VMD-based denoising methods for surface electromyography signals.

Area of Science:

• Biomedical Engineering
• Signal Processing

Background:

• Surface electromyographic (sEMG) signals are prone to noise during measurement.
• Existing denoising methods may not be optimal for sEMG signals.

Purpose of the Study:

• To propose and evaluate two novel denoising methods for sEMG signals based on Variational Mode Decomposition (VMD).
• To compare the performance of the proposed methods against established techniques like Empirical Mode Decomposition (EMD) and Wavelet methods.

Main Methods:

• Developed VMD-WST: Decomposes sEMG into variational mode functions (VMFs), then applies Wavelet Soft Thresholding (WST) to each VMF.
• Developed VMD-SIT: Decomposes sEMG into VMFs, then applies Soft Interval Thresholding (SIT) to each VMF.
• Evaluated methods using Signal-to-Noise Ratio (SNR), Root Mean Square Error, and R-squared value on sEMG data from 10 healthy subjects and 10 stroke patients.

Main Results:

• Both VMD-WST and VMD-SIT effectively filtered noise from sEMG signals.
• The proposed VMD methods outperformed EMD and Wavelet methods in denoising.
• VMD-SIT exhibited the best overall denoising performance.

Conclusions:

• VMD-based methods offer a promising approach for sEMG signal denoising.
• The VMD-SIT method is particularly effective and recommended for sEMG noise reduction.
• These techniques have potential applications in limb movement classification, disease diagnosis, and human-machine interaction.