Abstract
In this study, wavelet entropy, which is calculated from the wavelet transform coefficients obtained from heart rate variability data, is used to distinguish the control group from the patients with congestive heart failure. Wavelet entropies are obtained from 29 patients with congestive heart failure and 54 subjects in the control group. In addition, standard heart rate variability (HRV) indices are also calculated for the whole dataset. Then, the performance of these indices in classifying these two groups is evaluated using k-Nearest Neighbor classifier and genetic algorithm. As a result, the subset of the HRV indices that increase the performance of the classifier is obtained. Using the optimal subset of HRV measures gives discrimination accuracy of 97.59%.
