Wearable myoelectricity monitoring and electrical stimulation intervention integrated device

Abstract

The invention discloses a wearable myoelectricity monitoring and electrical stimulation intervention integrated device, and the device comprises an integrated function patch which collects surface myoelectricity signals in a motion state and applies electrical stimulation to muscles; the electromyographic signal preprocessing module is used for preprocessing a surface electromyographic signal sequence; the power generation period electromyographic signal processing module is used for extracting a power generation period signal sub-sequence in the muscle gait movement from the preprocessed surface electromyographic signal sequence; calculating the characteristic value of each force generation period signal subsequence, wherein the characteristic value of each force generation period signal subsequence reflects the deviation degree of each signal in the force generation period; the control module is used for creating an eigenvalue sequence based on the eigenvalue of each force generation period signal sub-sequence; generating an accumulated feature enhancement sequence based on the feature value sequence; judging whether the selected element is smaller than a preset shrinkage feature threshold value or not from the accumulated feature enhancement sequence; when the constriction characteristic value is smaller than the preset constriction characteristic threshold value, treatment stimulation electric pulses continue to be generated, and otherwise, relaxation electric pulses are generated.

Description

Technical Field

[0001] The present invention relates to the technical field of electromyography detection, and in particular to a wearable electromyography monitoring and electrical stimulation intervention integrated device. Background Art

[0002] Abnormal neuromuscular control or motor neuron damage often lead to muscle atrophy, weakness, fatigue, abnormal contraction pattern and other phenomena. Surface electromyography can reflect the function of neural control and muscles to a certain extent. Its essence is the superposition of motor unit action potentials (MUAP) generated by the activation of multiple motor units (MU) at the detection electrode. By collecting the surface electromyography signals of the corresponding muscles and extracting features, its neuromuscular control status can be judged to a large extent. Therefore, electromyography features are often used as neuromuscular control evaluation indicators or related intervention method effect evaluation indicators. Neuromuscular electrical stimulation (NMES) is a common method of neuromuscular control intervention. It mainly uses prefabricated patterns, postures or timing (such as gait cycles, etc.) to perform medium and low frequency electrical stimulation on the corresponding muscles (groups) to activate or inhibit neural control, improve muscle atrophy, and adjust the duration or amplitude of muscle contraction, thereby correcting the muscle contraction pattern in the gait movement cycle and improving neuromuscular control.

[0003] In the process of realizing the present invention, the inventors discovered the following technical problems: At present, when conducting neuromuscular monitoring and intervention, it is often necessary to first collect electromyographic signals and extract features, and then match the corresponding fixed electrical stimulation parameters for intervention treatment. Real-time adjustment cannot be achieved. Even if there are relatively complete monitoring and stimulation adaptive adjustment methods, the adjustment intervention is mainly based on gait cycle timing or body angle. There is a mismatch between the basis of electrical stimulation intervention and the target parameters, resulting in a certain deviation in monitoring and intervention, making it difficult to achieve real-time and accurate neuromuscular control intervention, reducing the effectiveness of actual rehabilitation applications such as walking learning or gait correction. Summary of the invention

[0004] The embodiment of the present invention provides a wearable electromyography monitoring and electrical stimulation intervention integrated device to solve the technical problem that neuromuscular monitoring and intervention in the prior art cannot be adjusted in real time.

[0005] In view of this, an embodiment of the present invention provides a wearable electromyography monitoring and electrical stimulation intervention integrated device, comprising:

[0006] An integrated functional patch for collecting surface electromyographic signals during exercise and applying electrical stimulation to muscles;

[0007] An electromyographic signal preprocessing module is used to preprocess the surface electromyographic signal sequence collected by the integrated functional patch;

[0008] The force-generating electromyographic signal processing module is used to extract the force-generating signal subsequence in the muscle gait movement from the preprocessed surface electromyographic signal sequence; calculate the characteristic value of each of the force-generating signal subsequences, and the characteristic value of the force-generating signal subsequence reflects the sum of the deviation degrees of each signal in the force-generating period;

[0009] A control module is used to create a feature value sequence based on the feature value of each force period signal subsequence; generate a cumulative feature enhancement sequence based on the feature value sequence; select a preset number of elements in a back-to-front order from the cumulative feature enhancement sequence, and determine whether the selected elements are less than a preset contraction feature threshold; when it is less than the preset contraction feature threshold, control the integrated functional patch to continue generating therapeutic electric pulses, otherwise, control the integrated functional patch to generate relaxation electric pulses.

