Wavelet Based Denoising for Suppression of Motion Artifacts in Impedance Cardiography

Introduction Impedance cardiography is a noninvasive technique for monitoring the changes in the electrical impedance of the thorax z(t), caused by variation in the blood volume during the cardiac cycle [1]. Time derivative of the thoracic impedance is known as the impedance cardiogram (ICG) and it is used for estimating the ventricular ejection time (Tlvet), the negative peak of ICG ((-dz/dt)max), the stroke volume, and some other cardiovascular indices. Respiratory and motion artifacts cause baseline drift in the sensed impedance waveform, particularly during or after exercise, and this drift results in errors in the estimation of the parameters [1], [2]. Ensemble averaging [2], generally employed for suppressing the artifacts, suppresses the beat-to-beat variations and tends to smear the peak in the ICG. It may blur or suppress the less distinctive characteristic points in the waveform and hence may result in error in their detection. Due to a partial overlap between the spectra of ICG and the artifacts, non-adaptive digital filters are not effective in removing the artifacts. Adaptive filtering may be used for canceling the respiratory artifacts [3], but it is generally difficult to sense the references related to the sources of various motion artifacts and to combine them. Earlier we have investigated a wavelet based denoising technique for suppression of the respiratory artifact from ICG signal [4], in order to facilitate estimation of stroke volume on beat-to-beat basis. In present study, we examine the applicability of the denoising technique for cancellation of the motion artifact from the ICG signals, without smearing the beat-to-beat variations.