Fatigue intervention method, system, equipment, medium and product

Abstract

The invention discloses a fatigue intervention method, system and device, a medium and a product, and relates to the field of electrocardiosignal processing. Preprocessing the electrocardiosignal by adopting wavelet transform; performing feature extraction on the preprocessed electrocardiosignals to obtain an initial feature set; screening the initial feature set based on an SHAP algorithm to obtain a key feature set; based on the key feature set, a fatigue recognition model is adopted to determine the grade of a fatigue state; the fatigue recognition model is obtained by training a random forest model through a sample key feature set; the sample key feature set is obtained by processing sample electrocardiosignals; and based on the level of the fatigue state, performing fatigue intervention by using different types of music and time-frequency interference signals with different frequencies. According to the invention, the level of the fatigue state can be identified, and personalized intervention is implemented according to the level of the fatigue state.

Description

Technical Field

[0001] The present application relates to the field of electrocardiogram signal processing, and in particular to a fatigue intervention method, system, device, medium and product. Background Art

[0002] In the modern society with rapid development of science and technology and economy, people are facing a faster pace of life and greater competitive pressure. This has led to the prevalence of long-term and high-intensity mental work, which in turn causes brain fatigue. Brain fatigue not only makes people feel tired and has difficulty concentrating, but long-term accumulation may also induce a series of physical and psychological problems, such as heart disease, insomnia, anxiety and depression, which seriously threaten people's physical and mental health. Summary of the invention

[0003] The purpose of this application is to provide a fatigue intervention method, system, device, medium and product that can identify the level of fatigue status and implement personalized intervention based on the level of fatigue status.

[0004] To achieve the above objectives, this application provides the following solutions.

[0005] In a first aspect, the present application provides a fatigue intervention method, comprising: Collect human ECG signals; Preprocessing the electrocardiogram signal by wavelet transform; Extracting features from the preprocessed ECG signal to obtain an initial feature set; the initial feature set includes heart rate features and heart rate variability features; The initial feature set is screened based on the SHAP algorithm to obtain a key feature set; Based on the key feature set, a fatigue recognition model is used to determine the level of fatigue state; the fatigue recognition model is obtained by training a random forest model with a sample key feature set; the sample key feature set is obtained by processing a sample electrocardiogram signal; Based on the level of the fatigue state, different types of music and time-frequency interference signals of different frequencies are used to perform fatigue intervention.

[0006] In a second aspect, the present application provides a fatigue intervention system, comprising: ECG signal acquisition module, used to collect human ECG signals; A preprocessing module, used for preprocessing the electrocardiogram signal by wavelet transform; A feature extraction module, used to extract features from the preprocessed ECG signal to obtain an initial feature set; the initial feature set includes heart rate features and heart rate variability features; A feature screening module, used for screening the initial feature set based on the SHAP algorithm to obtain a key feature set; A fatigue state level determination module is used to determine the level of fatigue state based on the key feature set using a fatigue recognition model; the fatigue recognition model is obtained by training a random forest model with a sample key feature set; the sample key feature set is obtained by processing a sample electrocardiogram signal; The fatigue intervention module is used to perform fatigue intervention based on the level of the fatigue state by using different types of music and time-frequency interference signals of different frequencies.

[0007] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned fatigue intervention method.

[0008] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned fatigue intervention method when executed by a processor.

[0009] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements the above-mentioned fatigue intervention method when executed by a processor.

[0010] According to the specific embodiments provided in this application, this application has the following technical effects.

[0011] (1) This application uses the SHAP algorithm to screen heart rate characteristics and heart rate variability characteristics, which can improve the accuracy of fatigue state level identification and the effectiveness of intervention.

[0012] (2) This application uses different types of music and time-frequency interference signals of different frequencies to intervene in fatigue states of different levels. It can penetrate deep into the brain and provide non-invasive stimulation to the human body, effectively avoiding the pain problems that may be caused by traditional direct current stimulation. At the same time, it improves the comfort and tolerance of the intervention, thereby minimizing the health and safety risks caused by fatigue and protecting people's physical and mental health. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0014] Figure 1 A flowchart of a fatigue intervention method provided in one embodiment of the present application.

