Stimulus artifact removal using a software-based two-stage peak detection algorithm

The analysis of stimulus evoked neuromuscular potentials or m-waves is a us eful technique for improved feedback control in functional electrical stimu lation systems. Usually, however, these signals are contaminated by stimulu s artifact. A novel software technique, which uses a two-stage peak detecti on algorithm, has been developed to remove the unwanted artifact from the r ecorded signal. The advantage of the technique is that it can be used on al l stimulation artifact-contaminated electroneurophysiologic data provided t hat the artifact and the biopotential are non-overlapping. The technique do es not require any estimation of the stimulus artifact shape or duration. W ith the developed technique, it is not necessary to record a pure artifact signal for template estimation, a process that can increase the complexity of experimentation. The technique also does not require the recording of an y external hardware synchronisation pulses. The method avoids the use of an alogue or digital filtering techniques, which endeavour to remove certain h igh frequency components of the artifact signal, but invariably have diffic ulty, resulting in the removal of frequencies in the same spectrum as the m -wave. With the new technique the signal is sampled at a high frequency to ensure optimum fidelity. Instrumentation saturation effects due to the arti fact can be avoided with careful electrode placement. The technique was ful ly tested with a wide variety of electrical stimulation parameters (frequen cy and pulse width) applied to the common peroneal nerve to elicit contract ion in the tibialis anterior. The program was also developed to allow batch processing of multiple files, using closed loop feedback correction. The t wo-stage peak detection artifact removal algorithm is demonstrated as an ef ficient post-processing technique for acquiring artifact free m-waves. (C) 2001 Elsevier Science BN. All rights reserved.