Time-frequency analysis of heart rate variability as an aid to diagnose pathologies

The heart rate variability is a diagnostic measure of great clinical interest. Previous results have shown that the position of the maximum peak of the heart rate power spectrum could be an index to distinguish in postmyocardial infarction (PMI) patients those that later develop malignant ventricular arrhythmia (MVA). The instantaneous frequencies in the low frequency band (LF, 0-0.07 Hz), mean frequency band (MF 0.07-0.15 Hz), and high frequency band (HF, 0.15-0.45 Hz) have been obtained by means of the time-frequency analysis using the Choi-Williams distribution. Two groups of patients have been selected in this study: group (a) consisted of 9 PMI patients who presented MVA during or after Holter recording, and group (b) consisted of 10 PMI patients without MVA. This work shows that the Choi-Williams time-frequency distribution of the R-R signal is useful to obtain new discriminatory variables for these patients. The difference between the mean frequencies in the MF and LF bands during the daytime allows to separate both groups, and the statistical significance is greatly increased (p < 0.001) in a multivariate analysis when the instantaneous frequencies in the LF. MF, HF bands at the daytime and nocturnal period are considered separately.