A method and system for removing central electroencephalic components from abdominal surface myoelectric signals

Abstract

The application discloses a method and system for removing central electrical components of abdominal muscle surface myoelectric signals, and comprises the following steps: obtaining an initial digital signal after pre-processing of the abdominal muscle surface myoelectric signals; extracting a template from the initial digital signal to obtain a R peak value horizontal and vertical coordinate set; constructing a training sample set according to the R peak value horizontal and vertical coordinate set; collecting an analog signal of the abdominal muscle surface myoelectric electrode again, and pre-processing the analog signal through signal amplification and power frequency filtering; intercepting a signal with the same length as the training sample as a to-be-processed signal; processing the to-be-processed signal and the training sample set through an N-order electrocardio filter to obtain a waveform of the abdominal muscle myoelectric signal with the electrocardio signal filtered out, and displaying the waveform in real time. The N-order electrocardio filter is used to greatly reduce the computing power requirement, effectively filter out the electrocardio signal while retaining the abdominal muscle myoelectric signal components with high fidelity, and facilitate detection of the abdominal muscle surface myoelectric signal of the upper abdomen.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a method and system for removing the electromyographic components of the electromyographic signal on the surface of the abdominal muscles. Background Technology

[0002] Electromyography (EMG) refers to the electrical activity of muscles, specifically the electrical signals generated during muscle contraction. These signals are produced by muscle fibers within the muscle. When a muscle contracts, the muscle proteins within the fibers slide against each other, causing changes in charge distribution and generating an EMG signal. This EMG signal can be captured using EMG electrodes, amplified, and processed to produce an electromyogram (EMG). EMG can be used to study muscle activity patterns, muscle fatigue levels, and muscle dysfunction.

[0003] Surface electromyography (sEMG) is widely used as an objective means and method to reflect the bioelectrical activity of the neuromuscular system. Its greatest feature is its non-invasiveness. sEMG signals are one-dimensional time-series signals of voltage changes obtained by guiding, amplifying, recording, and displaying the bioelectrical changes generated by single or multiple cells or parts of muscle tissue during activity, through electrode guidance. By placing electrode patches at the midline of the rectus abdominis muscle, the electromyographic signals of the abdominal muscles during movement and rest can be accurately detected and displayed in real time. However, because the position of the electrodes is relatively close to the heart, the collected abdominal muscle electromyographic signals are mixed with electrocardiogram (ECG) signals. In actual monitoring of abdominal muscle EMG, there is still a need to monitor the upper abdominal muscles, so it is unavoidable to place the electrode patches on the upper abdomen, thus introducing ECG interference. The closer the electrode patches are to the heart, the more severe the interference.

[0004] In the industry, interference from respiratory signals is typically addressed by using filters in specific frequency bands. However, crudely using filters can lead to several problems. First, filters can cause phase changes that do not correspond to the actual movement distance, resulting in inaccurate measurement of minute movements. Second, while filters are effective for signals during motion, they can cause signal distortion and abrupt changes when used on stationary signals. Furthermore, using high-order filters requires significant computation, making it impossible to display the signal waveform in real time.

[0005] Furthermore, the industry typically uses filters with specific frequency bands to filter out ECG noise introduced when using surface electromyography (EMG) for abdominal muscle monitoring. However, using filters indiscriminately can easily lead to problems. First, the frequency bands distributed by ECG signals largely overlap with the effective components of EMG signals, so using filters will weaken the amplitude of EMG signals, making measurements inaccurate. Second, when trying to remove signals with very narrow bandwidths, the selected filter order will be very high, resulting in a large amount of computation, making it impossible to process signals in real time and meet the real-time display requirements during product use.

[0006] In existing technologies, electrodes are attached to the lower abdomen or muscles far from the heart. However, when measuring upper abdominal muscles, electrocardiogram (ECG) noise is still introduced, failing to meet the requirements. Furthermore, using traditional filters to crudely remove ECG signals can easily affect the true components of the target electromyography (EMG) signal, thus impacting subsequent evaluation or model training. Moreover, using high-order filters requires higher computational power, which cannot meet the needs of real-time display. Summary of the Invention

[0007] Technical Objective: To address the shortcomings of existing technologies where electromyography (EMG) signals from the abdominal muscle surface are mixed with electrocardiogram (ECG) signals, thus causing interference, this invention discloses a method and system for removing ECG components from abdominal muscle surface EMG signals. Utilizing an N-order model, the computational requirements are significantly reduced. The method performs real-time processing on EMG signals from the abdominal muscles collected using surface electrodes, effectively filtering out ECG signals while preserving the EMG signal components of the abdominal muscles with high fidelity. This effectively removes ECG interference and facilitates the detection of EMG signals from the upper abdominal muscles.

