Recognition of arrhythmic electrocardiogram using wavelet based feature extraction

摘要

Arrhythmia is one of the most common cardiac diseases. Efficient methods of detecting arrhythmia have been proposed in literatures. Our study proposes a unique feature extraction approach with entropy and Hjorth descriptor to classify a set of ECG signals into normal and arrhythmic with a considerable amount of accuracy. The conventional approach involving wavelet decomposition as the primary feature extraction method yields classification accuracy of 81.8%. The method proposed in the study using entropy and Hjorth descriptor provides higher classification rate at 82.9%. Our study is validated by a reliable dataset.