Neurometrix Inc., Waltham, MA 02451, USA. brian_tracey@neurometrix.com
An algorithm for automated removal of stimulus artifact has been developed and tested on nerve conduction study data. The algorithm uses a hardware-based model of the stimulus artifact (SA). Model parameters are estimated from portions of the data that are judged to contain only the artifact. The model can be used to remove SA even when it is temporally overlapped with the nerve signal. Data are shown to demonstrate the algorithm's performance and to quantify the effect of SA removal on clinical parameters.