Myo electric classification using twin SVM and blind source separation techniques

Myo electrical activities also known as Surface electromyogram (sEMG) is a measure of the muscle activity from the skin surface, and is an excellent indicator of the strength of muscle contraction. It is an obvious choice for control of prostheses, and identification of body gestures. Using sEMG to identify posture and actions is rendered difficult by interference between different muscle activities making it a multi class classification problem. Multi-category classification problems are usually solved by solving many, one-versus-rest binary classification tasks. These sub-tasks naturally involve unbalanced data sets. Therefore, we require a learning methodology that can take into account unbalanced data sets, as well as large variations in the distributions of patterns corresponding to different classes. This paper reports the use of Twin Support Vector Machine for gesture classification based on sEMG, and shows that this technique is eminently suited to such applications.