A new strategy based on feature filtering technique for improving the real-time control performance of myoelectric prostheses

Surface electromyogram pattern recognition (EMG-PR) has been considered as a promising approach for predicting amputees’ motion intentions to control myoelectric prostheses. However, EMG recordings are mostly contaminated by various interferences, which decay the prediction of EMG-PR methods, thus affecting performance prosthesis. One common way solve this issue is improve quality signals via filtering methods. Unfortunately, complete attenuation effects caused remains major challenge till date. In study, besides signal filtering, an additional conditioning strategy, designated feature was proposed and applied features that were extracted from overall EMG-PR. The strategy’s investigated in both offline analysis real-time prosthesis test able bodied subjects amputees. experimental results showed that, using classification accuracy increased 6.4% 7.4% amputees, respectively, completion rate 12.5% 14.6%. Furthermore, strategy compared with commonly used post-processing known majority vote. It observed achieved significantly better prosthetic control. Findings study suggest could potentially enhance robustness reliability method prostheses other human–machine interaction systems.