Analysis of Electrocardiogram Signals Using Machine Learning Techniques – A Review

According to the most recent survey, 60% of patients die as a result of heart disorders, which accounts for a major cause of death in this century. Electrocardiogram (ECG) signals can be used to diagnose these conditions. Electrical activity that the heart produces is captured in an ECG signal. Building a model to do automatic ECG categorization has piqued interest in the analysis of the ECG signal for more than ten years. This research's major objective is to examine and provide a summary of using modern classification techniques such artificial neural networks (ANN), convolutional neural networks (CNN), discrete wavelet transforms (DWT), support vector machines (SVM), and K-Nearest Neighbour (KNN). The final product includes effective comparisons of classification techniques, feature extraction methods, datasets, contributions, and other factors. The outcome also demonstrates that the CNN has been the most popular method for classifying ECGs since it has a greater success rate than the remaining classification approaches.