[0010] Furthermore, the wearable electromyography monitoring and electrical stimulation intervention integrated device also includes:

[0011] A muscle contraction abnormality reference value generation module is used to generate a muscle contraction abnormality reference value according to a linear quantization estimation adjustment factor and elements in a cumulative feature enhancement sequence;

[0012] The muscle contraction abnormality reference value output module is used to output the muscle contraction abnormality reference value generated by the muscle contraction abnormality reference value generation module for reference.

[0013] Furthermore, the control module includes:

[0014] A stimulation electric pulse intensity calculation unit, used for calculating the stimulation electric pulse intensity according to the last element in the cumulative feature enhancement sequence, the fat layer and the skin thickness;

[0015] The relaxation electric pulse intensity calculation unit is used to calculate the relaxation electric pulse intensity according to the last element in the cumulative feature enhancement sequence, the fat layer and the skin thickness.

[0016] Furthermore, the electromyographic signal processing module during the force generation period includes:

[0017] The force-exerting period electromyographic signal division unit is used to divide the force-exerting period according to the degree of stable change of the collected surface electromyographic signal.

[0018] Furthermore, the electromyographic signal preprocessing module includes:

[0019] The normalization processing unit is used to normalize the collected surface electromyography signal sequence.

[0020] Furthermore, the electromyographic signal preprocessing module also includes:

[0021] The denoising unit is used to perform denoising on the normalized surface electromyography signal.

[0022] Furthermore, the denoising unit is used to:

[0023] Select a wavelet basis type and a wavelet decomposition level N, perform N-layer wavelet decomposition on the noisy signal, and obtain the wavelet coefficients of each layer;

[0024] The threshold quantization of high-frequency coefficients of wavelet decomposition adopts soft threshold quantization method and sets the threshold λ. When the wavelet coefficients converge to 0, the coefficients greater than λ are set as parameters related to λ. For the wavelet coefficients less than λ, they are directly set to 0 to obtain a smoother denoised signal.

[0025] Reconstruct the processed signal to obtain a new sequence after denoising.

[0026] The wearable electromyography monitoring and electrical stimulation intervention integrated device provided by the embodiment of the present invention is used to collect surface electromyography signals in a motion state and apply electrical stimulation to muscles by setting an integrated functional patch; an electromyography signal preprocessing module is used to preprocess the surface electromyography signal sequence collected by the integrated functional patch; a force period electromyography signal processing module is used to extract the force period signal subsequence in the muscle gait movement from the preprocessed surface electromyography signal sequence; calculate the characteristic value of each of the force period signal subsequences, and the characteristic value of the force period signal subsequence reflects the sum of the deviation degrees of each signal in the force period; a control module is used to create a characteristic value sequence based on the characteristic value of each force period signal subsequence; generate a cumulative characteristic enhancement sequence based on the characteristic value sequence; select a preset number of elements in a back-to-front order from the cumulative characteristic enhancement sequence, and judge whether the selected elements are less than a preset contraction characteristic threshold; when it is less than the preset contraction characteristic threshold, control the integrated functional patch to continue to generate therapeutic electric pulses, otherwise, control the integrated functional patch to generate relaxation electric pulses. The integrated myoelectric detection patch worn on the patient's muscle surface can collect muscle group signals in real time through the myoelectric detection electrodes and transmit them to the signal processing module for preprocessing; then the feature extraction module performs digital filtering and dynamic myoelectric feature calculation; the obtained myoelectric features are analyzed by the feature recognition unit and fed back to the main control, which outputs an electrical stimulation signal with a certain frequency and duty cycle, which is amplified and processed by the electrical stimulation circuit and applied to the patient's muscle parts by the electrical stimulation electrodes. Real-time monitoring can be achieved and intervention can be carried out in real time according to the monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0028] Figure 1 It is a structural schematic diagram of a computing device in a wearable electromyography monitoring and electrical stimulation intervention integrated device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0030] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0031] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and the like are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", and the like may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0032] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood by specific circumstances.