[0008] Technical solution: To achieve the above technical objectives, the present invention adopts the following technical solution.

[0009] A method for removing the central electrical component of electromyographic signals from the surface of abdominal muscles, comprising the following steps:

[0010] S1. Obtaining the initial digital signal after preprocessing the electromyographic signal on the surface of the abdominal muscle: The analog signal collected by the electromyographic electrodes on the surface of the abdominal muscle group is obtained, and the initial digital signal is obtained after signal amplification and power frequency filtering.

[0011] S2. Extract templates from the initial digital signal to obtain the set of R-peak values ​​and their horizontal and vertical coordinates: Extract templates from the initial digital signal to obtain several differential feature functions, and make a differential threshold judgment based on all differential feature functions to obtain the set of R-peak values ​​and their horizontal and vertical coordinates.

[0012] S3. Construct a training sample set based on the set of peak values ​​and their x and y coordinates for R peaks;

[0013] S4. The simulated signals from the electromyography electrodes on the surface of the abdominal muscle group are collected again. After signal amplification and power frequency filtering preprocessing, the signal with the same length as the training sample is truncated as the signal to be processed. The training sample set of the signal to be processed is processed by the constructed N-order ECG filter to obtain the waveform of the abdominal muscle electromyography signal after removing the ECG signal, and it is displayed in real time.

[0014] Preferably, step S1 specifically includes:

[0015] The analog signal X(t) collected by the electromyography electrodes on the surface of the abdominal muscle group is acquired. The analog signal X(t) is converted into a digital signal x(n′). The digital signal x(n′) is amplified to obtain the amplified digital signal x(n). The amplified digital signal x(n) is then filtered out of 50Hz power frequency noise by a power frequency notch filter to obtain the digital signal after power frequency noise removal, which is used as the initial digital signal x0(n).

[0016] Preferably, step S2 specifically includes:

[0017] S21. For the initial digital signal x0(n), set the signal starting point n0 and the window length WinLength, extract the signal x0(n0:n0+WinLength) within the window length and record it as the variable x. w (n), for variable x w (n) Detect the peak value of the heartbeat R-peak. Extract the peak value of the heartbeat R-peak (i.e., the heartbeat wave peak) using the multi-point difference method. The calculation formula is as follows:

[0018] Y(n) = 0.25*x w (n0+6)+0.5*x w (n0+5)+x w (n0+4)

[0019] -x w (n0+3)-0.5*x w (n0+2)-0.25*x w (n0+1)

[0020] Where Y(n) is the difference characteristic function;

[0021] S22. Set the difference threshold W. If the signal point n of the difference characteristic function Y(n) is... k If a signal point n passes through zero, and the absolute value of the local extreme values ​​of the difference before and after zero is greater than the set difference threshold W, then the signal point n... k The calculated R-peak P is located at point P. k The location of this peak point is denoted as P. k =(n k ,x0(n k), where k represents the index of the k-th R-peak in the initial digital signal, x0(n k ) represents the signal value at which the R peak is located, {P k} represents the set of peak values ​​of R-peaks in all signals, both horizontally and vertically.

[0022] Preferably, step S3 specifically includes:

[0023] S31. Set the order of the ECG removal function to N, and take the first N coordinates from the set of R peak values ​​as the training coordinate set Q, Q = {(n1, x0(n1)), (n2, x0(n2)), ..., (n...} N ,x0(n N ))};

[0024] S32. Based on the set of peak values ​​and coordinates of R obtained in S2, construct a training set by taking the same length before and after each peak; take the coordinates of the peak location before... The coordinates of the peak value at each point. The training sample set is composed of several points, and this time sequence is denoted as:

[0025]

[0026] Among them, the training sample set module k There are N training samples in total, and k usually takes the values ​​1, 2, ..., N.