[0015] Figure 2A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0017] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0018] In an exemplary embodiment, Figure 1 As shown, a fatigue intervention method is provided, which is executed by a computer device, and can be specifically executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In an embodiment of the present application, the method is applied to a server as an example for explanation, including the following steps S1 to S6.

[0019] S1: Collect human ECG signals.

[0020] ECG signals are electrical signals of human heart activity, which are captured by measuring the action potential of heart cells. Changes in the action potential of heart cells can reflect the physiological state of the heart, and these changes are closely related to the individual's fatigue state.

[0021] S2: Preprocessing the ECG signal using wavelet transform.

[0022] Wavelet transform is used to preprocess the collected ECG signals to ensure the accuracy and reliability of the ECG signals. By decomposing the ECG signals at multiple scales, useful signal components and noise can be effectively separated, and power frequency interference, high-frequency noise and baseline drift, which are caused by power frequency, electromagnetic interference, breathing and skin impedance, can be removed. After wavelet transform, useful ECG signal coefficients are retained and the signal is reconstructed, thereby significantly improving the quality of the ECG signal. The characteristics of the ECG signal at different times and scales are analyzed by formula (1).

[0023] (1) Among them, X(t) is the original time domain ECG signal. is the time shift parameter, which is used to adjust the position of the wavelet function on the time axis. is a scale parameter, which is used to adjust the width of the wavelet function to analyze the signal at different scales. represents the continuous wavelet transform coefficients, Denotes X(t) with respect to the wavelet function The continuous wavelet transform results are essentially different representations of the same transform result, which are used to characterize the ECG signal in different and The following features.

[0024] The ECG signal is decomposed at multiple scales through wavelet transform, the coefficients of useful signals are retained, and finally the signal is reconstructed through the extracted wavelet coefficients, thereby reducing noise interference and facilitating subsequent QRS complex detection.

[0025] S3: Extracting features from the preprocessed ECG signal to obtain an initial feature set; the initial feature set includes heart rate features and heart rate variability features. The heart rate variability features include: time domain features, frequency domain features and nonlinear features.

[0026] Step S3 specifically includes: performing R wave positioning detection based on the preprocessed ECG signal; and extracting heart rate characteristics and heart rate variability characteristics based on the R wave.

[0027] Heart rate index and heart rate variability index can be used to identify fatigue status. By observing the waveform of the ECG signal, it can be found that the R wave in the QRS complex has a large amplitude, a narrow pulse width, obvious characteristics and is easy to identify. The positioning of the R wave is the most important step in ECG signal recognition. This application uses a method for real-time detection of QRS complexes proposed by Pan and Willis J. Jiapu Tompkins to locate and detect the R wave: the denoised ECG signal is differentially processed to obtain slope information, and then the ECG signal that has been differentially processed is processed using a square operation. In order to remove the negative waveform that may be caused by the square operation to form a double peak that affects the detection and positioning of the R wave, the waveform can be smoothed by window sliding integration. After the above processing, the ECG signal uses an adaptive threshold method to locate and detect the R wave. The formulas involved are as follows: (2) (3) (4) Formula (2) represents the differential processing of ECG signal. is the original ECG signal of the nth signal sequence, is the signal after differentiation. and represents the value of the signal at different time points, and represents the previous differential value. This differential operation helps to extract the slope information of the signal. n represents the sequence number in the signal sequence, which is used to identify the signal values ​​at different times and other related calculations. The slope information of the ECG signal can be obtained by performing differential operations on the signal values ​​corresponding to different n values.

[0028] Formula (3) represents the differential signal Performs a square operation. is the squared signal, and T is the sampling period. The square operation can enhance the positive waveform of the signal, making the R wave more prominent.