[0033] Figure 1 is a schematic diagram of the structure of a wearable electromyography monitoring and electrical stimulation intervention integrated device provided by an embodiment of the present invention, see Figure 1 The wearable electromyography monitoring and electrical stimulation intervention integrated device comprises:

[0034] An integrated functional patch can be realized by combining a wire cap, a flexible patch, a surface electromyography sensor and an electrical stimulation treatment electrode. By applying the flexible patch to the surface of the user's muscles, the electromyography signal generated by the user's walking can be collected. The electrical stimulation treatment electrode can generate a corresponding pulse signal. The surface electromyography sensor adopts a multi-channel electrode system and can collect 6 to 8 channel signals. The electrode base uses a gel patch electrode, which is a bipolar type, and a reference electrode is inserted between the two electrodes to reduce noise and improve the ability to suppress common mode signals.

[0035] Accordingly, the wearable electromyography monitoring and electrical stimulation intervention integrated device may also include a main control system, which can process the collected surface electromyography signals in real time to determine whether the current muscle is in an abnormal contraction state. The main control system can generate a control signal based on the judgment structure, and use the control signal to control the electrical stimulation treatment electrode to switch between treatment and sending electrical pulses.

[0036] Exemplarily, the main control system may include:

[0037] The electromyographic signal preprocessing module is used to preprocess the surface electromyographic signal sequence collected by the integrated functional patch to meet the requirements of judging whether the muscle is currently in an abnormal contraction state. Optionally, the electromyographic signal preprocessing module can implement normalization and denoising. Exemplarily, the AD module can convert the line analog-to-digital to form a digital sequence, and the digital sequence can be preprocessed accordingly.

[0038] The force period electromyographic signal processing module is used to extract the force period signal subsequence in the muscle gait movement from the preprocessed surface electromyographic signal sequence; calculate the characteristic value of each force period signal subsequence, and the characteristic value of the force period signal subsequence reflects the sum of the deviation degrees of each signal in the force period.

[0039] First, the force generation period can be determined based on the smoothness of the signal in the surface electromyographic signal sequence obtained by the electromyographic signal preprocessing module. Since the muscle movement in the force generation period is relatively tighter, the electromyographic signal intensity is greater. Therefore, the smoothness can be used to extract the force generation period signal subsequence in the muscle gait movement from the preprocessed surface electromyographic signal sequence.

[0040] After obtaining the force-generating period signal subsequence, calculate the characteristic value of each of the force-generating period signal subsequences, and the characteristic value is used to reflect the degree of deviation of each signal in the force-generating period. Exemplarily, the degree of deviation can be calculated by calculating the average value of each signal in the force-generating period, and calculating the difference between each signal and the average value. Or the signal value of a certain action or moment in the force-generating period of the user's movement gait is used as the standard value, and the standard value of each user is different. The characteristic value is calculated using the standard value.

[0041] In this embodiment, the characteristic value of each of the force-generating phase signal subsequences is calculated in the following manner:

[0042]

[0043] Among them, S i represents the characteristic value of the force generation period in the i-th gait motion cycle in the electromyographic sequence X″′, x i ″′(p) is the pth myoelectric signal value during the force generation period of the i-th gait motion cycle, p∈[1,m],i∈[1,q], m,q are both positive integers.

[0044] Using this calculation method, the discrete degree of the signal quantity in the force period can be fully reflected, and the sequence structure and properties can be measured.

[0045] A control module is used to create a feature value sequence based on the feature value of each force period signal subsequence; generate a cumulative feature enhancement sequence based on the feature value sequence; select a preset number of elements in a back-to-front order from the cumulative feature enhancement sequence, and determine whether the selected elements are less than a preset contraction feature threshold; when it is less than the preset contraction feature threshold, control the integrated functional patch to continue generating therapeutic electric pulses, otherwise, control the integrated functional patch to generate relaxation electric pulses.

[0046] In this embodiment, the control module can continuously perform real-time detection based on the force-generating period signal subsequence corresponding to the real-time collected signal. Exemplarily, a feature value sequence can be created based on the feature value of each force-generating period signal subsequence; and a cumulative feature enhancement sequence can be generated based on the feature value sequence.

[0047] For example, the characteristic value sequence S can be composed of S=[S 1 ,S 2 ,……,S i ], where the number of elements in the eigenvalue sequence S is increasing. The two adjacent terms of the obtained eigenvalue sequence are added together to form a cumulative feature enhancement sequence. This is achieved by:

[0048] I=[I 1 ,I 2 ,...,Ij],j∈[1,i-1], where, I j =Sj+Sj +1 Similar to the eigenvalue sequence, the elements in the cumulative feature enhancement sequence are also increasing.

