Abstract:
An Electrocardiogram (ECG) signal gives a significant information for the cardiologist to detect cardiac diseases. A significant amount of research and development effort has been devoted on the detection of the cardiac disease but still cannot get control on death rate. ECG signal is a self-similar object. So, fractal analysis can be implemented for proper utilization of the gathered information. The aim of this dissertation is to present a Fractal Dimension method to analyze three specific heart diseases namely Premature Atrial Contraction (PAC), Premature Ventricular Contraction (PVC), and Atrial Fibrillation.
A Graphical User Interface (GUI) was designed using MATLAB to calculate the Fractal Dimension, to distinguish between the ECG signals of healthy person and patients with the three specific heart diseases from the raw ECG data. ECG signals used were taken from three databases, MIT-BIH Arrhythmia Database, the MIT-BIH Normal Sinus Rhythm Database, and the Intracardiac Atrial Fibrillation Database. The results show that the fractal dimension performs well in classifying the different ECG signals.