Classification of ECG signal with Support Vector Machine Method for Arrhythmia Detection

IOP Conf. Series: Journal of Physics: Conf. Series 970 (2018) 012012, doi :10.1088/1742-6596/970/1/012012

Selasa, 03 April 2018 13:48 | Sudah dibaca 960 kali

An electrocardiogram is a potential bioelectric record that occurs as a result of cardiac activity. QRS Detection with zero crossing calculation is one method that can precisely determine peak R of QRS wave as part of arrhythmia detection. In this paper, two experimental scheme (2 minutes duration with different activities: relaxed and, typing) were conducted. From the two experiments it were obtained: accuracy, sensitivity, and positive predictivity about 100% each for the first experiment and about 79%, 93%, 83% for the second experiment, respectively. Furthermore, the feature set of MIT-BIH arrhythmia using the support vector machine (SVM) method on the WEKA software is evaluated. By combining the available attributes on the WEKA algorithm, the result is constant since all classes of SVM goes to the normal class with average 88.49% accuracy.

Kata Kunci: ECG, QRS, MIT-BIH, SVM, WEKA