Heartbeat type classification with optimized feature vectors

Abstract

In this study, a feature vector optimization based method has been proposed forclassification of the heartbeat types. Electrocardiogram (ECG) signals of fivedifferent heartbeat type were used for this aim. Firstly, wavelet transform (WT)method were applied on these ECG signals to generate all feature vectors.Optimizing these feature vectors is provided by performing particle swarmoptimization (PSO), genetic search, best first, greedy stepwise and multi objectiveevoluationary algorithms on these vectors. These optimized feature vectors arelater applied to the classifier inputs for performance evaluation. A comprehensiveassessment was presented for the determination of optimized feature vectors forECG signals and best-performing classifier for these optimized feature vectors wasdetermined.