Binary coding of Arrhythmic beats for generation of simplified classification rule

Detection of Arrhythmic beats at the time of its occurrence is very essential for cardiac monitoring. In this work a feature based binary coded classification method is presented for detection of beats like Premature Ventricular Contraction, Bundle Branch Block and Normal. The method is based on mainly the QRS features. A binary coded word is formed representing the status of each ECG beat comprising of five binary bits and the bit status is checked for the one by one to indicate the type of the beat. The binary status word is formed using the temporal and morphological features of ECG signal. This approach of logical classification reduces the complexity of detection because of its binary nature resulting in fast and ready reference of arrhythmic disorder for standalone systems. Performance evaluation results indicate that the method has good accuracy and positive predictivity for all kind of test beats.