Estimation of noise level and signal to noise ratio of laplacian electrocardiogram during ventricular depolarization and repolarization

Body surface Laplacian ECG (LECG) has demonstrated its enhanced capability to localize cardiac electrical sources closest to the recording site. The aim of the present study was to evaluate the noise level and signal to noise ratio (SNR) in the LECG as compared to the potential ECG (PECG). Such evaluation is important to determine the applicability of the LECG to localizing and imaging of cardiac electrical activity in an experimental setting. Experimental studies were conducted in six healthy men. A 150-channel PECG was recorded from the anterolateral chest and the LECG was estimated using the finite difference algorithm. The noise level in the PECG and LECG was evaluated using multiple estimation protocols. The signal level during ventricular depolarization and repolarization was also estimated, and the corresponding SNR was calculated. Different filtering techniques were examined to evaluate their effects on the noise level and SNR of the LECG and PECG. The experimental results indicate that with basic signal processing techniques (baseline adjustment, three-point moving average filter, and Wiener spatial filter), the SNR of the LECG is about 30-40% of that of the PECG. Furthermore, the SNR estimated during ventricular depolarization is about three times that obtained during ventricular repolarization for the PECG and LECG. The present study indicates that the LECG derived from the PECG using a local finite difference estimation procedure has satisfactory SNR during the periods of ventricular depolarization and repolarization, and suggests the feasibility of estimating the LECG from the recorded PECG in human subjects in an experimental setting.