frequency characteristic data, allowing the system to monitor changes in the user's fatigue status in real time, promptly identify the risk of fatigue accumulation, and prevent potential health problems. Finally, combined with multi-dimensional data, the system can conduct a comprehensive analysis from multiple physiological and behavioral perspectives, provide a more comprehensive fatigue risk assessment, and help users better understand their own fatigue status.

[0035] The basic data collected provides a solid foundation for subsequent research and analysis, and can help related industries develop more scientific fatigue management strategies and improve work efficiency and safety.

[0036] Overall, the design and implementation of the acquisition module will significantly improve the effectiveness and practicality of the fatigue risk level assessment system.

[0037] A processing module, configured to extract features from the sleep feature data and the health behavior data respectively to obtain first feature data and second feature data; and to fuse the five-dimensional time-frequency feature data, the first feature data and the second feature data to obtain fused feature data; Specifically, the processing module is an important part of the fatigue risk level assessment system, and is responsible for in-depth analysis and processing of the collected sleep characteristic data, health behavior data, and five-dimensional time-frequency characteristic data.

[0038] By analyzing the sleep characteristic data, key features are extracted, such as average sleep duration, deep sleep duration, sleep delay, number of awakenings, etc. These indicators can reflect the individual's sleep quality and impact on fatigue, and generate the first characteristic data.

[0039] At the same time, through the analysis of health behavior data, relevant variables are extracted, such as daily exercise duration, body mass index (BMI), cumulative training duration, etc. These factors directly affect fatigue level and recovery ability, thereby generating the second characteristic data.

[0040] The acquired five-dimensional time-frequency feature data, the first feature data (sleep feature) and the second feature data (health behavior) are fused. In this step, appropriate data fusion algorithms (such as weighted average, principal component analysis, etc.) are used to transform these data from different sources into a comprehensive fusion feature data. This data set reflects the user's comprehensive fatigue state, taking into account the influence of multiple aspects such as physiology, behavior and sleep.

[0041] By integrating multiple data sources and features through the processing module, it can comprehensively reflect the fatigue status of individuals, avoid the one-sidedness and defects that may be caused by a single data source, and provide a more comprehensive and detailed perspective for fatigue assessment. The extracted and fused feature data can accurately capture the fatigue status of individuals and their influencing factors, helping the system to generate a more accurate fatigue risk level assessment, thereby providing a scientific basis for individual health management.

[0042] At the same time, after collecting new sleep characteristics and health behavior data, the processing module can update the characteristic data in a timely manner, adapt to changes in individual conditions, realize dynamic fatigue monitoring and evaluation, and improve the response speed and adjustment ability of the system. The fusion characteristic data provides rich input information for subsequent model training and fatigue risk prediction, which can improve the prediction accuracy of the model and promote the formulation and implementation of individual fatigue management strategies.

[0043] In summary, the processing module not only improves the comprehensiveness and accuracy of fatigue assessment through the workflow of feature extraction and data fusion, but also provides favorable support for individual health management and work safety.

[0044] A prediction module, used for inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data of the person to be tested; the preset fatigue prediction model is constructed based on a fixed effect regression model; Specifically, the prediction module is mainly responsible for inputting the processed fusion feature data into the preset fatigue prediction model to generate fatigue data of the person to be tested. The processed fusion feature data (including five-dimensional time-frequency features, sleep feature data and health behavior data) is used as input to provide comprehensive information about the individual's current physiological state, sleep quality and living habits.

[0045] This module can use a fatigue prediction model based on a fixed effect regression model. The fixed effect regression model can control the influence of individual invariant characteristics and emphasize the changes of individuals at different time points caused by the environment or time, so as to more accurately capture the relationship between fatigue and input characteristics.

[0046] The model will use the fusion feature data to calculate the fatigue data of the person under test through regression analysis. Based on historical data, individual behavior habits and physiological reactions, the model outputs a fatigue score or level to represent the individual's current fatigue status.

[0047] Ultimately, the fatigue data generated by the prediction module can be a quantitative fatigue score, probability value, or classification level, which is convenient for subsequent interpretation and application.

[0048] Based on the selection of the fixed effect regression model, the system can more accurately reflect the various factors that affect fatigue, while reducing the interference of individual inherent differences on the results, and improving the accuracy of fatigue level prediction. And because the fixed effect model can adapt to the changes in input data at different points in time, the prediction module can respond to changes in individual status in real time, providing individuals with more personalized and timely fatigue assessment.

[0049] Therefore, the prediction module can provide high-quality fatigue data by inputting the fused feature data into the fixed-effect regression model, which can provide strong support for individuals to identify and respond to fatigue status in a timely manner, thereby optimizing health management strategies.

