Comparison of pulse rate variability indices estimated from pressure signal and photoplethysmogram

Abstract

In this research, we proposed a method to estimate the time- and frequency-domain indices of pulse rate variability by using pulse waveforms separated from pressure signals measured during sleep. The pulse waveform is extracted from pressure signals based on wavelet transform, and heart beats are detected by an adaptive pulse template method. Through an evaluation based on 13 healthy subjects' pressure and photoplethysmography signals measured simultaneously during sleep, we found that the time- and frequency-domain indices of pulse rate variability estimated from pressure signals have a high correlation with those estimated from photoplethysmography signals. It suggests that pressure signals may serve as a convenient and low-cost tool to estimate pulse rate variability during sleep.