Muscle fatigue assessment method based on entropy and application thereof

Abstract

The invention provides an entropy-based muscle fatigue assessment method and application thereof, and particularly relates to the technical field of signal processing, and the method comprises the steps: firstly collecting a surface electromyogram (sEMG) signal, and carrying out the preprocessing; then, fractional calculus is combined with a fluctuation dispersion entropy (FDE) algorithm, and a multi-scale processing process is introduced, so that an RCMFFDE algorithm is innovatively proposed. Performing multi-scale analysis on the preprocessed sEMG signal by using the algorithm to generate a plurality of entropy curves; the entropy curve meeting the preset condition is selected as the final entropy curve, and the final entropy curve can accurately reflect the muscle fatigue state. According to the method, the muscle fatigue degree can be effectively and reliably evaluated, and the method has important kinematics and medical significance for preventing human body muscle diseases and sports injuries.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and in particular to an entropy-based muscle fatigue assessment method and application thereof. Background Art

[0002] In recent years, with the rapid development of sports science, rehabilitation medicine and artificial intelligence technology, muscle fatigue monitoring technology has become a hot area of ​​concern. In fields such as sports training and industrial production lines, accurate assessment of muscle fatigue is of great significance for improving training effects and optimizing work processes. However, in practical applications, muscle fatigue monitoring technology faces many challenges. On the one hand, muscle fatigue is a complex and dynamic physiological process involving the interaction of multiple physiological parameters, which makes accurate monitoring particularly difficult. On the other hand, there are significant differences in muscle fatigue characteristics between different individuals, which makes the development of universal monitoring technology more complicated. In addition, most of the existing monitoring methods rely on expensive equipment and complex operating procedures, which limits their wide application in actual scenarios. Therefore, how to develop an accurate, reliable, portable and cost-effective muscle fatigue monitoring technology has become an urgent need for the current development of the industry.

[0003] In order to solve the related problems in muscle fatigue monitoring, the traditional existing technology mainly uses a variety of signal analysis methods to extract feature information from sEMG signals. Among them, time domain analysis, frequency domain analysis and time-frequency domain analysis are three commonly used methods. Time domain analysis mainly evaluates the degree of muscle fatigue by calculating statistical parameters such as the mean, variance, and root mean square value of the signal; frequency domain analysis converts the signal into a frequency domain representation through methods such as Fourier transform, and then analyzes the spectral characteristics of the signal; time-frequency domain analysis combines the advantages of time domain and frequency domain, and can more comprehensively reflect the dynamic changes of the signal. In addition, with the development of nonlinear dynamics methods, entropy algorithms have gradually been applied to the analysis of sEMG signals. Such as sample entropy, fuzzy entropy, dispersed entropy, etc., these algorithms evaluate the fatigue state of muscles by quantifying the uncertainty of the signal. In particular, methods such as fractional-order fuzzy dispersed entropy and multi-scale entropy proposed in recent years have further improved the accuracy and robustness of feature extraction. These traditional existing technologies have provided strong support for muscle fatigue monitoring through continuous optimization and innovation. However, in practical applications, the above methods still have the problem of insufficient sensitivity and robustness of the algorithm for evaluating muscle fatigue using surface electromyography (sEMG) signals.

[0004] In summary, how to effectively and reliably assess the degree of muscle fatigue is a difficult problem that needs to be solved urgently in this technical field. Summary of the invention

[0005] The main purpose of the present invention is to provide an entropy-based muscle fatigue assessment method and its application to solve the technical problem that the prior art cannot effectively and reliably assess the degree of muscle fatigue, which has important kinematic and medical significance for preventing human muscle diseases and sports injuries.

[0006] In order to achieve the above-mentioned purpose, the present invention provides an entropy-based muscle fatigue assessment method and its application, which has important kinematic and medical significance for preventing human muscle diseases and sports injuries.

[0007] In a first aspect, the present invention provides an entropy-based muscle fatigue assessment method, the method comprising:

[0008] Collecting sEMG signals and preprocessing the sEMG signals;

[0009] Based on the FDE algorithm combined with fractional calculus and the introduction of multi-scale processing, the RCMFFDE algorithm is obtained;

[0010] Performing multi-scale analysis on the preprocessed sEMG signal using the RCMFFDE algorithm to obtain multiple entropy curves;

[0011] Selecting an entropy value curve that meets a preset condition from among the multiple entropy value curves as a final entropy value curve;

[0012] The degree of muscle fatigue was evaluated based on the final entropy curve.

