Feature extraction of ECG signal for detection of ventricular fibrillation

Abstract

Ventricular fibrillation (VF) is the intense arrhythmia condition which is the major cause of cardiac arrest. Quick and precise detection of VF is crucial for the success of delivering an electrical shock through defibrillator to save life. Feature extraction algorithms have been used in electrocardiogram (ECG) signal to extract temporal and spectral parameters for rhythm detection. In this paper, we present different arrhythmias detection algorithms for feature extraction of ECG signal. Seven parameters both temporal and spectral features are computed for normal and abnormal conditions of ECG signals. The algorithms are tested and the results are compared with widely recognized databases of MITBIH, SVDB. The extracted features may be used to improve the efficiency of machine learning algorithms for detection of life-threatening arrhythmias.