Suppression of Baseline Wander Artifact in Magnetocardiogram Using Breathing Sensor

Measurements of magnetocardiogram (MCG) are commonly contaminated by different types of artifacts like breathing, power line interference and subject’s movement, out of which artifact due to breathing is unavoidable. This breathing artifact is mainly caused by movement of the chest of the subject during respiration and the consequent changes of the distance between the superconducting quantum interference device (SQUID) sensors (used for recording the MCG signal) and the subject (lying in a supine position under the cryostat housing the SQUID sensors); because of this, the measured MCG signal is a superposition of the true MCG signal of cardiac origin and the breathing related artifact. In most previous works, the suppression of baseline wander artifact was implemented by using digital filters, wavelet based technique, empirical mode decomposition (EMD) and ensemble empirical mode decomposition (EEMD). We present here a novel approach for effective suppression of breathing related baseline wander artifacts by acquiring the MCG data contaminated by the breathing artifact simultaneously with the experimental recording of the actual breathing pattern of the subject using a temperature sensor. The temperature sensor monitors the very slight change in temperature that occurs near the nostril during inhalation and exhalation, and this data is used for the subsequent elimination of the breathing artifact from the contaminated MCG by using the regression approach. We illustrate the method and compare its performance with other conventional approaches for the suppression of baseline wander in the measurement of MCG using single channel data. This is the first time that a breathing sensor has been employed for the suppression of breathing related baseline wander artifact. Experimental results reveal the superiority of the proposed approach in suppression of the breathing related base-line wander artifact with the important advantage that the algorithm used for the suppression uses the actual breathing data derived from an experimental input.