[0049] In this embodiment, the reason why the elements in the cumulative feature enhancement sequence are used to evaluate the degree of muscle contraction, rather than directly using the eigenvalue calculation, is mainly because it is found in the actual operation process that the use of the eigenvalue of a force period alone is prone to misjudgment, that is, the eigenvalue is closely related to each force period and the walking environment, as well as the walking method. Even if the electromyographic signal is pre-processed in the early stage, a large error will still occur. Experiments have shown that a single eigenvalue sequence has the disadvantages of disorder, nonlinearity and randomness, which is not conducive to judging the degree of fatigue. However, the method of cumulative feature enhancement sequence processing has been proved by experiments to be linear and orderly. It can better and more accurately reflect the degree of contraction.

[0050] Exemplarily, since the number of elements in the cumulative feature enhancement sequence increases with the user's walking movement and real-time acquisition, and since the cumulative feature enhancement sequence corresponding to the electromyographic signal acquired earlier has little reference significance, it is possible to determine whether the selected element is less than the preset contraction feature threshold based on a preset number of elements in order from back to front, where the last element is the most recently generated.

[0051] For the analysis of muscle contraction abnormalities, the following non-absolute quantitative methods can be used: j When ≥α, the possibility of muscle contraction abnormality is the same as I j The size of is negatively correlated; when I j When <α, the muscle is in the abnormal contraction stage, where α is the normal contraction threshold of the muscle, which is usually 0.00716.

[0052] Optionally, after the relevant cumulative sequence is acquired, the cumulative sequence signal is transmitted to the PC via Bluetooth, and the electrical stimulation instruction is transmitted back after the PC makes a decision.

[0053] The intervention signal is in the form of a pulse, with a pulse frequency between 50-75 nz and a duration of 200-400 μs. The pulse signal is used to allow the muscle to better adapt to the current. The control module may include: a stimulation electric pulse intensity calculation unit, which is used to calculate the stimulation electric pulse intensity according to the last element in the cumulative feature enhancement sequence, the fat layer and the skin thickness; a relaxation electric pulse intensity calculation unit, which is used to calculate the relaxation electric pulse intensity according to the last element in the cumulative feature enhancement sequence, the fat layer and the skin thickness. Regarding real-time intensity adjustment, when I j ≥α, the unit applies the same j The relaxation mode electric pulse corresponding to the value, the electric stimulation intensity is within the voltage amplitude upper limit of 5V and the current amplitude upper limit of 100mA; when I j <α, the unit applies the same j The value corresponds to the enhanced mode electric pulse. j≥α when electric pulse intensity γ=10000×K 1 ×(I j -α)mA. When I j <α, the electrical stimulation intensity γ=10000×K 2 ×(I j -α-I j )mA. K 1 With K 2 The value of K is determined by the fat layer and skin thickness of the person being tested. 1 The value is between 0.23-0.30, K 2 The value is between 0.46-0.54.

[0054] The wearable electromyography monitoring and electrical stimulation intervention integrated device may also be provided with a muscle contraction abnormal reference value generation module, which is used to generate a muscle contraction abnormal reference value based on a linear quantization estimation adjustment factor and elements in an accumulated feature enhancement sequence; and a muscle contraction abnormal reference value output module, which is used to output the muscle contraction abnormal reference value generated by the muscle contraction abnormal reference value generation module for reference.

[0055] Exemplarily, the value of each element in the cumulative sequence is input into a specific calculation and comparison unit to evaluate the possibility of abnormal muscle contraction, and the calculation method is as follows: j =1-β×I j , where ρ j Represents the possibility of abnormal muscle contraction in the jth gait movement cycle during exercise. β can be a linear quantitative estimation parameter. The above evaluation results can provide a reference for doctors or users. For doctors, targeted treatment can be carried out, and for users, the movement mode and force mode can be adjusted in time to avoid injury.

