Resumen de Robust Signal Processing in Cardiac Signals: Applications in Heart Rate Variability, Heart Rate Turbulence and Fibrillatory Arrhythmias
The main objective of this doctoral Thesis, in the field of biomedical signal processing, is to develop robust methods for cardiac signal analysis. It has two specific objectives, namely, (1) to characterize atrial and ventricular fibrillation (AF, VF), and (2) to assess, noninvasively, the baroreflex and the Autonomic Nervous Systems (ANS) control of the heart rate. Spectral analysis of electrograms (EGM) has been used to characterize the average cycle and regularity of VF and AF. However, this approach discards relevant information of the spectrum. A parametric method based on an extension of Fourier Series is presented. Also, a method to estimate the AF fundamental frequency based on Correntropy, is proposed. Spectral analysis of beat-to-beat time series has been widely used to assess the ANS control of the heart rate using the Heart Rate Variability (HRV). However, this approach is very sensitive to the presence of noise and false beat detections. A robust method is proposed to interpolate HRV based on Support Vector Machine (SVM) regression. Heart rate turbulence (HRT) has been shown to be a strong risk stratification criterion. A method to denoise individual VPCs using SVM regression is proposed. Additionally, a nonlinear regression model is proposed to assess the influence of coupling interval and heart rate on HRT.
