A novel method of detection and classification of motor nerve late-wave activity

In motor nerve conduction studies, important diagnostic information is provided by the late-wave responses, comprised of F-waves, A-waves, and repeaters. Late-waves in addition to contamination from power line interference and baseline disturbance, are of low amplitude and random in nature. This makes computer-based analysis of late-wave activity very challenging, especially the computer-based F-wave onset latency assignment. Currently available algorithms assign latency independently on a trace-by-trace basis without considering the information present in an entire ensemble of traces. A novel algorithm that takes into account the ensemble information for segmenting and classifying regions of late-wave data is described in this paper, which in turn can be used to improve the performance of computer-based F-wave onset latency assignment.