StimEMG: An Electromyogram Recording System With Real-Time Removal of Time-Varying Electrical Stimulation Artifacts

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Signal Processing

Background:

• Closed-loop Functional Electrical Stimulation (FES) systems aid in motor function recovery for paralytic patients.
• Real-time FES parameter adjustments introduce time-varying stimulation artifacts into electromyogram (EMG) signals.
• Artifacts challenge accurate monitoring of muscle contraction status, hindering FES system effectiveness.

Purpose of the Study:

• To develop an EMG acquisition system, StimEMG, capable of mitigating time-varying stimulation artifacts.
• To enhance the accuracy of EMG signal monitoring in closed-loop FES applications.
• To compare the performance of the novel SAG-RLS strategy against existing methods.

Main Methods:

• Development of the StimEMG system embedding a stimulation artifact generation (SAG) circuit and a Recursive Least Squares (RLS) adaptive filter.
• Testing the SAG-RLS strategy with simulated EMG signals contaminated by artifacts.
• Experimental validation of the StimEMG system with 8 human subjects.

Main Results:

• The SAG-RLS method demonstrated significantly higher correlation between denoised and clean EMG signals compared to the Gram-Schmidt-based method (simulation: 0.98 vs. 0.65; experiment: 0.99 vs. 0.52).
• SAG-RLS achieved superior artifact suppression, resulting in higher signal-to-noise ratios (simulation: 12.83 vs. 1.54) and noise rejection ratios (experiment: 2.32 vs. 1.92).
• The system's performance improvement is attributed to the SAG unit's ability to track artifact variations.

Conclusions:

• The developed StimEMG system with the SAG-RLS strategy effectively removes time-varying stimulation artifacts from EMG signals.
• This robust EMG acquisition module enhances the reliability of closed-loop FES systems for patient rehabilitation.
• The SAG-RLS approach offers a significant advancement over previous artifact removal techniques.