Classification of ECG Signal using Artificial Neural Network

Abstract

Electrocardiogram (ECG) is one of the significant investigative tool used in determining the health condition of heart. The raise in number of heart patients has necessitated a technique for automatic determination of diverse abnormalities of heart for lessening the pressure on the specialists or sharing their work load. The work presented in this paper facilitates in generating a computer based system that assists in categorizing the ECG signals. Artificial Neural Network (ANN) is been used for the classification of the signal. The various steps used for the determination of type of ECG signal are preprocessing, Feature extraction & selection and classification. The considered neural network is used to classify the six categories of arrhythmias named Normal Sinus, Right Bundle Branch Block (RBBB), Atrial Premature Beat (APB), Left Bundle Branch Block (LBBB), Arterial fibrillation, PVC. The simulation is done in MATLAB. The obtained results shows that the proposed classifier shows the enhanced performance sensitivity 95%, Specificity99% and classification accuracy 98%. This work provides the comparative analysis of the performance of proposed classifier with KNN, ANFIS and Naive Bias. The results shows the performance of proposed technique is better than other techniques.