[0050] An evaluation module is used to determine the fatigue risk level of the person to be tested based on preset risk level data, the individual basic data and the fatigue data.

[0051] Specifically, the main function of the assessment module is to assess the fatigue risk level of the person to be tested based on pre-set risk level data, individual basic data and fatigue data.

[0052] The module first obtains the individual basic data (such as age, gender, occupation, health status, etc.) and fatigue data (fatigue score or level generated by the prediction module) of the person to be tested from the system. Based on the preset risk level model, the module considers individual differences and fatigue influencing factors, and combines the obtained individual basic data with the fatigue data through an algorithm. These risk level data are usually compiled based on historical data and research results, and include risk level divisions corresponding to different fatigue scores.

[0053] At the same time, individual basic data and fatigue data are analyzed, and statistical models or machine learning algorithms (such as logistic regression models, decision trees, etc.) can be used to calculate the fatigue risk level of the person to be tested. In this process, some weighting factors may be involved to better reflect the contribution of different data sources to fatigue risk.

[0054] Finally, through calculation and analysis, the assessment module can generate the fatigue risk level of the person being tested, which may be presented in the form of "low risk", "medium risk", "high risk" or specific scores. This result will provide guidance for subsequent health intervention and management.

[0055] That is, by comprehensively considering individual basic data and fatigue data, the assessment module can provide a scientific and reliable fatigue risk level to help users accurately understand their own fatigue status. And based on the assessment results, the system can provide users with targeted health management suggestions, such as improving sleep habits, adjusting workload, increasing exercise, etc., so as to effectively prevent and deal with fatigue problems.

[0056] The assessment module can quickly identify potential high-risk individuals, allowing companies or organizations to take timely intervention measures to reduce the risk of health problems and work accidents caused by fatigue. At the same time, the assessment module can dynamically adjust the risk assessment as new data is input to reflect the changes in the fatigue status of individuals at different time points. This dynamic monitoring capability is crucial for real-time health management and the implementation of intervention measures.

[0057] In summary, the assessment module provides accurate fatigue risk level assessment for the tested personnel through scientific calculation and analysis, supports personalized health management, and promotes early risk identification and organizational health decision-making, making positive contributions to improving the health level and work efficiency of individuals and groups.

[0058] In this embodiment, the multi-dimensional basic data of the person to be tested is obtained through the acquisition module to provide necessary information for subsequent feature extraction and analysis. Among them, basic data types: individual basic data: including basic information such as age, gender, weight, height, etc., which are helpful to understand the individual's physical condition and fatigue performance. Sleep feature data: data such as sleep quality, duration, depth, etc. within a preset time interval (for example, the past week). These data can be obtained through devices such as smart bracelets and sleep monitors. Five-dimensional time-frequency feature data: five-dimensional features extracted by analyzing physiological signals (such as brain waves, heart rate, etc.) to reflect the individual's physiological state in different frequency bands. Health behavior data: including data such as body mass index and average heart rate to evaluate the impact of individual health behaviors on fatigue.

[0059] The processing module extracts the sleep feature data. For example, it can extract key indicators in the sleep feature data through algorithms (such as time domain analysis and frequency domain analysis) to form the first feature data. For the health behavior data processing process, statistical analysis or machine learning methods can be used to extract important features in the health behavior data to form the second feature data. The extracted first feature data, second feature data and five-dimensional time-frequency feature data are then fused to form a comprehensive fused feature data set. This process may use techniques such as feature splicing and weighted fusion to ensure maximum integration of information between different features.

[0060] The prediction module inputs the fused feature data into the preset fatigue prediction model to generate fatigue data of the person to be tested. The preset fatigue prediction model built based on the fixed effect regression model can control the influence of individual invariant features, emphasize the changes of individuals caused by the environment or time at different time points, and thus more accurately capture the relationship between fatigue and input features. That is, based on the selection of the fixed effect regression model, the system can more accurately reflect the various factors affecting fatigue, while reducing the interference of individual inherent differences on the results, and improving the accuracy of fatigue level prediction.

[0061] The assessment module determines the fatigue risk level of the person being tested based on individual basic data and fatigue data, combined with preset risk level data (such as mild, moderate, severe fatigue, etc.).

[0062] Finally, the system will output the individual's fatigue risk level and may provide corresponding suggestions or intervention measures to help the individual improve his or her fatigue status.

