Multivariate And Multidimensional Signal Analysis Of Cardiovascular Time Series

During the past two decades, a great progress has been made in the field of cardiovascular physiological signal analysis, and the indices of heart rate variability (HRV) have been well recognized as a "window" onto autonomic modulation of the heart. However, to complete traditional analysis, new methods derived from classical signal processing or non-linear dynamics have recently been developed. Therefore, it is more appropriate to use a panel of multivariate, multi-domain, dynamic and nonlinear dynamics methods to obtain the nonduplicative information about the autonomic regulation under both physiological and pathological conditions. And one may expect that a multivariate and multidimensional analysis will also enhance the clinical applications of HRV and BPV (blood pressure variability) analysis.Based on previous work in this laboratory, the aim of the present study was: 1) to improve pre-processing methods of HRV/SBPV; 2) to establish a multiple analysis panel for cardiovascular time series, including multivariate parametric system identification, time-frequency analysis and nonlinear dynamics; 3) to elucidate the cardiovascular integrative regulation mechanism under different orthostatic stresses and the effects of aging on it; 4) to compare the dynamic spectra of both HRV and SBPV during each of three kinds of stress, i.e., the...