A real time approach of cardiac profiling scheme for ECG beat classification using shared counters

The QRS complex is the most striking waveform within the electrocardiogram (ECG). QRS provides the fundamentals for almost all automated ECG analysis algorithms. We proposed a novel method for classification of ECG beats using shared counters. We first developed a real-time QRS detection technique using two-phase hashing to extract precise QRS points. Then we proposed a classifier for profiling an ECG of normal patient. Our proposed technique depends much on the series of data corresponding to particular feature. For minimize the memory requirement shared counters concept has used. The algorithm automatically adjusts parameters periodically to adapt to ECG changes as QRS morphology and heart rate. The significance of such profiling is for diagnosis of various heart diseases. Automatic abnormal cardiac behavior of any individual can be detected by using this system.