Classification of Supraventricular Arrhythmias using Wavelet Decomposition

Accurate observation of cardiac dysrhythmias are extremely important for clinical applications. Arrhythmias are one of the leading causes of cardiovascular mortality, direct evidence of clinical records has been lacking. Authors of this paper present a unified approach of discrete wavelet transform (DWT) for extracting the features from ECG signals, cross recurrence quantification analysis (CRQA) for calculating the recurrent rate valuesusing the cross-recurrence plot (CRP) toolboxof MATLABand probabilistic neural network (PNN) concept. Our aim in this study is to classify supraventricular arrhythmias (SVA) signals using the combined approach of DWT, CRQA and PNN. Authors achieved 100% accuracy by using this approach. The MIT-BIHsupraventricular arrhythmias (SVA)was used for experimental results.