Estimation of the Neural Drive to Forearm Muscles in Periodic Movements of Fingers

This study explored the feasibility of estimating the neural drive to forearm muscles during finger movements by decomposition of high-density surface electromyography (HD sEMG). Five healthy adults performed eight periodic movements of flexion and extension of fingers at two different frequencies. The Gradient Convolution Kernel Compensation (gCKC) algorithm was employed to decompose HD-sEMG signals into motor unit (MU) spike trains. The findings revealed a high correlation between the estimated neural drive and reference signal input, suggesting that HD sEMG decomposition is a promising approach for the proportional and simultaneous control of myoelectric prostheses. These results support the idea that this technique is an effective control strategy, paving the way for developing real-time human-machine interfaces.