[0063] In summary, the system of the present invention can comprehensively reflect the fatigue state of an individual by evaluating through a combination of multiple data sources (basic data, sleep characteristics, five-dimensional time-frequency characteristics, and health behaviors), overcoming the information loss that may be caused by a single data source. And through feature extraction and fusion, the system can capture the complex characteristics of the fatigue state, so that the accuracy of the fatigue prediction model is significantly improved, thereby more effectively identifying the fatigue risk of an individual. Finally, based on the individual basic data and fatigue data, the evaluation can provide personalized fatigue risk levels for different individuals, that is, improve the accuracy and effectiveness in different occupational backgrounds, and help formulate more targeted health management strategies. And the system can continuously monitor the fatigue state of an individual within a preset time interval, and timely feedback the risk level, so as to promote the individual's attention to and management of their own health status. And by identifying the individual's fatigue risk level, the system can provide users with corresponding suggestions, such as improving sleep quality, adjusting daily activities, etc., so as to intervene before the fatigue problem worsens and reduce the potential harm of fatigue to health.

[0064] Therefore, the individual fatigue risk level assessment system provides a comprehensive, accurate and personalized fatigue status assessment solution through multi-dimensional data collection, feature extraction and fusion, fatigue prediction and risk assessment. Its beneficial effects are significant, and it can provide important support for individual health management, promote the improvement of healthy behavior and effective management of fatigue.

[0065] Optionally, the individual basic data includes occupation information and basic information; the evaluation module is specifically used for: Filtering the preset risk level data by the occupation information to obtain a corresponding risk level data set; The risk level data set is filtered through the basic information to obtain temporary risk level data, and based on the fatigue data, the temporary risk level data is screened to obtain the fatigue risk level of the person to be tested.

[0066] Specifically, during the process of individual fatigue risk level assessment, the specific assessment module systematically screens and filters data to determine the fatigue risk level of the person to be tested.

[0067] Individual basic data mainly include: occupational information: for example, the individual's position, industry, nature of work (physical labor, mental labor, etc.) and working environment (such as high-pressure work, shift work, etc.).

[0068] Basic information mainly includes age, gender, health status (whether there are chronic diseases), etc.

[0069] First, through occupational information screening: According to the occupational information of the person to be tested, the risk level data set related to the occupation is screened from the stored preset risk level data. Occupational characteristics are usually strongly correlated with fatigue risk. For example, the risk level data corresponding to high-risk occupations (doctors, firefighters, transport drivers, etc.) may show that they have higher fatigue scores under specific working conditions. Low-risk occupations (office workers) have lower risk scores under normal working conditions. This step ensures that the assessment is based on highly relevant occupational data, which enhances the accuracy of the assessment.

[0070] Then filter through basic information: Individuals may have different susceptibility to fatigue depending on age and gender. Therefore, older individuals may need to consider age-related fatigue risks. Previous health conditions (such as high blood pressure, diabetes, etc.) will affect the assessment results of fatigue scores, so they will be taken into account at this stage. This process helps eliminate irrelevant or outdated risk data and ensure the accuracy and timeliness of temporary data.

[0071] Finally, the actual fatigue score of the person to be tested is combined with the temporary risk level data for final screening. The final individual fatigue risk level is determined based on the applicability of each temporary risk level data. That is, the fatigue score is matched with the temporary risk level to identify the risk level that best reflects the actual fatigue status of the person to be tested. For example, if the fatigue score of the person to be tested is high, but the basic information shows that he or she belongs to a high-risk occupation, the system will classify him or her as a high fatigue risk level.

[0072] Based on the analysis results of the assessment module, the fatigue risk level of the person to be tested is finally output, and targeted suggestions are provided, such as health intervention plans for high-risk personnel, or warnings and suggestions for medium-risk personnel, encouraging them to take active health maintenance measures.

[0073] This assessment module can provide personalized fatigue risk analysis for each individual, taking into full account the impact of occupation and basic information. This makes the assessment more accurate and can reflect the actual situation of the individual. And through multi-layer screening and filtering, it ensures that factors related to the person to be tested are taken into consideration to avoid erroneous judgments caused by data noise. Ensure that the reliability of the assessment results is improved.

[0074] At the same time, after identifying high-risk individuals, timely formulating corresponding intervention measures can reduce work errors and health problems caused by fatigue, improve overall work efficiency and individual quality of life. And the standardization of the assessment process allows regular tracking and re-evaluation, so that individuals can get timely feedback and help them make timely adjustments in work and life.

[0075] Therefore, the assessment module systematically screens and filters occupational information, basic information and fatigue data to ensure that the individual's fatigue risk level is obtained. This process not only improves the accuracy and personalization of the assessment, but also provides strong support for health management and intervention, helping to improve the work efficiency and quality of life of the tested personnel.

