A Novel Approach to Monitor Nonstationary Dynamics in Physiological Signals: Application to Blood Pressure, Pulse Oximeter, and Respiratory Data

摘要

We propose a parametric time-varying (TV) algorithm which utilizes sinusoids as the basis functions which are then projected onto sets of Legendre and Walsh functions for the purpose of monitoring nonstationary dynamics. The proposed algorithm is a general-purpose algorithm that has the potential to be widely applicable to various physiological signals, but is especially well-suited for tracking blood pressure (BP), pulse oximeter, and respiratory signals, as they all exhibit periodic oscillations with TV dynamics. The proposed algorithm’s efficacy was verified using both simulation examples and application to experimental data from all of the above-mentioned sources. Our results show that the method can: (1) accurately monitor abrupt frequency changes even when the data are contaminated with significant noise, (2) accurately monitor the BP and pulse oximeter signals, and (3) provide accurate estimation of respiratory rates derived directly from pulse oximeter recordings.