[0027] Preferably, step S4 specifically includes:

[0028] S41. Define the joint correlation coefficient calculation function Comcoefficient(X, Y), and its calculation formula is as follows:

[0029]

[0030] Where X and Y represent two training samples in the time series, and Y′ is the transpose of Y; and during the calculation, X and Y have the same dimension.

[0031] S42. Construct an N-order ECG filter, collect the analog signal of the electromyography electrode on the surface of the abdominal muscle group again, and after signal amplification and power frequency filtering preprocessing, extract the signal with the same length as the training sample as the signal to be processed module0. The signal to be processed module0 and the training sample set in S3 are processed by the constructed N-order ECG filter to obtain the waveform of the abdominal muscle electromyography signal after the ECG signal is filtered out from the signal to be processed.

[0032] S43. Downsample the waveform of the abdominal muscle electromyography signal after removing the electrocardiogram signal, and display the downsampled waveform in real time.

[0033] Preferably, the mathematical formula for the ECG filter is expressed as follows:

[0034]

[0035] Where N is the function order, which is the value after the last iteration. It is a one-dimensional time series of length before+after+1, which is the waveform of the signal to be processed after removing the electrocardiogram signal, that is, the waveform of the abdominal muscle electromyography signal after removing the electrocardiogram signal.

[0036] A system for removing the central electrical component of electromyographic signals on the surface of abdominal muscles, used to implement any of the methods described above for removing the central electrical component of electromyographic signals on the surface of abdominal muscles, includes an electromyographic acquisition module, a signal amplification module, a signal processing module, and a real-time display module connected in sequence;

[0037] The electromyography (EMG) acquisition module is used to acquire analog signals from EMG electrodes on the surface of the abdominal muscle group. The signal amplification module is used to amplify and preprocess the analog signals output by the EMG acquisition module to obtain initial digital signals. The signal processing module is used to process the digital signals output by the signal amplification module to obtain the waveform of the abdominal muscle EMG signal after filtering out the ECG signal. The real-time display module is used to display the waveform of the abdominal muscle EMG signal after filtering out the ECG signal output by the signal processing module in real time.

[0038] Beneficial effects: This invention utilizes an N-order electrocardiogram (ECG) filter, which greatly reduces the computational power requirement. It performs real-time processing of electromyographic (EMG) signals from abdominal muscles collected using surface electrodes. While preserving the EMG signal components of the abdominal muscles with high fidelity, it effectively filters out ECG signals, effectively removing interference from ECG signals and facilitating the detection of EMG signals on the surface of the upper abdominal muscles. It also enables real-time display of abdominal EMG signal detection and greatly preserves the effective components of the EMG signal, accurately reflecting the amplitude of the abdominal EMG signal. Attached Figure Description

[0039] Figure 1 This is a flowchart of the method of the present invention;

[0040] Figure 2 This is a schematic diagram of the system signal flow processing of the present invention;

[0041] Figure 3 This is a system structure framework diagram of the present invention;

[0042] Figure 4 The following are waveform examples of the method of the present invention for processing the electromyographic signals of the abdominal muscles before and after surgery. Detailed Implementation

[0043] The present invention will be further described and explained below with reference to the accompanying drawings.

[0044] As attached Figure 1 As shown, a method for removing the central electrical component of electromyographic signals on the surface of abdominal muscles includes the following steps:

[0045] S1. Obtaining initial digital signals after preprocessing of abdominal muscle surface electromyography signals: Analog signals collected by electromyography electrodes on the surface of the abdominal muscle group are acquired, and initial digital signals are obtained after signal amplification and power frequency filtering preprocessing; specifically including:

[0046] The analog signal X(t) collected by the electromyography electrodes on the surface of the abdominal muscle group is acquired. The analog signal X(t) is converted into a digital signal x(n′). The digital signal x(n′) is amplified to obtain an amplified digital signal x(n). The amplified digital signal x(n) is then filtered out of 50Hz power frequency noise by a power frequency notch filter to obtain a digital signal after power frequency noise removal, which is used as the initial digital signal x0(n).

[0047] S2. Extract templates from the initial digital signal to obtain the set of R-peak values ​​and their horizontal and vertical coordinates: Extract templates from the initial digital signal to obtain several differential feature functions, and make a differential threshold judgment based on all differential feature functions to obtain the set of R-peak values ​​and their horizontal and vertical coordinates.

