Detection and classification of motor nerve late-wave activity

The health of peripheral nerves is commonly assessed using nerve conduction studies, in which the nerve is stimulated electrically and the evoked responses are analyzed. In motor nerve studies, important diagnostic information is provided by the late-wave responses, comprised of F-waves, A-waves and repeaters. Computer-based analysis of late-wave activity is challenging due to the random nature and low amplitude of these signals. This paper describes an algorithm for segmenting portions of late-wave activity and classifying the type of evoked response. The algorithm may be used to improve the robustness of computer-based late-wave assessment.