Portable sEMG-based Multi-Channel Hand Contralateral Controlled Functional Electrical Stimulation System

Contralateral functional electrical stimulation (CCFES) has been clinically proven to be effective in improving hand motor control and dexterity. This paper introduces a portable CCFES system based on Surface electromyography (sEMG) for active rehabilitation training of multiple movements. Applying the classifier based on Support Vector Machine (SVM) for the categorization of four distinct gestures, achieving an average F1-score of 0.928 during offline assessments. By employing a voting mechanism for real-time classification, the system achieved an outstanding average F1-score of 0.935 exceeding 0.9, demonstrating exceptional performance. In addition, the system detected sEMG in the affected hand in real time to characterize its muscle activity status, which in turn enabled adaptive amplitude modulation of the FES to reduce unnecessary stimulation. The experimental results showed that the system could enable the subjects to realize the hand rehabilitation training with different FES intensity, which showed the potential of active stroke rehabilitation.