[0013] Optionally, the collecting of sEMG signals and preprocessing of the sEMG signals include:

[0014] The sEMG signal is collected and denoised by a signal denoising method based on VMD.

[0015] Optionally, the FDE algorithm is combined with fractional calculus and a multi-scale processing process is introduced to obtain a RCMFFDE algorithm, including:

[0016] Based on the FDE algorithm and combined with fractional calculus, the FFDE algorithm is obtained;

[0017] A multi-scale processing process is introduced on the basis of the FFDE algorithm to obtain the RCMFFDE algorithm.

[0018] Optionally, the multi-scale processing process includes:

[0019] The sEMG signal is subjected to coarse-graining processing by using the median of the sEMG signal through different scale factors, wherein the coarse-graining processing is used to evaluate the entropy characteristics of the sEMG signal at various scales.

[0020] Optionally, the RCMFFDE algorithm is used to perform multi-scale analysis on the pre-processed sEMG signal to obtain multiple entropy curves, including:

[0021] Performing coarse-graining processing on the sEMG signal to obtain a plurality of coarse-grained time series, wherein the plurality of coarse-grained time series are used to indicate the sEMG signal after coarse-graining, and the coarse-graining processing is used to indicate the multi-scale processing process;

[0022] According to the plurality of coarse-grained time series, a plurality of entropy value curves are obtained through the FFDE algorithm, wherein the RCMFFDE algorithm is used to indicate an algorithm for introducing a multi-scale processing process on the basis of the FFDE algorithm.

[0023] Optionally, selecting an entropy value curve that meets a preset condition from among the multiple entropy value curves as a final entropy value curve includes:

[0024] Performing curve fitting on each of the entropy value curves to obtain the absolute value of the slope and the coefficient of determination of the fitting curve corresponding to each of the entropy value curves;

[0025] Multiply the absolute value of the slope and the coefficient of determination to obtain a curve evaluation index value;

[0026] The entropy curve corresponding to the maximum curve evaluation index value is selected as the final entropy curve.

[0027] Optionally, the muscle fatigue degree is evaluated according to the final entropy value curve, including:

[0028] When the entropy value in the final entropy value curve gradually decreases, the degree of muscle fatigue gradually increases.

[0029] In a second aspect, the present invention provides an application of the entropy-based muscle fatigue assessment method as described in the first aspect in assessing muscle fatigue status.

[0030] The entropy-based muscle fatigue assessment method and its application provided by the present application realize accurate assessment of muscle fatigue degree by collecting surface electromyography (sEMG) signals. First, the system collects sEMG signals and preprocesses them to eliminate noise and interference and improve signal quality. Subsequently, the present invention proposes a new method combining fractional calculus with fluctuation dispersion entropy (FDE) algorithm and introducing multi-scale processing process, namely RCMFFDE algorithm, which is a groundbreaking method to conduct in-depth analysis of pre-processed sEMG signals through multi-scale processing process to generate multiple entropy value curves. Among multiple entropy value curves, the present invention selects the entropy value curve that best reflects the muscle fatigue state as the final entropy value curve through preset conditions. The final entropy value curve can intuitively and accurately show the entropy value changes of muscles under different fatigue levels, and provides a reliable basis for muscle fatigue assessment. The method can effectively and reliably assess the degree of muscle fatigue, and has important kinematic and medical significance for preventing human muscle diseases and sports injuries. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0032] Figure 1 Flowchart of the entropy-based muscle fatigue assessment method provided for this application.

[0033] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0035] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in sequences other than those illustrated or described herein.

[0036] In the present invention, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0037] In recent years, with the rapid development of sports science, rehabilitation medicine and artificial intelligence technology, muscle fatigue monitoring technology has become a hot area of ​​concern. In fields such as sports training and industrial production lines, accurate assessment of muscle fatigue is of great significance for improving training effects and optimizing work processes. However, in practical applications, muscle fatigue monitoring technology faces many challenges. On the one hand, muscle fatigue is a complex and dynamic physiological process involving the interaction of multiple physiological parameters, which makes accurate monitoring particularly difficult. On the other hand, there are significant differences in muscle fatigue characteristics between different individuals, which makes the development of universal monitoring technology more complicated. In addition, most of the existing monitoring methods rely on expensive equipment and complex operating procedures, which limits their wide application in actual scenarios. Therefore, how to develop an accurate, reliable, portable and cost-effective muscle fatigue monitoring technology has become an urgent need for the current development of the industry.

