Analysis of the heart signal dynamics to identify pathological cardiac patients with sudden cardiac death risk

Nowadays, one of the main reasons of people’s death in developed countries is due to cardiac causes. The analysis of cardiac electrical series by signal processing using different linear and non-linear techniques could give predictors or markers of sudden cardiac death risk stratification in cardiac pathologies. One of the most analyzed time series is the heart rate variability (HRV). HRV refers to the variations of both instantaneous heart rate and RR intervals (i.e. beat-to-beat intervals) controlled by the Autonomic Nervous System (ANS). In this research project, a set of subjects with different cardiac pathologies has been analyzed using conventional time-domain and frequency-domain linear techniques as well as a nonlinear time-domain technique. The phase rectified signal averaging (PRSA), a non-linear signal processing technique, was applied to assess the risk of suffering sudden cardiac death. Also, subjects classified into different groups of risk (normal or reduced left ventricular ejection fraction, NYHAII or NYHAIII classification and presence or absence of ischemic pathology) were analyzed. Different measures were defined from PRSA, acceleration and deceleration based capacities of the HRV series. The phase-rectified signal averaging, or PRSA, has been demonstrated to be a method that shows clear advantages compared to the conventional analysis methods in the time and frequency domains. While these conventional methods have not been capable of providing results that showed statistically significant differences between the groups of low and high risk of sudden cardiac death at night period HRV recordings and also during 24h HRV recordings, the PRSA method not only proved differences between these groups but also allowed to characterize subjects in other groups