Premature Ventricular Contraction Classification based on ECG Signal using Multilevel Wavelet entropy

Abstract

One of the abnormalities in the heart that can be assessed from an ECG signal is premature ventricle contraction (PVC). PVC is a form of arrhythmia in the form of irregularity in beat ECG signals. In this study, a multilevel wavelet entropy method was developed to distinguish PVC and normal ECG signals automatically. Data was taken from the MIT-BIH arrhythmia database with the process carried out is normalization, median filtering, beat-parsing, MWE calculation and classification using SVM. The results of the experiment showed that MWE level 5 with DB2 as mother wavelet and Quadratic SVM as classifier resulted in the highest accuracy of 94.9%. MWE level 5 means only five features needed for classification. The number of features is very little compared to previous research with a quite high accuracy.

Classification of Premature Ventricular Contraction (PVC) based on ECG Signal using Convolutional Neural Network

“…There were two processes conducted in the ECG signal. The first one was the mean removal as in Equation (1) [9].…”

“…Meanwhile, other research used DNN for the biometric ECG classification [19]. Ulah, et al transformed 1d ECG signal to 2-D spectrograms through short-time Fourier transform [9]. 2D-CNN was used as classifier to produce 99.11% of accuracy.…”

“…Meanwhile, for the classifier, it used K-nearest neighbor (KNN), neural network (NN), decision tree (DT), and support vector machine (SVM). Rizal, et al used multilevel wavelet entropy for PVC signal classification [9]. Wavelet entropy was calculated on five levels of wavelet packet decomposition and produced the highest accuracy of 94.9% using SVM as classifier.…”

Premature Ventricular Contraction Detection Based on Chebyshev Polynomials and K Nearest Neighbours Classifier

“…The PhysioNet/CINC 2020 and 2021 Challenges [17,18] provide an opportunity to discuss the complexities of ECG classification from several perspectives and the impact of analysing large numbers of leads. Algorithms for ECG classification can be divided into two groups: morphologybased methods [1,[13][14][15][19][20][21][22][23] and deep learning-based methods [16,[24][25][26][27][28][29][30].…”

“…These methods achieve high accuracy in recognising PVC arrhythmias and different heart rhythms. Rizal et al employed a technique based on multilevel wavelet entropy to differentiate premature ventricular contraction (PVC) beats [19]. Jenny et al [1] utilised Independent Component Analysis (ICA) to extract features, which is one of the most reliable techniques in this domain and is based on the assumption that most measured signals must be mixtures of independent signals.…”