[0038] In order to solve the related problems in muscle fatigue monitoring, the traditional existing technology mainly uses a variety of signal analysis methods to extract feature information from sEMG signals. Among them, time domain analysis, frequency domain analysis and time-frequency domain analysis are three commonly used methods. Time domain analysis mainly evaluates the degree of muscle fatigue by calculating statistical parameters such as the mean, variance, and root mean square value of the signal; frequency domain analysis converts the signal into a frequency domain representation through methods such as Fourier transform, and then analyzes the spectral characteristics of the signal; time-frequency domain analysis combines the advantages of time domain and frequency domain, and can more comprehensively reflect the dynamic changes of the signal. In addition, with the development of nonlinear dynamics methods, entropy algorithms have gradually been applied to the analysis of sEMG signals. Such as sample entropy, fuzzy entropy, dispersed entropy, etc., these algorithms evaluate the fatigue state of muscles by quantifying the uncertainty of the signal. In particular, methods such as fractional-order fuzzy dispersed entropy and multi-scale entropy proposed in recent years have further improved the accuracy and robustness of feature extraction. These traditional existing technologies have provided strong support for muscle fatigue monitoring through continuous optimization and innovation. However, in practical applications, the above methods still have the problem of insufficient sensitivity and robustness of the algorithm for evaluating muscle fatigue using surface electromyography (sEMG) signals.

[0039] In summary, how to effectively and reliably assess the degree of muscle fatigue is a difficult problem that needs to be solved urgently in this technical field.

[0040] The entropy-based muscle fatigue assessment method and its application provided in this application collects and preprocesses sEMG signals, innovatively integrates the FDE algorithm and fractional calculus, and introduces multi-scale processing to form the RCMFFDE algorithm. The algorithm analyzes sEMG signals at multiple scales and screens the entropy value curves that meet the conditions, thereby achieving accurate assessment of muscle fatigue.

[0041] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0042] Figure 1 The flowchart of the entropy-based muscle fatigue assessment method provided in this application is intended to explain the entropy-based muscle fatigue assessment method in detail, such as Figure 1 As shown, a muscle fatigue assessment method based on entropy is provided in this embodiment, and the method includes:

[0043] S101: collecting sEMG signals and preprocessing the sEMG signals.

[0044] The collecting of sEMG signals and preprocessing of the sEMG signals may include the following steps:

[0045] Step 1: Prepare the acquisition equipment:

[0046] Use a sEMG acquisition device containing multiple electrodes, such as the MyoWare sensor or similar commercially available devices.

[0047] The electrode array should be placed over the target muscle group, such as the biceps in the arm or the thigh muscles in the leg.

[0048] Use a conductive gel or patch to ensure good contact between the electrodes and the skin.

[0049] Step 2: Signal acquisition:

[0050] Connect the sEMG acquisition device to a data acquisition system, such as an electromyograph or a dedicated data acquisition card.

[0051] Set the sampling frequency between 1000Hz and 5000Hz to capture enough signal details.

[0052] Make sure that the subject is in a relaxed state, and then perform specific muscle activities (such as flexion or extension) to record sEMG signals. In the specific implementation, this embodiment selected 10 subjects aged between 23 and 26 years old, who were in good health and had no neuromusculoskeletal diseases. The subjects avoided participating in any high-intensity physical activities within 24 hours before the experiment. sEMG signals were collected using RTLab software and RunE series wireless devices with a sampling rate of 1000Hz. The experiment used 3kg dumbbells as the load, and 75% medical alcohol was used to wipe the skin surface of the brachioradialis muscle of the subject's upper limb to remove grease and dead skin on the skin surface. The entire experimental process was carried out on a windows11 system laptop, and program programming and data analysis were performed on MATLAB R2022b.

[0053] During the experiment, the subject sat in front of the experimental table with his back upright, and the electromyographic sensor was precisely placed in the developed area of ​​the brachioradialis muscle of the arm. The subject's upper arm was vertical to the ground, and the forearm and upper arm were at a 90° angle. Before the experiment began, the subject relaxed the upper arm muscles and then exerted force through the brachioradialis muscle. During the experiment, the subject lifted a 3kg dumbbell with one hand to carry the weight until the muscle felt sore and weak, and then stopped signal collection.

