Automatic Analysis And Classification Of ECG Abnormalities Using C4.5 Decision Tree Classifier
The analysis of electrocardiogram (ECG) signal is the most important task for diagnosis of heart diseases. Any subtle changes in the peaks and valleys of the ECG signal can be difficult to monitored manually. Hence an automated diagnosis approach for detection of cardiac abnormalities is preferable for early detection. In this work, we have used empirical mode decomposition (EMD) method for R-peaks and QRS complex detection. Several morphological features are analyzed from the signal and entered to the classifier for the detection of abnormal beats. A set of eighteen attributes were extracted from each cardiac cycle and used for classifying cardiac abnormalities such as: Normal (N), Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB), Atrial Premature beats (APB) and Premature Ventricular Contractions (PVCs) using C4.5 classifier. The effectiveness of the classifier in terms of Sensitivity (Se), Specificity (Sp), Positive predictivity (Pp) and Accuracy (Acc) are estimated using the physionet arrhythmia database. The best performance result we obtained at fold 10 with an accuracy of 99.23% in C4.5 classifier.