A Capacitive Electrocardiography System With Dedicated Noise-Cancellation Algorithms for Morphological Analysis

Objective:
Capacitive electrocardiography (cECG) has been proposed for ambulatory cardiac health monitoring. However, due to the high sensitivity of capacitive electrodes to various noise sources, particularly the power-line interference (PLI) and motion artifacts (MA), the existing capacitive systems are only verified in terms of RR interval. The aim of the present study is to explore the feasibility of using cECG for morphological analysis, and thus to extract clinical meaningful parameters.
Methods:
A capacitive electrode with active guarding is realized. A phase-locked-loop (PLL) based adaptive canceller is employed to remove PLI from the cECG. Wavelet analysis is adopted to cancel other noises. The developed capacitive system and algorithms are evaluated by real ECG measurements on 7 volunteers using 3-lead configuration. The correlation coefficient (CC) between the processed cECG and the wet ECG is calculated in two different conditions: with and without the QRS complex. Several frequently used diagnostic parameters, i.e., RR interval, QRS interval, P segment, T segment, ST segment, are extracted and compared with that obtained from the wet ECG.
Results:
High CCs are observed between the cECG and the wet ECG in both conditions, i.e.,
$0.97\pm 0.03$
with the QRS complex and
$0.92\pm 0.07$
without the QRS complex. Besides, RR interval extracted from the two different ECG signals are identical for each subject. Other diagnostic parameters are quite similar.
Significance:
Our results suggest cECG to be reliable for ambulatory heart rate monitoring. The results also indicate the feasibility of using cECG for clinical diagnosis.