Telemedicine has started to be beneficial to patients in remote regions. It is very important to monitor the ECG signals of these patients with heart disorders. Developments in information technology have started to provide important contribution to the clinical decision support systems for early detection and diagnosis. This study aimed to be part of clinical decision support systems and used easily calculated features for detection of ECG arrhythmia. Different classification methods are compared using these features. The performance of the method is tested on data used obtained from the PhysioNet database.