[0054] Step 3: Filtering:

[0055] The acquired sEMG signals were filtered using a bandpass filter to remove low-frequency noise (such as baseline drift) and high-frequency noise (such as electromagnetic interference).

[0056] The frequency range of the filter was set to 10 Hz to 500 Hz, which is the main frequency component range of the sEMG signal.

[0057] Step 4: Denoising:

[0058] Independent component analysis (ICA) can be applied, or a signal denoising method based on VMD or an algorithm such as principal component analysis (PCA) can be used to separate and remove the noise component from the signal. The specific method to be used is selected according to the actual situation and is not limited here.

[0059] Adaptive filters are used to adjust the filter parameters according to the dynamic characteristics of the signal to further reduce noise.

[0060] With respect to the denoising process, the present invention preferably uses a signal denoising method based on VMD, wherein the sEMG signal is collected and preprocessed, including:

[0061] Collect sEMG signals and perform denoising on the sEMG signals using a VMD-based signal denoising method. The following methods may be used for specific implementation:

[0062] The present invention proposes a new signal denoising method based on VMD, which can effectively separate useful sEMG signals from noise and provide reliable signal data for subsequent feature extraction and analysis.

[0063] First, the sEMG signal is decomposed into a set of band-limited modes in different frequency bands, called variational mode functions (VMFs). The key parameter of VMD is the number of decomposition modes K. In order to balance the computational complexity and denoising accuracy, K is set to 10. According to the characteristics of VMD, baseline drift noise is often limited in the first VMF. Therefore, the first VMF of the low-frequency part of the signal is removed to eliminate the baseline drift noise. For Gaussian white noise, the present invention uses an adaptive soft threshold to denoise the remaining VMF. The setting of soft noise is as follows:

[0064]

[0065] Where sgn(y) is the soft thresholding function, y is the amplitude of the signal, and T is a local threshold calculated based on the signal segment in each sliding window. The purpose is to provide an adaptive threshold for the denoising algorithm. Therefore, T can be selected as follows:

[0066] T = a*std(|z i (t)|) (2)

[0067] In the formula, z i (t) is the sEMG signal at time t, a is an empirical parameter used to determine the strength of denoising; std is the standard deviation.

[0068] Finally, the 9 modes denoised by soft threshold are reconstructed to achieve denoising.

[0069] Step 5: Standardization:

[0070] The denoised sEMG signals were normalized to eliminate the differences in signals between different subjects or between the same subject at different times.

[0071] Normalization methods include converting the signal to a standardized signal with a mean of 0 and a standard deviation of 1, or normalizing the signal based on its peak value or root mean square value.

[0072] Step 6: Signal verification:

[0073] The preprocessing process is verified using known standard signals or simulated signals to ensure the accuracy and reliability of the preprocessing method.

[0074] Compare the signal quality before and after preprocessing to evaluate the effectiveness of filtering, denoising, and normalization steps.

[0075] It can be understood that through the above steps, the quality of sEMG signals has been significantly improved, and noise and interference have been effectively removed. The pre-processed signals are clearer and more stable, providing a reliable basis for subsequent signal analysis and application.

[0076] S102: Based on the FDE algorithm, fractional calculus is combined and a multi-scale processing process is introduced to obtain the RCMFFDE algorithm.

[0077] Among them, the FDE algorithm refers to Fluctuation Dispersion Entropy, which is an entropy measurement method used to measure the complexity of time series signals. It is based on the statistical characteristics of signal fluctuations and evaluates the complexity and irregularity of signals by calculating the degree of dispersion of signals on different time scales. This application uses the FDE algorithm as the abbreviation of Fluctuation Dispersion Entropy.

[0078] Specifically, the RCMFFDE algorithm is obtained based on the FDE algorithm combined with fractional calculus and the introduction of multi-scale processing, including:

[0079] Based on the FDE algorithm and combined with fractional calculus, the FFDE algorithm is obtained;

[0080] On the basis of the FFDE algorithm, a multi-scale processing process is introduced to obtain the RCMFFDE algorithm. Among them, the FFDE algorithm refers to the Fractional Fluctuation-Based Dispersion Entropy (FFDE) algorithm, which is a muscle fatigue assessment method that combines the fractional order with the FDE algorithm to improve the sensitivity of the FDE algorithm. And on the basis of the FFDE algorithm, a new multi-scale factor is introduced to perform multi-scale analysis, and a Refined Composite Multiscale Fractional Fluctuation-Based Dispersion Entropy (RCMFFDE) algorithm is proposed, which is the RCMFFDE algorithm.

