Classifying Hand Movement Intentions Using Surface EMG signals and SVM

This article presents a complete system for classifying hand movement intentions by means of Support Vector Machines (SVM). The proposed approach is able to classify up to 52 intentions of hand movement through electromyographic signals, which are acquired by surface electrodes placed on the forearm of a subject. Experimental results have shown that the proposed approach can identify movement intentions even in subjects with transradial amputation with an accuracy above 90%.