[0029] Formula (4) represents the window sliding integration of the signal to achieve smoothing. Indicates the time period, is the smoothed signal, is the window size, is the original ECG signal at time This integration operation helps to reduce the negative waveform that may be generated after the squaring operation, thereby improving the accuracy of R wave detection.

[0030] Based on the positioning of the R wave, the extraction of heart rate indicators and heart rate variability (HRV) indicators is a systematic process used to evaluate the autonomic nervous regulation function of the heart. It includes time domain features, frequency domain features, and nonlinear features. The time domain features focus on the direct measurement of heart beat intervals. These indicators can reflect the activity of the autonomic nervous system and the adaptability of the heart. The frequency domain features evaluate the balance of the sympathetic and parasympathetic nervous systems by analyzing the frequency components of the heart rate. These indicators help identify changes in the autonomic nervous system under fatigue conditions. Nonlinear features provide a deeper understanding of the complexity of heart rate and can reveal the complexity and regularity of heart rate variability, thereby more accurately evaluating fatigue status.

[0031] The specific steps are as follows: First, the time interval between two consecutive R waves, the RR interval, is calculated, and then these intervals are used to calculate the heart rate, usually obtained by the formula heart rate (bpm) = 60 / average RR interval (seconds).

[0032] Next, extract heart rate indicators, such as average heart rate, maximum heart rate, minimum heart rate, and standard deviation of heart rate. The extraction of HRV indicators includes time domain indicators, such as SDNN, RMSSD, and NN50; frequency domain indicators, analyze the power of different frequency bands (such as very low frequency VLF, low frequency LF, and high frequency HF) through Fourier transform, calculate the total power, LF / HF ratio, etc.; and nonlinear indicators, such as Poincare diagram analysis and sample entropy (SampEn).

[0033] The meanings of time domain features and frequency domain features are shown in Table 1 and Table 2 respectively.

[0034] Table 1

[0035] Table 2

[0036] Due to the inconsistency of dimensions and units of each feature index, in order to avoid the impact of related factors on the calculation speed of the model, the feature index data must be normalized and mapped to [-1,1] to obtain the initial feature set.

[0037] S4: Screening the initial feature set based on the SHAP algorithm to obtain a key feature set, which specifically includes: calculating the SHAP value of each initial feature in the initial feature set based on the SHAP algorithm; sorting the initial features in descending order according to the average absolute value of the SHAP value, and selecting the first N initial features to construct a key feature set.

[0038] In order to remove redundant features, avoid multicollinearity problems and improve the overall quality of the data, feature selection will be performed in this step. The SHAP (SHapley Additive exPlanations) algorithm evaluates the importance of features by calculating the contribution of each feature to model prediction. First, calculate the SHAP value of each initial feature in the initial feature set obtained in step S3. The SHAP value can be used to analyze the average impact of each initial feature on model prediction. Generally, a SHAP value with a larger absolute value means that the feature has a greater impact on model prediction. The initial features are sorted according to the average absolute value of the SHAP value, and the top N initial features with the greatest impact on model prediction are selected to construct the key feature set.

[0039] S5: Based on the key feature set, a fatigue recognition model is used to determine the level of fatigue status; the fatigue recognition model is obtained by training a random forest model with a sample key feature set; the sample key feature set is obtained by processing a sample electrocardiogram signal.

[0040] The random forest model improves the accuracy and stability of the overall model by constructing multiple decision trees and combining their prediction results. The advantage of the random forest model in processing high-dimensional data is that it does not require feature selection because each tree is trained on a randomly selected sample key feature set. The random forest model is designed with four output nodes, corresponding to a mild fatigue state, a moderate fatigue state, a severe fatigue state, and a non-fatigue state. During the training and prediction process of the random forest model, the sample key feature set is input into the random forest model. These sample key feature sets combine heart rate indicators and heart rate variability (HRV) indicators, providing a strong physiological basis for assessing the fatigue state of an individual. With its powerful pattern recognition ability, the random forest model can accurately classify the sample key feature set into the above four fatigue states, thereby achieving the classification of fatigue state levels.