[0081] In specific implementation, the specific principles of obtaining the RCMFFDE algorithm include the following steps:

[0082] Step 1: First, based on the FDE algorithm and combined with fractional calculus, to obtain the FFDE algorithm, including:

[0083] Given a sEMG signal x={x 1 ,x 2 ,K,x n}, the FFDE algorithm is as follows:

[0084] (1) First, each data point x in the original sEMG signal j (j=1,2,...,N) is mapped to c categories with integer indexes from 1 to c, and the classified signal is u j (j=1, 2, ..., N). Five mapping methods are used in the invention.

[0085] The first mapping method is linear mapping (LM), which is simple and fast to calculate. The second method uses sorting mapping (SORT), which first calculates x j (j=1,2,...,N) are sorted, and then the sorted data points are evenly distributed into c categories, ensuring that each category contains the same number of x j .

[0086] Then, two nonlinear mapping formulas, log-sigmoid (LOGSIG) and tan-sigmoid (TANSIG), are used as (3) and (4):

[0087]

[0088] Where σ and μ are the standard deviation and mean of the time series (i.e., sEMG signal) x, respectively.

[0089] The last mapping method is the normal distribution (Normal Cumulative Distribution Function, NCDF), as shown in formula (5):

[0090]

[0091] In the formula, μ and σ 2 They represent the expectation and variance of the time series (i.e., sEMG signal) x, respectively, and t is the time.

[0092] (2) Further, y is mapped to a time series u through linear transformation i :

[0093]

[0094] Where R is the rounding function and c is the number of categories. After steps (1) and (2), each element in the time series x is mapped to [1, 2, ..., c].

[0095] (3) Then use equation (7) to calculate the embedding vector, where m is the embedding dimension, d is the time delay, and i = 1, 2, N-(m-1)d.

[0096]

[0097] (4) Each time series Mapping to dispersion mode The number of dispersion modes is equal to c m .

[0098] (5) Next, calculate the probability of each dispersion mode:

[0099]

[0100] In the formula, Indicates that each The number of dispersion modes is divided by the total number of embedded signals with embedding dimension m.

[0101] (6) According to the definition of Shannon entropy, DispEn is defined as follows:

[0102]

[0103] (7) The FDE algorithm considers the differences between adjacent elements of the dispersion pattern, which is called the fluctuation-based dispersion pattern. In this way, there is a vector of length m-1, each of whose elements changes from -c+1 to c-1. Therefore, there exists (2c-1) m-1 The potential dispersion mode based on fluctuations is as follows:

[0104]

[0105] (8)Finally, based on fractional calculus, the FFDE algorithm is defined as:

[0106]

[0107] Where q is the fractional order. Γ(·) and Ψ(·) represent the gamma function and the digital matrix function, respectively.

[0108] Finally, using FFDE / ln((2c-1) m-1 ) for standardization.

[0109] Step 2: Based on the FFDE algorithm, a multi-scale processing process is introduced to obtain the RCMFFDE algorithm. Furthermore, the multi-scale processing process refers to: coarse-graining the sEMG signal using the median of the sEMG signal through different scale factors, wherein the coarse-graining process is used to evaluate the entropy characteristics of the sEMG signal at various scales. Specifically, it includes:

[0110] In order to further analyze the complexity of sEMG signals at different time scales, a multi-scale processing step is introduced into the FFDE algorithm in step 1 above. In this step, the original sEMG signal is coarse-grained through a series of different scale factors to evaluate the entropy characteristics of the signal at various scales. The present invention proposes for the first time to use the median to achieve coarse-graining. Compared with the traditional average and time-shift methods, this method is not only faster and more efficient in calculation, but also effectively reduces the influence of extreme points. Through this method, the original signal can be accurately coarse-grained at various scales, thereby generating a series of representative coarse-grained time series.

[0111] For the original sEMG signal x={x 1 ,x 2 ,K,x n} for coarse-graining, select the scale factor τ, and construct a moving window sequence based on the median in The calculation method is:

[0112]

[0113] Wherein, median is the median function, 1≤j≤N-τ+1, 1≤k≤τ.

[0114] For the moving window sequence The probability of each fluctuation mode in is calculated as:

[0115]

[0116] In the formula, is The probability of a medium volatility pattern.

