EMG Pattern Analysis and Classification for a Prosthetic Arm

Abstract

This paper deals with the statistical analysis and pattern classification of electromyographic signals from the biceps and triceps of a below-the-humerus amputated or paralyzed person. Such signals collected from a simulated amputee are synergistically generated to produce discrete lower arm movements. The purpose of this study is to utilize these signals to control an electrically driven prosthetic or orthotic arm with minimum extra mental effort on the part of the subject. The results show very good separability of classes of movements when a learning pattern classification scheme is used, and a superposition principle seems to hold which may provide a means of decomposition of any composite motion to the six basic primitive motions, e.g., humeral rotation in and out, elbow flexion and extension, and wrist pronation and supination. Since no synergy was detected for the hand movements, different inputs have to be provided for a grip. The method described is not limited by the location of the electrodes. For amputees with shorter stumps, synergistic signals could be obtained from the shoulder muscles. However, the presentation in this paper is limited to bicep-tricep signal classification only.