[0056] An integrated functional patch is set up to collect surface electromyographic signals in motion and apply electrical stimulation to muscles; an electromyographic signal preprocessing module is used to preprocess the surface electromyographic signal sequence collected by the integrated functional patch; a force period electromyographic signal processing module is used to extract the force period signal subsequence in the muscle gait movement from the preprocessed surface electromyographic signal sequence; the characteristic value of each force period signal subsequence is calculated, and the characteristic value of the force period signal subsequence reflects the sum of the deviation degrees of each signal in the force period; a control module is used to create a characteristic value sequence based on the characteristic value of each force period signal subsequence; a cumulative feature enhancement sequence is generated based on the characteristic value sequence; a preset number of elements in a back-to-forward order are selected from the cumulative feature enhancement sequence, and it is determined whether the selected elements are less than a preset real-time contraction feature threshold; when it is less than the preset contraction feature threshold, the integrated functional patch is controlled to generate a stimulating electric pulse, otherwise, the integrated functional patch is controlled to generate a relaxing electric pulse. The integrated myoelectric detection patch worn on the patient's muscle surface can collect muscle group signals in real time through the myoelectric detection electrodes and transmit them to the signal processing module for preprocessing; then the feature extraction module performs digital filtering and dynamic myoelectric feature calculation; the obtained myoelectric features are analyzed by the feature recognition unit and fed back to the main control, which outputs an electrical stimulation signal with a certain frequency and duty cycle, which is amplified and processed by the electrical stimulation circuit and applied to the patient's muscle parts by the electrical stimulation electrodes. Real-time monitoring can be achieved and intervention can be carried out in real time according to the monitoring results.

[0057] In addition, in this embodiment, the electromyographic signal preprocessing module includes: a normalization processing unit, which is used to normalize the collected surface electromyographic signal sequence. Exemplarily, the electromyographic signal collected by the flexible circuit group is first converted into digital form by the AD module, and a certain length of original signal x is collected each time. 1 ,x 2 ,……,x n The signal length depends on the actual situation. Assume that the maximum value of the original signal during the signal detection process is x t , that is, in x t After that, the size of the original electrical signal data is significantly reduced, and the judgment criteria are when When the detection cycle is judged to be a valid cycle, otherwise it is regarded as a noise signal filtering. Time determination x n This is the last electromyographic data of this cycle.

[0058] The collected single cycle electromyographic signal is normalized. The normalization is calculated based on the following formula: Where X′=x(n)′∈[0,1] is the normalized EMG signal sequence, n is the position of the element in the sequence, and X=x(n) is the EMG signal sequence before normalization. The purpose of normalization is that the existence of singular sample data will increase the data processing time, resulting in the main control processing timeout, and the real-time integration of detection and stimulation cannot be achieved. It may also cause the algorithm to fail to converge during training. Therefore, the preprocessed data needs to be normalized before data processing; if normalization is not performed, the objective function feature will be missing due to the large difference in the values ​​of different features in the feature vector. In this way, when the gradient is descended, the direction of the gradient will deviate from the direction of the minimum value, which significantly affects the efficiency of real-time electrical stimulation.

[0059] In addition, the electromyographic signal preprocessing module also includes: a denoising unit, which is used to denoise the surface electromyographic signal after normalization. The denoising unit is used to: select a wavelet basis type and a wavelet decomposition level N, perform N-layer wavelet decomposition on the noisy signal, and obtain the wavelet coefficients of each layer; threshold quantization of the high-frequency coefficients of the wavelet decomposition, adopt a soft threshold quantization method, set the threshold λ, when the wavelet coefficients converge to 0, the coefficients greater than λ are set to parameters related to λ, and the wavelet coefficients less than λ are directly set to 0 to obtain a smoother denoised signal; reconstruct the processed signal to obtain a new denoised sequence.

[0060] For example, the normalized EMG signal sequence can be subjected to wavelet denoising. The expression of the noisy signal is: x(t) = y(t) + n(t). Wavelet denoising is divided into three steps. The first step is the wavelet decomposition of the signal. A wavelet basis type and a wavelet decomposition level N are selected. The noisy signal is subjected to N-layer wavelet decomposition to obtain the wavelet coefficients of each layer. The second step is the threshold quantization of the high-frequency coefficients of the wavelet decomposition. A soft threshold quantization method is adopted to set a threshold MAD is the median absolute deviation, and n is the length of the signal. When the wavelet coefficients converge to 0, the coefficients greater than λ are set to sgn(W)(|W|-λ), and the wavelet coefficients less than λ are directly set to 0, which can obtain a smoother denoised signal. The third step is wavelet signal reconstruction, which reconstructs the processed signal to obtain a new denoised sequence.

