ANALYSIS AND CLASSIFICATION OF CARDIAC ARRHYTHMIA USING ECG SIGNAL

Abstract

The main objective of this study is to provide recognition of cardiac arrhythmic pathologies from the classification of ECG recordings. ECG is a graphical record which is a result of electrical tension of heart and is the most important biosignal used by cardiologists for diagnostic purposes. The difficulty faced in interpretation of ECG signals forced researchers to study about automatic detection of cardiac arrhythmia disorders. Using data analysis techniques, computer programs could easily interpret complex ECG signals, predict presence or absence of cardiac arrhythmia and provide real time analysis and diagnosis. In this thesis, Support Vector Machine (SVM) technique using LIBSVM 2.98 has been applied to ECG dataset for arrhythmia classification in five categories one normal, and four arrhythmic beat categories. The dataset used in this study is 3003 arrhythmic beats out of which 2101 beats (70%) for training 902 beats (30%) have been used for testing purpose. Whole dataset is obtained from MIT-BIH Arrhythmia dataset. Filtering of signal is important for artifact removal. For feature extraction AcqKnowledge software is used. The results are presented in the terms of sensitivity and positive predictions. Total performance is found to be around 96 % in this case