[0117] Finally, the RCMFFDE algorithm is defined as:

[0118]

[0119] As a preferred embodiment, this embodiment also provides an optimal fractional-order FFDE experimental idea:

[0120] Using analog signals and sEMG signals generated by one-dimensional logical mapping, the influence of fractional order q on the entropy value of the FFDE algorithm is studied.

[0121] The one-dimensional logical mapping can be expressed as follows:

[0122] x n+1 =ux n (1-x n ) (15)

[0123] Where 0<x 0<1 and 0<u≤4. x n is the simulated signal after the nth iteration, x 0 It is the initial value before the iteration starts, that is, the value of x when n=0. Studies have shown that when u>3.56995, the one-dimensional logical mapping will exhibit chaotic behavior.

[0124] In this embodiment, the parameter x 0 Set to 0.4; parameter u is set to 3.5~4, with a step size of 0.1, and a periodic sequence and five chaotic sequences are obtained; the fractional order q is set to -0.5~0.5, with a step size of 0.05, in order to verify the sensitivity of the FFDE algorithm to dynamic changes under different fractional orders and determine the optimal q value.

[0125] S103: Perform multi-scale analysis on the pre-processed sEMG signal using the RCMFFDE algorithm to obtain multiple entropy curves.

[0126] Furthermore, the RCMFFDE algorithm is used to perform multi-scale analysis on the pre-processed sEMG signal to obtain multiple entropy curves, including:

[0127] Performing coarse-graining processing on the sEMG signal to obtain a plurality of coarse-grained time series, wherein the plurality of coarse-grained time series are used to indicate the sEMG signal after coarse-graining, and the coarse-graining processing is used to indicate the multi-scale processing process;

[0128] According to the plurality of coarse-grained time series, a plurality of entropy curves are obtained by the FFDE algorithm, wherein the RCMFFDE algorithm is used to indicate an algorithm for introducing a multi-scale processing process on the basis of the FFDE algorithm. In specific implementation, the principles and applications of the multi-scale processing process, the FFDE algorithm, and the RCMFFDE algorithm can refer to the principles or application methods in step S103, which will not be described in detail here.

[0129] S104: Selecting an entropy value curve that meets a preset condition from among the multiple entropy value curves as a final entropy value curve.

[0130] Further, the selecting an entropy value curve that meets a preset condition from among the multiple entropy value curves as a final entropy value curve comprises:

[0131] Performing curve fitting on each of the entropy value curves to obtain the absolute value of the slope and the coefficient of determination of the fitting curve corresponding to each of the entropy value curves;

[0132] Multiply the absolute value of the slope and the coefficient of determination to obtain a curve evaluation index value;

[0133] The entropy curve corresponding to the maximum curve evaluation index value is selected as the final entropy curve. The final entropy curve is used to indicate the entropy curve corresponding to the best fitting line after curve fitting of the multiple entropy curves.

[0134] When implementing it, you can refer to the following methods:

[0135] The entropy value curve was fitted using the least squares method to determine the coefficients of the best fitting line in the data set.

[0136] The entropy curve z changes with time t, so there are data points (t 1 ,z 1 ),(t 2 ,z 2 ),...,(t n ,z n ), where t n It's time, n is the entropy value at the corresponding time point. Use the least squares method to fit a straight line z=mt+b, where m is the slope and b is the intercept.

[0137] (1) Calculate the average value of time t and entropy value z:

[0138]

[0139] (2) Calculate the slope m:

[0140]

[0141] (3) Calculate the intercept b:

[0142]

[0143] (4) Substituting m and b into the linear equation, we obtain the linear model of entropy changing with time:

[0144]

[0145] in, is the entropy of the model prediction.

[0146] (5) Calculation of the coefficient of determination R 2 :

[0147]

[0148]

[0149] Among them, SS total is the total variation of entropy values ​​(the sum of squared differences from their mean); SS resis the sum of squares of the residuals of the fit (i.e., the sum of squared differences between the actual value and the predicted value); R 2 It measures the goodness of model fitting. The closer the value is to 1, the better the fitting effect is and the more the curve conforms to the linear change trend.

