Classification of grasp modes based on electromyographic patterns of preshaping motions

The specificity of forearm surface EMG patterns during four grasps modes: cylindrical, spherical, pinch and key, were studied for hand preshaping motions. Surface EMG signals from four extensor muscles were squared and filtered to obtain smooth temporal envelopes of the EMG bursts. The maximum amplitudes of the envelopes were processed using principal component analysis to obtain a two-dimensional representation of the four clusters associated with the four grasp modes. The Mahalanobis distance function was used to classify the grasp modes. Initial results suggest that the classification of four grasp modes can be accomplished with a success rate above 90%