Determining best HRV indices for PAF screening using genetic algorithm

Heart Rate Variability (HRV) analysis is used for diagnosis of various cardiac diseases. In this study, the genetic algorithm (GA) is used for the selection of optimal subset of HRV features to detect paroxysmal atrial fibrillation (PAF) patients from their ectopic-free ECG records. The K-nearest neighbors (K-NN) algorithm is used as the classifier. This GA-based method reduced the number of HRV features from 33 to 7 and increased the classifier's performance from 90.2% to 92.2%.