Method for detecting ECG arrhythmia / 医疗卫生装备

ABSTRACT

Usually, the frequency of heartbeat can be used to discriminate different kinds of arrhythmia from a normal cardiac rhythm, but the result is not satisfying. This paper presents a method that can differ ventricular fibrillation (VF) and ventricular tachycardia (VT) from a normal sinus rhythm (NSR). VT and VF are fatal arrhythmia for patients, but timely electroshock is a good remedy for them. The method above can be integrated into the monitor system or telemedicine to diagnose patients and then treat them properly. With auto regressive (AR) model applied to modeling, Itakura and Euclidian distance measurements are used to classify data. With this method, VF and VT conditions are detected with error less than 10%.