[0076] Optionally, the individual basic data includes occupation information and basic information, and the evaluation module specifically includes: Establishing a risk level model, and processing the occupational information and the basic information described in the risk level model to obtain risk assessment data; Based on the preset risk level data, the fatigue risk level of the person to be tested is determined according to the risk assessment data.

[0077] Specifically, the occupational information includes the occupation of the individual, which is classified into high-risk occupations (such as medical staff, pilots, shift workers, etc.) and low-risk occupations (such as office clerks, teachers, etc.).

[0078] Work environment: For example, whether overtime is frequent, working hours (weekly working hours), nature of work (physical labor, mental labor), etc. This information helps to understand the impact of occupation on fatigue.

[0079] The basic information includes the individual's age, gender, height, weight, health history (such as previous fatigue symptoms, chronic diseases), etc. This basic information helps to judge the individual's health status and sensitivity to fatigue.

[0080] The collected individual basic data (occupational information and basic information) and fatigue data are integrated to form a comprehensive data set. This data set will contain fields such as individual ID, occupational category, age, gender, health status, fatigue data, etc.

[0081] Establish a preset risk level model, and establish risk level standards (preset risk level data) corresponding to different fatigue scores based on clinical research and industry data, such as: Low risk: Fatigue score 0-30, indicating little or no fatigue.

[0082] Medium risk: Fatigue score of 31-60 indicates increased fatigue, which may affect work efficiency.

[0083] High risk: Fatigue scores of 61-100 indicate high levels of fatigue that require immediate attention and intervention.

[0084] Using the preset risk level standard, input individual basic data and fatigue data, and run the assessment algorithm. Analyze individual basic data: for example, identify whether the individual belongs to a high-risk occupation. High-risk occupations are usually associated with higher fatigue scores. Analyze fatigue data: Put the individual's fatigue data into the preset model, combine it with the basic data, and conduct a comprehensive analysis.

[0085] Through the above process, effective health management suggestions are provided to individuals to help them improve their fatigue and enhance the quality of life and work. This data-driven model can strengthen the focus on individual health and provide targeted intervention strategies.

[0086] Alternatively, if Figure 2 As shown, the processing module also includes an analysis unit; The analysis unit is used to perform correlation analysis on the first feature data and the second feature data respectively to obtain corresponding temporary feature data.

[0087] Optionally, the processing module further includes a fusion unit; The fusion unit is used to fuse all the temporary feature data with the five-dimensional time-frequency feature data to obtain the fused feature data.

[0088] Optionally, the five-dimensional time-frequency feature data includes time-domain type features and frequency-domain type features; and fusing all the temporary feature data with the five-dimensional time-frequency feature data to obtain the fused feature data includes: Evaluate the time domain type features, the frequency domain type features and each of the temporary feature data respectively to obtain a feature weight matrix; Based on the feature weight matrix, all the temporary feature data are fused with the five-dimensional time-frequency feature data to obtain the fused feature data.

[0089] Specifically, in individual fatigue risk assessment, the fusion of five-dimensional time-frequency feature data and temporary feature data can help improve the depth and accuracy of data analysis.

[0090] Time domain type features usually include the performance of the signal in the time domain, such as: Mean: the average value of the signal, which can reflect the overall level. Variance: the degree of signal change, describing the volatility of the signal. Peak: reflects the highest point of the signal, usually used to evaluate extreme fatigue status.

[0091] Frequency domain features usually include Power Spectral Density (PSD): the distribution of the frequency components of the signal, which can often reveal features that help understand fatigue. Frequency Distribution: Analyzing the energy distribution of each frequency component in the signal helps determine the dominant frequency of the fatigue state.

[0092] Through correlation analysis, variance analysis or modeling techniques (such as linear regression, logistic regression, etc.), the importance of these features in assessing fatigue risk is obtained, and the feature weight matrix is ​​calculated.

[0093] Statistical analysis methods are used to evaluate each feature (including time domain, frequency domain and temporary features) to determine its relative importance in fatigue risk assessment.

[0094] A weight matrix is ​​established, in which each element represents the weight value of the corresponding feature, reflecting its contribution to the overall evaluation result.

[0095] For example, assuming that the weight of the time domain feature is 0.4, the weight of the frequency domain feature is 0.3, and the weight of the temporary feature is 0.3, the following weight matrix is ​​formed: ; Different features are weighted according to the feature weight matrix to integrate all temporary feature data and five-dimensional time-frequency feature data. The fusion of any feature can be achieved by the following formula: ; Among them, w1, w2, and w3 are the weight values ​​of the features respectively.

[0096] Through the above method, the fused feature data finally obtained is the result of comprehensive consideration of time domain, frequency domain and temporary features, which can improve the accuracy of fatigue state assessment.