[0048] Specifically, the following steps are included:

[0049] S21. For the initial digital signal x0(n), set the signal starting point n0 and the window length WinLength. Starting from the signal starting point, divide the initial digital signal according to the window length. The signal length within each window length is equal. The size of the window length is set based on experience and actual conditions. Extract the signal x0(n0: n0 + WinLength) within the window length and record it as the variable x. w (n), x w (n) = {x0(n0), x0(n0+1), ..., x0(n0+WinLength)}, for the variable x w (n) Detect the peak value of the heartbeat R-peak. Extract the peak value of the heartbeat R-peak (i.e., the heartbeat wave peak) using the multi-point difference method. The calculation formula is as follows:

[0050] Y(n) = 0.25*x w (n0+6)+0.5*x w (n0+5)+x w (n0+4)

[0051] -x w (n0+3)-0.5*x w (n0+2)-0.25*x w (n0+1)

[0052] Where Y(n) is the difference characteristic function;

[0053] S22. Set the difference threshold W. If the signal point n of the difference characteristic function Y(n) is... k If a signal point n passes through zero, and the absolute value of the local extreme values ​​of the difference before and after zero is greater than the set difference threshold W, then the signal point n... k The calculated R peak P is located at [location]. k The location of this peak point is denoted as P. k =( n k, x0(n) k ), where k represents the index of the k-th R-peak in the initial digital signal, x0(n k ) represents the signal value at which the R peak is located, {P k} represents the set of peak-to-peak values ​​of all signals, which is the set of peak-to-peak values ​​of all signals within each group of windows divided according to the window length.

[0054] S3. Construct a training sample set based on the set of peak values ​​and their x and y coordinates for R, specifically including:

[0055] S31. Set the order of the ECG removal function to N, and take the first N coordinates from the set of R peak values ​​as the training coordinate set Q, Q = {(n1, x0(n1)), (n2, x0(n2)), ..., (n...} N ,x0(n N ))};

[0056] S32. Based on the set of horizontal and vertical coordinates of the R peak obtained in S2, take the same length before and after each peak to form a training set. In some embodiments of the present invention, the coordinates of the peak location are taken from the previous... The coordinates of the peak point are as follows: The training sample set is composed of several points, and this time sequence is denoted as:

[0057]

[0058] Among them, the training sample set module k There are N training samples in total, and k usually takes the values ​​1, 2, ..., N.

[0059] S4. Acquire the analog signal from the electromyography electrodes on the surface of the abdominal muscle group again. After signal amplification and power frequency filtering preprocessing, extract the signal with the same length as the training sample as the signal to be processed (module0). The signal to be processed and the training sample set in S3 are processed by the constructed N-order ECG filter to obtain the waveform of the abdominal muscle electromyography signal after removing the ECG signal from the signal to be processed, and display it in real time. The specific steps include:

[0060] S41. Define the joint correlation coefficient calculation function Comcoefficient(X, Y), and its calculation formula is as follows:

[0061]

[0062] Where X and Y represent two training samples in the time series, and Y′ is the transpose of Y; and during the calculation, X and Y have the same dimension.

[0063] S42. Construct an N-order ECG filter, where N ranges from 3 to 8; re-acquire analog signals from the electromyographic electrodes on the surface of the abdominal muscle group, and after signal amplification and power frequency filtering preprocessing, extract the signal with the same length as the training sample as the signal to be processed, module0. The signal to be processed, module0, and the training sample set in S3 are processed by the constructed N-order ECG filter to obtain the waveform of the abdominal muscle electromyographic signal after removing the ECG signal from the signal to be processed; the mathematical formula for the N-order ECG filter is expressed as:

[0064]

[0065] Where N is the function order, after the last iteration. It is a one-dimensional time series of length before+after+1, which is the waveform of the signal to be processed after removing the electrocardiogram signal, that is, the waveform of the abdominal muscle electromyography signal after removing the electrocardiogram signal.

[0066] In some embodiments of the present invention, a sixth-order electrocardiogram (ECG) filter is employed. The mathematical formula for the sixth-order ECG filter is as follows:

[0067]

[0068] S43. Downsample the waveform of the abdominal muscle electromyography signal after removing the electrocardiogram signal, and display the downsampled waveform in real time.

