Abstract:
The surface myoelectric signal (MES) has proven to be an effective control input for powered prostheses. Pattern recognition based controllers use multi-channel surface MES as inputs to discriminate between the desired classes of limb activation. There are two major methods which may be pursued to increase the accuracy of the controller: 1) use signal processing to extract more information from the input signals; or 2) provide more informative raw signals to the controller. As a result of recent technological advances, it is reasonable to assume that there will soon be implantable myoelectric sensors which will enable the internal MES to be used as inputs to controllers [1]. An internal MES measurement should have less muscular crosstalk allowing for more independent control sites. However, it remains unclear if this benefit outweighs the loss of the more global information contained in the surface MES. This study compares the classification accuracy of several pattern based myoelectric controllers which use information extracted from surface MES to the same controllers which use information extracted from fine-wire intramuscular MES.