[0097] It should be noted that the weights of the time domain features in the weight matrix are the weights corresponding to the number of different features. For example, if the number of time domain features is 3, then there will be corresponding weight values, and the weights of the frequency domain features and the temporary features are the same as described above.

[0098] The fusion of multi-dimensional features can more comprehensively reflect the fatigue status of an individual, especially considering the different effects of different features on the fatigue status, which can improve the accuracy of the assessment. And by combining time domain, frequency domain and temporary features, the model can capture more behavioral patterns and changing trends, providing more detailed analysis. At the same time, through the personalized evaluation of feature weights, targeted analysis can be performed on the differences and characteristics of different individuals, improving the flexibility of personalized health management.

[0099] After feature fusion, the system can dynamically monitor changes in individual fatigue levels, adjust health management strategies or clinical interventions in a timely manner, and improve the timeliness of management. The specific fusion feature data provides data support for fatigue management decisions, so that intervention measures can be formulated more reasonably, work and life arrangements can be optimized, and the impact of fatigue on life can be reduced.

[0100] Therefore, by comprehensively integrating temporary feature data with five-dimensional time-frequency feature data and using detailed calculations of the feature weight matrix, this process effectively improves the scientificity and accuracy of fatigue risk assessment, thereby promoting the effectiveness of individual health management.

[0101] Optionally, the preset fatigue prediction model includes a fatigue non-sleep module and a fatigue sleep module.

[0102] Specifically, the preset fatigue prediction model can be designed for different usage scenarios: Single-day work (no long sleep at night): In this case, the input data is processed through the fatigue non-sleep module to assess the fatigue level.

[0103] Continuous work for multiple days (with long sleep at night): In this case, the input data is processed by the fatigue sleep module to analyze the impact of sleep on fatigue.

[0104] In these two different scenarios, a fixed-effect regression method that varies with individuals but not with time in spatial panel data is used to construct a fatigue-sleep model and a fatigue-non-sleep model.

[0105] In some embodiments, a single day of work (without long sleep at night), i.e., a fatigue non-sleep module; The features used in the fatigue non-sleep module are shown in Table 1.

[0106] Table 1 Application data of fatigue non-sleep module

[0107] These three characteristic variables are physiological indicators related to fatigue and can reflect the individual's fatigue level at different time points. The explanations in the model are as follows: Cumulative training time (s22, obtained through health behavior data): This is an individual characteristic that does not change over time and is considered as a control variable in the model. It reflects the individual's training accumulation over the past period of time and has a certain impact on the individual's fatigue level. Introducing cumulative training time as a control variable can enable the model to more accurately evaluate the impact of other explanatory variables on fatigue level.

[0108] Average heart rate within 1 hour (s20, obtained through health behavior data): This is a physiological indicator that changes over time. The change in heart rate is closely related to the individual's fatigue state and is one of the important indicators for assessing the degree of fatigue.

[0109] Wavelet energy ratio (s7, obtained through frequency domain type data, that is, five-dimensional time-frequency feature data): This physiological indicator reflects the frequency component and energy distribution of the PPG signal, thereby providing richer physiological information and further improving the fatigue prediction accuracy of the model.

[0110] The calculation process of the fatigue non-sleep module is as follows, and the corresponding parameters are shown in Table 2.

[0111] Table 2 Parameters of fatigue non-sleep module

[0112] The final fatigue state model calculation formula is:

[0113] Continuous work for many days (with long sleep at night), that is, fatigue sleep module.

[0114] The features used in the fatigue sleep module are shown in Table 3.

[0115] Table 3 Application data of fatigue sleep module

[0116] In the fatigue sleep module it is explained as follows: BMI (s21, obtained through health behavior data): This is an individual characteristic that does not change over time, but is considered as a control variable in the model. BMI reflects the relationship between an individual's weight and height and is one of the important indicators for assessing an individual's health status. Although BMI itself does not directly reflect the degree of fatigue, it may indirectly affect the fatigue state by affecting the individual's basal metabolic rate, cardiopulmonary function, etc.

[0117] Cumulative training time (s22, obtained through health behavior data): This is an individual characteristic that does not change over time and is an important control variable in the model. Cumulative training time reflects the individual's training accumulation over the past period of time and has an important impact on the individual's physical reserve and fatigue recovery ability.

[0118] The number of deep sleeps (s16, obtained through sleep characteristic data) and the number of light sleeps (s17, obtained through sleep characteristic data): These two indicators are physiological indicators that change over time and can directly reflect the sleep quality and fatigue level of an individual in the current time period. Deep sleep is an important stage in the sleep cycle to restore physical strength, while light sleep is relatively light and easily disturbed by the outside world. Therefore, changes in the number of deep sleeps and light sleeps can reflect the fatigue state of an individual at different time points.