[0041] During the training process of the random forest model, the key feature set of samples is continuously iteratively improved to find the key feature set that optimizes the model performance.

[0042] S6: Based on the level of the fatigue state, different types of music and time-frequency interference signals of different frequencies are used to perform fatigue intervention.

[0043] When the level of the fatigue state is a non-fatigue state, no fatigue intervention is performed.

[0044] When the fatigue level is mild, fatigue intervention is performed by playing soothing music and weak high-frequency (frequency is 1000Hz) time-frequency interference signals to keep the mind clear and concentrate on completing the work.

[0045] When the fatigue state is at a moderate level, fatigue intervention is performed by playing rock music with a stronger sense of rhythm and applying a stronger high-frequency (frequency is 2000 Hz) time-frequency interference signal.

[0046] When the fatigue state is severe, fatigue intervention is performed by applying a time-frequency interference signal with a frequency of 3000 Hz. Increasing the intensity of the time-frequency interference signal can regain the necessary energy and attention for work.

[0047] This application uses time domain interferometry (TI) technology for fatigue intervention. This technology is a non-invasive deep brain stimulation method that is achieved by placing two electrodes on the scalp and applying two pairs of high-frequency stimulation signals with similar but slightly different frequencies. The interference of these signals in specific areas inside the brain forms a strong overlapping electric field and produces a low-frequency envelope wave. This low-frequency envelope wave can penetrate deep into the brain and regulate brain activity. Through precise electric field control, brain activity can be deeply regulated and fatigue symptoms can be improved.

[0048] In the fatigue intervention method of the present application, the parameters of the time-frequency interferometry (TI) signal are precisely adjusted according to the severity of the individual's fatigue state to achieve personalized intervention. This includes adjusting the intensity of the stimulation by changing the intensity of the current delivered by the electrode to achieve effective intervention in severe fatigue states; changing the frequency of the low-frequency envelope wave by adjusting the frequency difference between the two pairs of high-frequency stimulation signals, thereby affecting the synchronization of the neural network and achieving different neural regulation effects; in addition, combining different styles of music to further adjust the intensity of neural stimulation to meet the needs of different fatigue levels. This multi-dimensional adjustment strategy ensures the flexibility and effectiveness of the intervention measures, aiming to optimize the intervention effect and maintain and promote the physical and mental health of individuals.

[0049] The fatigue intervention method provided in this application has the following advantages.

[0050] 1. The electric field stimulation generated by the external electric field in this application can more accurately locate specific areas of the brain. Compared with traditional current stimulation, its positioning is more accurate and can act more deeply on the brain. This precise stimulation method can more effectively regulate brain activity and provide more significant fatigue intervention effects, while reducing unnecessary stimulation to non-target areas and reducing potential adverse reactions.

[0051] 2. This application introduces feature selection technology to optimize the performance of the fatigue recognition model. By using SHAP values ​​to quantify the contribution of each feature to the model prediction, the most critical features for fatigue status assessment can be identified and selected. This feature selection step not only improves the accuracy and efficiency of the model, but also enhances the model's interpretability, making fatigue intervention more accurate and personalized.

[0052] Based on the same inventive concept, the embodiment of the present application also provides a system for implementing the fatigue intervention method involved above. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more fatigue intervention system embodiments provided below can refer to the limitations of the fatigue intervention method above, and will not be repeated here.

[0053] In an exemplary embodiment, a fatigue intervention system is provided, comprising the following modules.

[0054] The ECG signal acquisition module is used to acquire human ECG signals.

[0055] The preprocessing module is used to preprocess the electrocardiogram signal by using wavelet transform.

[0056] The feature extraction module is used to extract features from the preprocessed ECG signal to obtain an initial feature set; the initial feature set includes heart rate features and heart rate variability features.

[0057] The feature screening module is used to screen the initial feature set based on the SHAP algorithm to obtain a key feature set.