[0061] The purpose of wavelet denoising is that in the process of integrated detection and treatment, the process of detecting electromyographic signals is also accompanied by the interference of electrical stimulation signals, and the wavelet denoising algorithm is simple and clear, and has a fast calculation speed. If N is the length of the signal, its calculation speed is O(N). In addition, the electrical stimulation signal is controlled by the main control unit and can be regarded as noise in a known frequency range. The frequency bands of the signal and the noise are separated from each other, and the denoising effect is excellent.

[0062] By using the above-mentioned means, the collected electromyographic signals can be effectively preprocessed, thereby providing a judgment basis for the subsequent real-time and accurate electrical stimulation intervention.

[0063] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A wearable electromyography monitoring and electrical stimulation intervention integrated device, characterized in that: include: An integrated functional patch for collecting surface electromyographic signals during exercise and applying electrical stimulation to muscles; An electromyographic signal preprocessing module is used to preprocess the surface electromyographic signal sequence collected by the integrated functional patch; The force-generating electromyographic signal processing module is used to extract the force-generating signal subsequence in the muscle gait movement from the pre-processed surface electromyographic signal sequence; Calculating a characteristic value of each of the force-generating period signal subsequences, wherein the characteristic value of the force-generating period signal subsequence reflects the degree of deviation of each signal in the force-generating period; A control module, configured to create a feature value sequence based on the feature value of each force-generating period signal subsequence; and generate a cumulative feature enhancement sequence based on the feature value sequence; Select a preset number of elements in a backward order from the cumulative feature enhancement sequence, and determine whether the selected elements are less than a preset contraction feature threshold; if the value is less than the preset contraction feature threshold, control the integrated functional patch to continue to generate therapeutic electrical pulses, otherwise, control the integrated functional patch to generate relaxation electrical pulses.

2. The wearable electromyography monitoring and electrical stimulation intervention integrated device according to claim 1, characterized in that: The wearable electromyography monitoring and electrical stimulation intervention integrated device also includes: A muscle contraction abnormality reference value generation module is used to generate a muscle contraction abnormality reference value according to a linear quantization estimation adjustment factor and elements in a cumulative feature enhancement sequence; The muscle contraction abnormality reference value output module is used to output the muscle contraction abnormality reference value generated by the muscle contraction abnormality reference value generation module for reference.

3. The wearable electromyography monitoring and electrical stimulation intervention integrated device according to claim 1, characterized in that: The control module comprises: A stimulation electric pulse intensity calculation unit, used for calculating the stimulation electric pulse intensity according to the last element in the cumulative feature enhancement sequence, the fat layer and the skin thickness; The relaxation electric pulse intensity calculation unit is used to calculate the relaxation electric pulse intensity according to the last element in the cumulative feature enhancement sequence, the fat layer and the skin thickness.

4. The wearable electromyography monitoring and electrical stimulation intervention integrated device according to claim 1, characterized in that: The electromyographic signal processing module during the power generation period includes: The force-exerting period electromyographic signal division unit is used to divide the force-exerting period according to the degree of stable change of the collected surface electromyographic signal.

5. The wearable electromyography monitoring and electrical stimulation intervention integrated device according to claim 1, characterized in that: The electromyographic signal preprocessing module comprises: The normalization processing unit is used to normalize the collected surface electromyography signal sequence.

6. The wearable electromyography monitoring and electrical stimulation intervention integrated device according to claim 5, characterized in that: The electromyographic signal preprocessing module also includes: The denoising unit is used to perform denoising on the normalized surface electromyography signal.

7. The wearable electromyography monitoring and electrical stimulation intervention integrated device according to claim 6, characterized in that: The denoising unit is used for: Select a wavelet basis type and a wavelet decomposition level N, perform N-layer wavelet decomposition on the noisy signal, and obtain the wavelet coefficients of each layer; The threshold quantization of high-frequency coefficients of wavelet decomposition adopts soft threshold quantization method and sets the threshold λ. When the wavelet coefficients converge to 0, the coefficients greater than λ are set as parameters related to λ. For the wavelet coefficients less than λ, they are directly set to 0 to obtain a smoother denoised signal. Reconstruct the processed signal to obtain a new sequence after denoising.

Images