Classification of Normal and Abnormal ECG Signals Based on their PQRST Intervals

In this paper, we propose a system that is capable of automatically differentiating between normal and abnormal heartbeats of patients using signals acquired from electrocardiography (ECG). The components of the ECG signals, that are PQRST intervals, were studied to acquire features for classification. Different time intervals of p-wave, QRS complex and t-wave were used as features. These features were fed to a linear discriminant analysis to classify the normal and abnormal heartbeats. The classification accuracy was above 80% on average. The results demonstrate the feasibility of development of a machine that is able to automatically detect all potential heart related diseases that can be identified from ECG signals manually.