A Novel FrWT Based Arrhythmia Detection in ECG Signal Using YWARA and PCA

In general, Electrocardiogram (ECG) signal gets corrupted by variety of noise at the time of its acquisition. Unfortunately, these noise tend to mask the crucial information. Consequently, it may endanger life of the subject (patient) due to delayed diagnosis of heart health. In critical situations, proper analysis of ECG signals is very important for correct and timely detection of heart diseases. This situation motivated the present authors to develop an efficient arrhythmia detection algorithm. In this paper, a novel fractional wavelet transform (FrWT), Yule–Walker Autoregressive Analysis (YWARA), and Principal Component Analysis (PCA) are used for preprocessing, feature extraction, and detection, respectively. The type of arrhythmia detected has been interpreted based on variance estimation theory. For performance evaluation, various statistical parameters such as mean square error (MSE), detection accuracy (Acc), & output signal-to-noise ratio (SNR) are used. The proposed algorithm achieved a MSE of 0.1656%, Acc of 99.89%, & output SNR of 25.25 dB for MIT-BIH Arrhythmia database. For complete validation of this proposed work, other databases such as ventricular tachyarrhythmia, MIT-BIH long-term, and atrial fibrillation are also utilized.