[0069] As attached Figure 4 As shown, attached Figure 3 The above figure shows a waveform example of the abdominal muscle electromyography signal before and after processing by the method of the present invention. The upper figure is a schematic diagram of the waveform before processing, and the lower figure is a schematic diagram of the waveform after processing.

[0070] Compared to existing filtering techniques, this invention can more accurately and sensitively filter out ECG signals mixed in with the electromyography (EMG) signals on the abdominal muscle surface. Traditional filter methods often suffer from overlapping frequencies between ECG noise and the effective components of the EMG signal, leading to the removal of some effective EMG components and significantly impacting subsequent evaluation and analysis. While higher-order filters can effectively remove ECG interference without considering the reduction of effective signal components, this method involves extremely high computational demands, making real-time signal processing and display impossible. The N-order ECG filter used in this invention has a lower order, ranging from 3 to 8, meaning only N iterations are performed, resulting in less computation. Furthermore, the method described in this invention does not require additional training of the ECG filter. When collecting abdominal muscle EMG signals for each patient, a training sample set is directly constructed using the patient's historical data and input directly into the ECG filter along with the data to be processed to obtain the abdominal muscle EMG signal with ECG signals removed, further reducing computational load.

[0071] Compared to existing machine learning filtering techniques, this invention has low computational cost and very fast computation speed, and does not require a large sample set for training or pre-training. The rhythm and waveform of heartbeats vary from user to user, and there are significant differences between patients with heart disease and healthy individuals. Machine learning filtering methods can only be trained on large sample sizes without specificity or targeting, which can lead to poor filtering results for patients with heart disease. The pre-training of this invention requires a very small training set sample size, and only a few seconds of electromyography (EMG) signal data from the patient are needed to complete the pre-training. This allows for targeted filtering of the user's EMG signal, with good and stable filtering results.

[0072] As attached Figure 2 and attached Figure 3 As shown, a system for removing the central electrical component of electromyographic signals on the surface of abdominal muscles is used to implement the above-described method for removing the central electrical component of electromyographic signals on the surface of abdominal muscles. The system includes an electromyographic acquisition module, a signal amplification module, a signal processing module, and a real-time display module connected in sequence.

[0073] The electromyography (EMG) acquisition module is used to acquire analog signals from EMG electrodes on the surface of the abdominal muscle group. The signal amplification module is used to amplify and preprocess the analog signals output by the EMG acquisition module to obtain initial digital signals. The signal processing module is used to process the digital signals output by the signal amplification module to obtain the waveform of the abdominal muscle EMG signal after filtering out the ECG signal. The real-time display module is used to display the waveform of the abdominal muscle EMG signal after filtering out the ECG signal output by the signal processing module in real time.

[0074] This invention performs real-time processing of electromyography (EMG) signals from abdominal muscles collected using surface electrodes. While preserving the EMG signal components of the abdominal muscles with high fidelity, it effectively filters out electrocardiogram (ECG) signals, effectively removing ECG interference and facilitating the detection of EMG signals on the surface of the upper abdominal muscles. Utilizing an N-order model significantly reduces computational requirements, enabling real-time display of abdominal muscle EMG signal detection. It greatly preserves the effective components of the EMG signal, accurately reflecting the amplitude of the abdominal muscle EMG signal. Applications include medical device technology, EMG signal processing, vital sign monitoring, sports rehabilitation, and abdominal muscle detection.