[0119] The calculation process of the fatigue sleep module is as follows, and the corresponding parameters are shown in Table 4.

[0120] Table 4 Parameters of fatigue sleep module

[0121] The final fatigue state model calculation formula is:

[0122] It should be noted that, on the one hand, the fixed effects model assumes that individual effects are fixed, that is, the specific effects produced by different individuals in the study are constant and do not change over time. A unique intercept term is set for each individual to capture this individual difference that does not change over time. These intercept terms are regarded as fixed parameters in the model, representing the effects that are unique to each individual and do not change over time.

[0123] On the other hand, although the intercept term of each individual is different, the marginal effect of the explanatory variable on the explained variable is the same among individuals. This means that the model focuses on how the explanatory variable affects the explained variable over time after controlling individual differences. Therefore, the fixed effect model can reflect the characteristics of changing with individuals but not changing with time, that is, individual differences are constant, while the impact of the explanatory variable changes with time.

[0124] Optionally, the five-dimensional time-frequency feature data includes time-domain type features and frequency-domain type features; and the step of obtaining the five-dimensional time-frequency feature data of the person to be tested includes: Collecting volume pulse wave data of the person to be tested through a smart wearable device, and preprocessing the volume pulse wave data to obtain temporary pulse wave data; Extracting the temporary pulse wave data to obtain the time domain type feature; The temporary pulse wave data is subjected to waveform conversion to obtain the frequency domain type feature.

[0125] Specifically, the volume pulse wave data of the person to be tested is collected in real time through smart wearable devices (such as smart watches or health monitoring devices). These devices usually monitor the fluctuation of blood in blood vessels through photoelectric sensors to obtain pulse waveform information.

[0126] The collected volume pulse wave data is preprocessed, mainly including: De-noising: using filtering technology to remove noise in the signal, such as power frequency interference, high-frequency clutter, etc. De-trending: eliminating trend interference in the data to ensure that the data can reflect the actual changes in the pulse waveform. Smoothing: applying smoothing technology to improve the readability of the data and make subsequent feature extraction more accurate. After preprocessing, temporary pulse wave data is obtained. Figure 3 As shown, it is a schematic diagram of temporary pulse wave data, wherein the horizontal axis represents time, which is marked as "time" and the unit is second; the vertical axis represents the light energy corresponding to the sampling point, which is marked as "power" and the unit is dimensionless.

[0127] Then analyze the temporary pulse wave data to extract time-related features (time domain features), such as: Mean: the average value of the pulse wave signal. Variance: the degree of change of the pulse wave, reflecting the stability of the waveform. Kurtosis and skewness: used to evaluate the characteristics of the pulse wave morphology, etc. These time domain features can provide direct information about the morphology and changes of the pulse wave.

[0128] At the same time, the temporary pulse wave data is converted into a waveform, usually using methods such as Fourier transform to convert the time domain signal into a frequency domain signal, and extract frequency domain features, including: wavelet energy ratio, center of gravity frequency standard deviation, etc. Through frequency domain features, we can have a deeper understanding of the frequency characteristics of the pulse wave and its relationship with physiological states such as fatigue.

[0129] Acquiring real-time data through smart wearable devices and preprocessing can reduce interference, ensure that the acquired data is more accurate, and thus improve the accuracy of subsequent analysis and model prediction. The combination of time domain and frequency domain features can fully reflect the individual's physiological state. Time domain features provide real-time change information, and frequency domain features reveal potential health problems, which helps to comprehensively evaluate fatigue and other physiological characteristics. After accurately acquiring five-dimensional time-frequency feature data, the model can provide personalized fatigue and health management suggestions for each individual, helping users to effectively predict and manage fatigue.

[0130] Alternatively, if Figure 2 , the fatigue risk level assessment system also includes an early warning module; The early warning module is used to issue early warning prompts based on the fatigue risk level of the person to be tested, and the early warning prompts include language broadcasts and information prompts.

[0131] Specifically, the early warning module sets corresponding early warning thresholds for different risk levels. When the fatigue risk level of the person to be tested reaches or exceeds the preset threshold, the system will activate the early warning prompt function.

[0132] Early warning alerts can be provided through multiple channels, such as mobile phone notifications, in-app reminders, audio alarms, etc., to ensure that users can obtain information in a timely manner.

[0133] In addition to issuing early warnings, the system can provide personalized suggestions, such as recommendations for rest, adjusting work schedules, and engaging in appropriate relaxation activities, to help individuals effectively cope with fatigue.

