VMD-based denoising methods for surface electromyography signals

Research on the recognition model of exercise fatigue based on the fusion of sEMG and ECG signals

This study significantly enhances the accuracy of exercise state identification in wearable devices through improved denoising techniques for sEMG and ECG signals. By adopting an optimized Variational Mode Decomposition (VMD) method, combined with the Improved Sparrow Search Algorithm and Second Generation Wavelet Transform (ISSA-VMD-SWT), and introducing chaos mapping to strengthen the algorithm's initial population, this approach effectively reduces noise while preserving key fatigue-related features. In tests conducted on data from 32 participants, the method achieved accuracy rates of 93.25%, 95.16%, and 93.05% for identifying "Easy," "Transition," and "Tired" exercise states, respectively, showing significant advantages over traditional denoising techniques. These results indicate that the denoising technology developed in this study represents a significant technological advancement for the application of ECG and sEMG fatigue identification technologies in wearable health monitoring devices.

Reducing Noise, Artifacts and Interference in Single-Channel EMG Signals: A Review

Electromyography (EMG) is gaining importance in many research and clinical applications, including muscle fatigue detection, control of robotic mechanisms and prostheses, clinical diagnosis of neuromuscular diseases and quantification of force. However, EMG signals can be contaminated by various types of noise, interference and artifacts, leading to potential data misinterpretation. Even assuming best practices, the acquired signal may still contain contaminants. The aim of this paper is to review methods employed to reduce the contamination of single channel EMG signals. Specifically, we focus on methods which enable a full reconstruction of the EMG signal without loss of information. This includes subtraction methods used in the time domain, denoising methods performed after the signal decomposition and hybrid approaches that combine multiple methods. Finally, this paper provides a discussion on the suitability of the individual methods based on the type of contaminant(s) present in the signal and the specific requirements of the application.

Deep Convolutional Generative Adversarial Network with LSTM for ECG Denoising

The electrocardiogram (ECG), as an essential basis for the diagnosis of cardiovascular diseases, is usually disturbed by various noise. To obtain accurate human physiological information from ECG, the denoising and reconstruction of ECG are critical. In this paper, we proposed an ECG denoising method referred to as LSTM-DCGAN which is based on an improved generative adversarial network (GAN). The overall network structure is composed of multiple layers of convolutional networks. Furthermore, the convolutional features can be connected to their time series order dependence by adding LSTM layers after each convolutional layer. To verify the effectiveness and the denoising performance of the improved network structure, we test the proposed algorithm on the famous MIT-BIH Arrhythmia Database with different levels of noise from the MIT-BIH Noise Stress Test Database. Experimental results show that our method can remove the single noise and the mixed noise while retaining the complete ECG information. For the mixed noise removal, the average SNR_imp, RMSE, and PRD are 19.254 dB, 0.028, and 10.350, respectively. Compared with the state-of-the-art methods, DCGAN, and the LSTM-GAN methods, our method obtains the higher SNR_imp and the lower RMSE and PRD scores. These results suggest that the proposed LSTM-DCGAN approach has a significant advantage for ECG processing and application in complex scenes.

Real-time modeling and feature extraction method of surface electromyography signal for hand movement classification based on oscillatory theory

OBJECTIVE

Research of surface electromyography (sEMG) signal modeling and feature extraction is important in human motion intention recognition, prosthesis and exoskeleton robot. However, the existing methods mostly use the signal segmentation processing way rather than point-to-point signal processing way and lack physiological mechanism support.

APPROACH

In this study, a real-time sEMG signal modeling and separation method is developed based on oscillatory theory. On this basis, a sEMG signal feature extraction method is constructed and ensemble learning method is combined to achieve the real-time human hand motion intention recognition.

MAIN RESULTS

The experimental results show that the average root mean square difference (RMSD) value of sEMG signal modeling is 0.3838±0.0591, and the average accuracy of human hand motion intention recognition is 96.03±1.74%. On a computer with an Intel (R) Core (TM) i5-8250U CPU running Matlab 2016Rb, the execution time for the sEMG signal with an actual duration of 2 second is 0.66 second.

SIGNIFICANCE

Compared with several existing methods, the proposed method has better modeling accuracy, motion intention recognition accuracy and real-time performance. The method developed in this study may provide a new perspective of sEMG modeling and feature extraction for hand movement classification.

Degradation Trend Prediction of Pumped Storage Unit Based on MIC-LGBM and VMD-GRU Combined Model

The harsh operating environment aggravates the degradation of pumped storage units (PSUs). Degradation trend prediction (DTP) provides important support for the condition-based maintenance of PSUs. However, the complexity of the performance degradation index (PDI) sequence poses a severe challenge of the reliability of DTP. Additionally, the accuracy of healthy model is often ignored, resulting in an unconvincing PDI. To solve these problems, a combined DTP model that integrates the maximal information coefficient (MIC), light gradient boosting machine (LGBM), variational mode decomposition (VMD) and gated recurrent unit (GRU) is proposed. Firstly, MIC-LGBM is utilized to generate a high-precision healthy model. MIC is applied to select the working parameters with the most relevance, then the LGBM is utilized to construct the healthy model. Afterwards, a performance degradation index (PDI) is generated based on the LGBM healthy model and monitoring data. Finally, the VMD-GRU prediction model is designed to achieve precise DTP under the complex PDI sequence. The proposed model is verified by applying it to a PSU located in Zhejiang province, China. The results reveal that the proposed model achieves the highest precision healthy model and the best prediction performance compared with other comparative models. The absolute average (|AVG|) and standard deviation (STD) of fitting errors are reduced to 0.0275 and 0.9245, and the RMSE, MAE, and R2 are 0.00395, 0.0032, and 0.9226 respectively, on average for two operating conditions.

EMG Signal Filtering Based on Variational Mode Decomposition and Sub-Band Thresholding

Surface electromyography (EMG) signals are inevitably contaminated by various noise components, including powerline interference (PLI), baseline wandering (BW), and white Gaussian noise (WGN). These noises directly degrade the efficiency of EMG processing and affect the accuracy and robustness of further applications. Currently, most of the EMG filters only target one category of noise. Here, we propose a novel filter to remove all three types of noise. The noisy EMG signal is first decomposed into an ensemble of band-limited modes using variational mode decomposition (VMD). Each category of noise is located within specific modes and is separately removed in sub-bands. In particular, WGN is suppressed by soft thresholding with a noise level-dependent threshold. The denoising performance was assessed from simulated and experimental signals using three performance metrics: the root mean square error ([Formula: see text]), the improvement in signal-to-noise ratio ([Formula: see text]), and the percentage reduction in the correlation coefficient ( η). Other methods, including traditional infinite impulse response (IIR) filters, empirical mode decomposition (EMD) method, and ensemble empirical mode decomposition (EEMD) method, were examined for comparison. The proposed method achieved the best performance to remove BW or WGN. It also effectively reduced PLI noise when the signal-to-noise ratio (SNR) was low. The SNR was improved by 18.6, 19.2, and 8.0 dB for EMG signals corrupted with PLI, BW, and WGN at -6 dB SNR, respectively. The experimental results illustrated that noise was completely removed from resting states, and obvious spikes were distinguished from action states. For two of the ten subjects, the improved SNR reached 20 dB. This study explores the special characteristics of VMD and demonstrates the feasibility of using the VMD-based filter to denoise EMG signals. The proposed filter is efficient at removing three categories of noise and can be used for any application that requires EMG signal filtering at the preprocessing stage, such as gesture recognition and EMG decomposition.