Cardiac Disease Classification Using Denoising and Wavelet Transformation of ECG Signals

Abstract

Cardiovascular disease (CVD) is one of the leading causes of death worldwide, and it is well understood that early diagnosis of the occurrence is critical for successful preventive therapies. While aortic stiffness has been shown to be an independent indicator of CVD, determining it is difficult and time-consuming. Looking for arterial properties such as arterial stiffness is another, much easier way to go about it. Traditional signal processing technologies, as well as machine learning and its sub-branches, such as deep learning, are common techniques for analyzing and classifying ECG signals, with the aim of developing applications for the early detection and treatment of cardiac conditions and arrhythmias. Several forms of classifiers have been used in previous study works to classify pathological CVDs attributable to conventional risk factors such as cigarette smoking, including Artificial Neural Networks (ANN), Fuzzy Logic Systems, Linear Discriminant Analysis (LDA), and Support Vector Machine (SVM). In their study, the majority of the researchers used SVM and Fuzzy Logic systems. The CVD detection model based on Dirichlet classification as well as logistic regression is presented in this article. Before starting the training phase, we use a few denoising approaches to smooth out the ECG signals.