[0134] The early warning module can also record early warning events and participants’ responses for subsequent analysis. These data will be used to optimize the assessment model and early warning mechanism to achieve more accurate fatigue management.

[0135] Timely fatigue warning can prevent safety accidents caused by fatigue, especially in work scenarios that require high concentration (such as driving, operating machines, etc.), which can significantly reduce risks. The warning system constantly reminds individuals to pay attention to their own fatigue status, promotes their attention to health management and self-monitoring, and helps to cultivate good living and working habits.

[0136] By recording and feeding back fatigue warning data, an effective data basis can be provided for researchers and health management experts for future research on the relationship between fatigue and health, as well as for improving the effectiveness of the warning system.

[0137] Through the integration of early warning modules, the individual fatigue risk level assessment system can not only timely identify and reduce fatigue-related risks, but also provide more comprehensive health management and support. The role of this system goes far beyond simple monitoring, allowing users to actively manage their own health and safety.

[0138] In some specific embodiments, the fatigue risk level assessment system is verified, and data of 365 volunteers who participated in fatigue questionnaire annotation are selected, as shown in Table 5, wherein the data test (level assessment) shows that 2 is the best value, 1 and 5 are good values, then a value range of 1-5 can be given, and the best value is fuzzy processed.

[0139] Table 5 Experimental verification data table (excerpt)

[0140] The total dataset of this study contains 365 samples, covering data related to fatigue status of different individuals in various situations. Each sample contains characteristics related to fatigue risk, such as physiological indicators, behavioral data, and time-frequency characteristics.

[0141] In order to evaluate the generalization ability and stability of the model, we used cross-validation technology. Specifically, the commonly used K-fold cross-validation method divides the data set into K subsets. In this case, 5-fold or 10-fold cross-validation may be used, that is, the data set is evenly divided into K parts, and then the model is trained on the K-1 parts and validated on the remaining 1 part. This process is repeated K times, and finally the average accuracy of each validation is calculated.

[0142] In each fold, the model is trained and its performance is evaluated on the validation set. The accuracy of each validation is calculated by recording the consistency between the model's prediction results and the actual labels. Finally, a comprehensive evaluation index is obtained by averaging the accuracy of all folds.

[0143] After verification of all 365 data, the final average cross-validation accuracy was 80%, which means that the model can correctly predict the fatigue risk level of the sample with an 80% probability.

[0144] This indicates that the fatigue risk level assessment system has good predictive ability in fatigue risk assessment of different individuals. This provides strong evidence for the practical application of the algorithm. The clinical significance of this result is that the fatigue risk assessment model based on this algorithm can help professionals identify potential high-risk individuals in a timely manner, and then formulate corresponding intervention measures to reduce health problems and accident risks caused by fatigue.

[0145] like Figure 4 As shown, a fatigue risk level assessment device 400 provided in an embodiment of the present disclosure includes a memory 410 and a processor 420; the memory 410 is used to store a computer program; the processor 420 is used to implement the fatigue risk level assessment system as described above when executing the computer program.

[0146] In other words, a fatigue risk level assessment device 400 includes a memory 410 and a processor 420 coupled to the memory 410; the memory 410 is configured to store a computer program; and the processor 420 is configured to perform the following operations when executing the computer program: Obtaining basic data of the person to be tested, the basic data including individual basic data, sleep characteristic data at preset time intervals, five-dimensional time-frequency characteristic data and health behavior data; Extracting features from the sleep feature data and the health behavior data respectively to obtain first feature data and second feature data; Fusing the five-dimensional time-frequency feature data, the first feature data, and the second feature data to obtain fused feature data; Inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data of the person to be tested; Based on the preset risk level data, the fatigue risk level of the person to be tested is determined according to the individual basic data and the fatigue data.

[0147] An embodiment of the present disclosure provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the fatigue risk level assessment system as described above is implemented.

[0148] In other words, a non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor performs the following operations: Obtaining basic data of the person to be tested, the basic data including individual basic data, sleep characteristic data at preset time intervals, five-dimensional time-frequency characteristic data and health behavior data; Extracting features from the sleep feature data and the health behavior data respectively to obtain first feature data and second feature data; Fusing the five-dimensional time-frequency feature data, the first feature data, and the second feature data to obtain fused feature data; Inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data of the person to be tested; Based on the preset risk level data, the fatigue risk level of the person to be tested is determined according to the individual basic data and the fatigue data.

[0149] An evaluation device 400 for fatigue risk level that can serve as a server or client of the present disclosure will now be described, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The evaluation device 400 for fatigue risk level is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The evaluation device 400 for fatigue risk level can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0150] The fatigue risk level assessment device 400 includes a computing unit, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for the operation of the device can also be stored. The computing unit, ROM, and RAM are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus.