[0150] It is understandable that the change trend of EMG signal indicators over time is usually not completely linear. However, in the current EMG fatigue analysis research, the determination coefficient R of linear fitting is often used. 2 The value of m reflects the ability of the algorithm to distinguish changes in muscle fatigue. The larger the absolute value of m, the stronger the ability of the algorithm to assess muscle fatigue. 2 It reflects the anti-interference ability of the algorithm. The larger the determination coefficient, the stronger the anti-interference ability. In this embodiment, |m|×R 2 The maximum is used as the preset condition of the judgment standard, and the best scale factor of the entropy value curve corresponding to the best fitting line is selected for analysis and comparison.

[0151] S105: Evaluate the degree of muscle fatigue according to the final entropy value curve.

[0152] Furthermore, the evaluating the degree of muscle fatigue according to the final entropy value curve includes: when the entropy value in the final entropy value curve gradually decreases, the degree of muscle fatigue gradually increases.

[0153] It is understandable that the final entropy curve changes with muscle fatigue, and has a clear downward trend as muscle fatigue occurs. During the process of muscle fatigue, the decrease in entropy reflects the transition of muscle activity from a more complex and random state to a more ordered state, which is a typical feature of muscle fatigue. As muscle fatigue intensifies, the complexity of the sEMG signal decreases, and the entropy value gradually decreases, which effectively reflects the process of muscle fatigue and can capture the changes in the sEMG signal during muscle fatigue.

[0154] The present application also provides a method of using Figure 1 The entropy-based muscle fatigue assessment method described in the embodiment is applied to assessing muscle fatigue status.

[0155] Specifically include:

[0156] Experimental subjects: 10 healthy volunteers were selected as experimental subjects, with an age range of 20-30 years old.

[0157] Experimental process: The subjects were asked to perform muscle contraction exercises of a specific intensity while collecting sEMG signals. During the exercise, sEMG signals were recorded at regular intervals and preprocessed and analyzed using the RCMFFDE algorithm.

[0158] Experimental results: According to the final entropy curve, the degree of muscle fatigue of the experimental subjects was successfully evaluated. The experimental results show that with the increase of exercise time, the entropy value gradually increases, reflecting the deepening of muscle fatigue. At the same time, the changes in entropy values ​​between different experimental subjects also show certain differences, reflecting the impact of individual differences on muscle fatigue assessment. Specifically, the experiment provided in this embodiment uses four multi-scale entropy methods to extract features from 10 groups of sEMG signals and analyze muscle fatigue information. According to the data analysis in Table 1, the average absolute value of the slope of the RCMFFDE algorithm (5.9667) is significantly higher than that of R2CMSE (3.3359), RCMDE (1.0372) and RCMFDE (1.7754), where R2CMSE refers to the Rapid Refined Composite Multiscale Sample Entropy (R2CMSE) algorithm, RCMDE refers to the Refined Composite Multiscale Dispersion Entropy (Refined Composite Multiscale Dispersion Entropy), which is a nonlinear dynamic analysis method, and RCMFDE refers to the Refined Composite Multiscale Fluctuation Dispersion Entropy (Refined Composite Multiscale Fluctuation Dispersion Entropy), which is a nonlinear dynamic method for analyzing complex signal characteristics. This shows that the RCMFFDE algorithm performs best in distinguishing muscle fatigue, while the RCMDE performs worst. In addition, the average determination coefficient of RCMDE is R 2 The value of RCMFFDE is 0.7634, slightly higher than 0.7337 of RCMFFDE. Although the introduction of fractional order makes RCMFFDE slightly inferior in anti-interference performance, it can still reflect the changes of muscle fatigue well. The results of independent sample t-test show that there is a significant statistical difference between RCMFFDE and other multi-scale entropy algorithms in muscle fatigue assessment (P<0.01), indicating that RCMFFDE is more sensitive to the dynamic changes of muscle fatigue, the linear relationship of the evaluation process is more obvious, and the overall effect is better than other algorithms.

[0159] Table 1 Comparison of four multi-scale entropy curve fitting results

[0160]

[0161]

[0162] Application areas and intended effects:

[0163] (1) Application and beneficial effects in sports training

[0164] application:

[0165] Evaluate the training effect of athletes: By monitoring the muscle fatigue of athletes during training, the training effect can be evaluated and scientific training guidance can be provided to coaches.

[0166] Prevent sports injuries: Timely understanding of athletes’ muscle fatigue status can prevent sports injuries caused by overtraining.

[0167] Develop personalized training plans: According to the athlete's muscle fatigue, a personalized training plan can be developed to improve training efficiency.

