| Summary: Cardiac arrhythmia is one of the most important indicators of heart disease. Heart arrhythmias are caused by any disruption in the regularity, rate, or transmission of the cardiac electrical impulse. Premature ventricular contractions (PVCs) are a common form of cardiac arrhythmia caused by ectopic heartbeats. The detection of arrhythmias by means of ECG (electrocardiogram) signals is important for the prediction of possible heart failure. This study was first focused on PVC classification using time series of ECG signal. Moreover, the performance effects of several dimension reduction approaches were also tested. In addition, the work is extended to classify more common arrhythmia types. Statistical features are calculated from one beat signal for the classification. Feature size were reduced to a lower size using size reduction algorithms. Experiments were carried out using well-known machine learning methods, including neural networks, k-nearest neighbor, decision trees, and support vector machines. Findings were expressed in terms of accuracy, sensitivity, specificity, selectivity and running time. The data used in the tests in this study were taken from the MIT-BIH arrhythmia database that has become the standard in this area.
Key Words: ECG signal analysis, Premature Ventricular Contraction, ECG arrhythmia classification, Feature size reduction, Signal processing, Machine learning. |