[0058] The fatigue state level determination module is used to determine the level of the fatigue state based on the key feature set and a fatigue recognition model; the fatigue recognition model is obtained by training a random forest model with a sample key feature set; the sample key feature set is obtained by processing a sample electrocardiogram signal.

[0059] The fatigue intervention module is used to perform fatigue intervention based on the level of the fatigue state by using different types of music and time-frequency interference signals of different frequencies.

[0060] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program. The computer device may be a server or a terminal, and its internal structure diagram may be as shown in FIG. Figure 2 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data to be processed. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a fatigue intervention method is implemented.

[0061] Those skilled in the art will understand that Figure 2 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0062] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0063] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0064] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0065] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0066] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0067] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0068] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A fatigue intervention method, characterized in that: include: Collect human ECG signals; Preprocessing the electrocardiogram signal by wavelet transform; Extract features from the preprocessed ECG signal to obtain an initial feature set; The initial feature set includes heart rate features and heart rate variability features; The initial feature set is screened based on the SHAP algorithm to obtain a key feature set; Based on the key feature set, a fatigue recognition model is used to determine the level of fatigue state; The fatigue recognition model is obtained by training a random forest model with a sample key feature set; the sample key feature set is obtained by processing a sample electrocardiogram signal; Based on the level of the fatigue state, different types of music and time-frequency interference signals of different frequencies are used to perform fatigue intervention.

2. The fatigue intervention method according to claim 1, characterized in that: The preprocessed ECG signal is subjected to feature extraction to obtain an initial feature set, which specifically includes: Perform R wave location detection based on preprocessed ECG signals; Heart rate features and heart rate variability features are extracted based on the R wave.

3. The fatigue intervention method according to claim 2, characterized in that: The heart rate variability characteristics include: time domain characteristics, frequency domain characteristics and nonlinear characteristics.

4. The fatigue intervention method according to claim 1, characterized in that: The initial feature set is screened based on the SHAP algorithm to obtain a key feature set, which specifically includes: Calculate the SHAP value of each initial feature in the initial feature set based on the SHAP algorithm; The initial features are sorted in descending order according to the average absolute value of the SHAP value, and the first N initial features are selected to construct a key feature set.

5. The fatigue intervention method according to claim 1, characterized in that: The levels of fatigue state include: non-fatigue state, mild fatigue state, moderate fatigue state and severe fatigue state.

6. The fatigue intervention method according to claim 5, characterized in that: Based on the level of the fatigue state, different types of music and time-frequency interference signals of different frequencies are used for fatigue intervention, specifically including: When the level of fatigue status is non-fatigue status, no fatigue intervention is performed; When the fatigue state is at a mild level, fatigue intervention is performed by playing soothing music and applying a time-frequency interference signal with a frequency of 1000 Hz; When the fatigue state is moderate, fatigue intervention is performed by playing rock music and applying a time-frequency interference signal with a frequency of 2000 Hz; When the level of the fatigue state is a severe fatigue state, fatigue intervention is performed by applying a time-frequency interference signal with a frequency of 3000 Hz.

7. A fatigue intervention system, characterized in that: include: ECG signal acquisition module, used to collect human ECG signals; A preprocessing module, used for preprocessing the electrocardiogram signal by wavelet transform; A feature extraction module, used to extract features from the preprocessed ECG signal to obtain an initial feature set; the initial feature set includes heart rate features and heart rate variability features; A feature screening module, used for screening the initial feature set based on the SHAP algorithm to obtain a key feature set; A fatigue state level determination module is used to determine the level of fatigue state based on the key feature set using a fatigue recognition model; the fatigue recognition model is obtained by training a random forest model with a sample key feature set; the sample key feature set is obtained by processing a sample electrocardiogram signal; The fatigue intervention module is used to perform fatigue intervention based on the level of the fatigue state by using different types of music and time-frequency interference signals of different frequencies.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the fatigue intervention method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the fatigue intervention method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the fatigue intervention method according to any one of claims 1 to 6 is implemented.

Images