[0168] Beneficial effects:

[0169] Improve training effect: Through scientific evaluation methods, it can ensure that athletes train in the best condition, thereby improving training effect.

[0170] Reduce the risk of sports injuries: timely understanding of athletes' muscle fatigue status can help prevent sports injuries and protect athletes' physical health.

[0171] Promoting the personalized development of athletes: Developing personalized training plans based on the individual differences of athletes will help promote the personalized development of athletes.

[0172] (2) Application and beneficial effects in the field of human-computer interaction

[0173] application:

[0174] Assessing muscle fatigue during human-computer interaction: During human-computer interaction, users need to use certain muscle groups for a long time. By monitoring the fatigue level of these muscle groups, the comfort and efficiency of human-computer interaction can be assessed.

[0175] Optimize human-computer interaction design: Based on the muscle fatigue assessment results, the human-computer interaction design can be optimized to improve user experience and interaction efficiency.

[0176] Beneficial effects:

[0177] Improve user experience: By optimizing the human-computer interaction design, the user's muscle fatigue during use can be reduced and the user experience can be improved.

[0178] Promote the development of human-computer interaction technology: The entropy-based muscle fatigue assessment method provides new ideas and methods for the development of human-computer interaction technology, which helps to promote innovation and development in this field.

[0179] In summary, the entropy-based muscle fatigue assessment method has broad application prospects and important practical significance in the fields of sports training and human-computer interaction. Through the application of this method, the training effect can be improved, the risk of sports injuries can be reduced, the human-computer interaction design can be optimized, and the user experience can be improved.

[0180] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0181] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A muscle fatigue assessment method based on entropy, characterized in that: include: Collecting sEMG signals and preprocessing the sEMG signals; Based on the FDE algorithm combined with fractional calculus and the introduction of multi-scale processing, the RCMFFDE algorithm is obtained; Performing multi-scale analysis on the preprocessed sEMG signal using the RCMFFDE algorithm to obtain multiple entropy curves; Selecting an entropy value curve that meets a preset condition from among the multiple entropy value curves as a final entropy value curve; The degree of muscle fatigue was evaluated based on the final entropy curve.

2. The method according to claim 1, characterized in that The collecting of sEMG signals and preprocessing of the sEMG signals include: The sEMG signal is collected and denoised by a signal denoising method based on VMD.

3. The method according to claim 1, characterized in that The FDE algorithm is combined with fractional calculus and a multi-scale processing process is introduced to obtain the RCMFFDE algorithm, including: Based on the FDE algorithm and combined with fractional calculus, the FFDE algorithm is obtained; A multi-scale processing process is introduced on the basis of the FFDE algorithm to obtain the RCMFFDE algorithm.

4. The method according to claim 1, characterized in that: The multi-scale processing process includes: The sEMG signal is subjected to coarse-graining processing by using the median of the sEMG signal through different scale factors, wherein the coarse-graining processing is used to evaluate the entropy characteristics of the sEMG signal at various scales.

5. The method according to claim 3, characterized in that: The RCMFFDE algorithm is used to perform multi-scale analysis on the pre-processed sEMG signal to obtain multiple entropy curves, including: Performing coarse-graining processing on the sEMG signal to obtain a plurality of coarse-grained time series, wherein the plurality of coarse-grained time series are used to indicate the sEMG signal after coarse-graining, and the coarse-graining processing is used to indicate the multi-scale processing process; According to the plurality of coarse-grained time series, a plurality of entropy value curves are obtained through the FFDE algorithm, wherein the RCMFFDE algorithm is used to indicate an algorithm for introducing a multi-scale processing process on the basis of the FFDE algorithm.

6. The method according to claim 1, characterized in that The step of selecting an entropy value curve that meets a preset condition from among the plurality of entropy value curves as a final entropy value curve comprises: Performing curve fitting on each of the entropy value curves to obtain the absolute value of the slope and the coefficient of determination of the fitting curve corresponding to each of the entropy value curves; Multiply the absolute value of the slope and the coefficient of determination to obtain a curve evaluation index value; The entropy curve corresponding to the maximum curve evaluation index value is selected as the final entropy curve.

7. The method according to claim 1, wherein the step of evaluating the degree of muscle fatigue according to the final entropy curve comprises: When the entropy value in the final entropy value curve gradually decreases, the degree of muscle fatigue gradually increases.

8. Use of the method according to any one of claims 1 to 7 in assessing muscle fatigue status.

Images