A comparison of adaptive filter and artificial neural network results in removing electrocardiogram contamination from surface EMGs

Abstract

Surface electromyograms (EMGs) are valuable in the pathophysiological study and clinical treatment. These recordings are critically often contaminated by cardiac artifact. The purpose of this article was to evaluate the performance of an adaptive filter and artificial neural network (ANN) in removing electrocardiogram (ECG) contamination from surface EMGs recorded from the pectoralismajor muscles. Performance of these methods was quantified by power spectral density, coherence, signal to noise ratio, relative error and cross correlation in simulated noisy EMG signals. In between these two methods the ANN has better results.