[0151] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc. In the present application, the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme of the present disclosure. In addition, each functional unit in each embodiment of the present disclosure can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0152] Although the disclosure is disclosed as above, the protection scope of the disclosure is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the disclosure, and these changes and modifications will fall within the protection scope of the disclosure.

Claims

1. A fatigue risk level assessment system, characterized in that: include: The acquisition module is used to obtain basic data of the person to be tested, wherein the basic data includes individual basic data, sleep characteristic data at preset time intervals, five-dimensional time-frequency characteristic data and health behavior data; A processing module, configured to extract features from the sleep feature data and the health behavior data respectively to obtain first feature data and second feature data; and fuse the five-dimensional time-frequency feature data, the first feature data and the second feature data to obtain fused feature data; A prediction module, used for inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data of the person to be tested; The preset fatigue prediction model is constructed based on a fixed effect regression model; An evaluation module is used to determine the fatigue risk level of the person to be tested based on preset risk level data, the individual basic data and the fatigue data.

2. The fatigue risk level assessment system according to claim 1, characterized in that: The individual basic data includes occupation information and basic information; the evaluation module is specifically used for: Filtering the preset risk level data by the occupation information to obtain a corresponding risk level data set; The risk level data set is filtered through the basic information to obtain temporary risk level data, and the temporary risk level data is screened based on the fatigue data to obtain the fatigue risk level of the person to be tested.

3. The fatigue risk level assessment system according to claim 1, characterized in that: The processing module also includes an analysis unit; The analysis unit is used to perform correlation analysis on the first feature data and the second feature data respectively to obtain corresponding temporary feature data.

4. The fatigue risk level assessment system according to claim 3, characterized in that: The processing module also includes a fusion unit; The fusion unit is used to fuse all the temporary feature data with the five-dimensional time-frequency feature data to obtain the fused feature data.

5. The fatigue risk level assessment system according to claim 4, characterized in that: The five-dimensional time-frequency feature data includes time domain type features and frequency domain type features; The step of fusing all the temporary feature data with the five-dimensional time-frequency feature data to obtain the fused feature data includes: Evaluate the time domain type features, the frequency domain type features and each of the temporary feature data respectively to obtain a feature weight matrix; Based on the feature weight matrix, all the temporary feature data are fused with the five-dimensional time-frequency feature data to obtain the fused feature data.

6. The fatigue risk level assessment system according to claim 1, characterized in that: The preset fatigue prediction model includes a fatigue non-sleep module and a fatigue sleep module.

7. The fatigue risk level assessment system according to claim 1, characterized in that: The five-dimensional time-frequency feature data includes time domain type features and frequency domain type features; The step of obtaining five-dimensional time-frequency characteristic data of the person to be tested includes: Collecting volume pulse wave data of the person to be tested through a smart wearable device, and preprocessing the volume pulse wave data to obtain temporary pulse wave data; Extracting the temporary pulse wave data to obtain the time domain type feature; The temporary pulse wave data is subjected to waveform conversion to obtain the frequency domain type feature.

8. The fatigue risk level assessment system according to claim 1, characterized in that: The fatigue risk level assessment system also includes an early warning module; The early warning module is used to issue early warning prompts based on the fatigue risk level of the person to be tested, and the early warning prompts include language broadcasts and information prompts.

9. A fatigue risk level assessment device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is configured to implement the following steps when executing the computer program: Obtaining basic data of the person to be tested, the basic data including individual basic data, sleep characteristic data at preset time intervals, five-dimensional time-frequency characteristic data and health behavior data; Extracting features from the sleep feature data and the health behavior data respectively to obtain first feature data and second feature data; Fusing the five-dimensional time-frequency feature data, the first feature data, and the second feature data to obtain fused feature data; Inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data of the person to be tested; Based on the preset risk level data, the fatigue risk level of the person to be tested is determined according to the individual basic data and the fatigue data.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program. When the computer program is executed by the processor, the following steps are implemented: Obtaining basic data of the person to be tested, the basic data including individual basic data, sleep characteristic data at preset time intervals, five-dimensional time-frequency characteristic data and health behavior data; Extracting features from the sleep feature data and the health behavior data respectively to obtain first feature data and second feature data; Fusing the five-dimensional time-frequency feature data, the first feature data, and the second feature data to obtain fused feature data; Inputting the fused feature data into a preset fatigue prediction model to obtain fatigue data of the person to be tested; Based on the preset risk level data, the fatigue risk level of the person to be tested is determined according to the individual basic data and the fatigue data.

Images