[0075] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for removing the central electrical component of electromyographic signals from the surface of abdominal muscles, characterized in that, Includes the following steps: S1. Obtaining the initial digital signal after preprocessing the electromyographic signal on the surface of the abdominal muscle: The analog signal collected by the electromyographic electrodes on the surface of the abdominal muscle group is obtained, and the initial digital signal is obtained after signal amplification and power frequency filtering preprocessing. S2. Extract templates from the initial digital signal to obtain the set of R-peak values ​​and their horizontal and vertical coordinates: Extract templates from the initial digital signal to obtain several differential feature functions, and make a differential threshold judgment based on all differential feature functions to obtain the set of R-peak values ​​and their horizontal and vertical coordinates. S3. Construct a training sample set based on the set of peak values ​​and their x and y coordinates for R peaks; S4. The simulated signal of the electromyography electrode on the surface of the abdominal muscle group is collected again. After signal amplification and power frequency filtering preprocessing, the signal with the same length as the training sample is truncated as the signal to be processed. The signal to be processed and the training sample set are processed by the constructed N-order ECG filter to obtain the waveform of the abdominal muscle electromyography signal after removing the ECG signal, and displayed in real time. Step S4 specifically includes: S41. Define the function for calculating the joint correlation coefficient. The calculation formula is as follows: ， Where X and Y represent two training samples in the time series, X is the transpose of Y; and during the calculation, X and Y have the same dimensions. S42. Construct an N-order ECG filter, and again acquire simulated signals from the electromyographic electrodes on the surface of the abdominal muscle group. After signal amplification and power frequency filtering preprocessing, extract the signal with the same length as the training sample as the signal to be processed. Signal to be processed After the training sample set in S3 is processed by the constructed N-order ECG filter, the waveform of the abdominal muscle electromyography signal after filtering out the ECG signal is obtained. S43. Downsample the waveform of the abdominal muscle electromyography signal after removing the electrocardiogram signal, and display the downsampled waveform in real time. The mathematical formula for the Nth-order ECG filter is expressed as follows: ， Where N is the function order, obtained after the last iteration. It is a one-dimensional time series with the same length as the training sample, which is the waveform of the signal to be processed after removing the electrocardiogram signal, that is, the waveform of the abdominal muscle electromyography signal after removing the electrocardiogram signal.

2. The method for removing the central electrical component of electromyographic signals on the surface of abdominal muscles according to claim 1, characterized in that: Step S1 specifically includes: Analog signals acquired by electromyography electrodes on the surface of the abdominal muscle group The analog signal Convert to digital signal digital signals The signal is amplified to obtain the amplified digital signal. Amplified digital signal The 50Hz power frequency noise is then filtered out using a power frequency notch filter to obtain the digital signal after power frequency noise removal, which is used as the initial digital signal. .

3. The method for removing the central electrical component of electromyographic signals on the surface of abdominal muscles according to claim 1, characterized in that: Step S2 specifically includes: S21. Regarding the initial digital signal Set signal start point and window length Extract the signal within the window length And denote it as a variable. , for variables The peak value of the heartbeat R-wave was detected, and the peak value was extracted using a multi-point difference method. The calculation formula is as follows: ， in, It is the difference characteristic function; S22. Set the difference threshold W, if the difference characteristic function signal point If the signal point passes through zero, and the absolute value of the local extreme values ​​of the difference before and after zero is greater than the set difference threshold W, then the signal point... The calculated R peak is located at [location]. Location, peak point is denoted as Where k represents the index of the k-th R-peak in the initial digital signal. This indicates the signal value at which the R peak is located. This is the set of horizontal and vertical coordinates of the peak values ​​of the R peaks in all signals.

4. The method for removing the central electrical component of electromyographic signals on the surface of abdominal muscles according to claim 1, characterized in that: Step S3 specifically includes: S31. Set the order of the ECG removal function to N, and take the first N coordinates from the set of R peak values ​​as the training coordinate set Q. ; S32. Based on the set of peak values ​​and coordinates of R obtained in S2, construct a training set by taking the same length before and after each peak; take the first part of the coordinates of the peak location. The latter part of the coordinates of the peak location The training sample set is composed of this time series, denoted as: ， Among them, the training sample set There are N training samples in total, and k takes a value of N. .

5. A system for removing the central electrical component of electromyographic signals on the surface of abdominal muscles, used to implement the method for removing the central electrical component of electromyographic signals on the surface of abdominal muscles as described in any one of claims 1-4, characterized in that: It includes an electromyography (EMG) acquisition module, a signal amplification module, a signal processing module, and a real-time display module connected in sequence; The electromyography (EMG) acquisition module is used to acquire analog signals from EMG electrodes on the surface of the abdominal muscle group. The signal amplification module is used to amplify and preprocess the analog signals output by the EMG acquisition module to obtain initial digital signals. The signal processing module is used to process the digital signals output by the signal amplification module to obtain the waveform of the abdominal muscle EMG signal after filtering out the ECG signal. The real-time display module is used to display the waveform of the abdominal muscle EMG signal after filtering out the ECG signal output by the signal processing module in real time.

Images