A Neuro-fuzzy Based Model for Analysis of an ECG Signal Using Wavelet Packet Tree

Detection and classification of electrocardiogram (ECG) signals are critically linked to the diagnosis abnormalities. Any abnormality in the wave shape and duration of the wave features of the ECG is considered as arrhythmia. This paper presents a diagnostic system for classification of cardiac arrhythmia from ECG data, using hybrid model of Artificial Neural Network and Fuzzy Logic. In an ECG, clinically useful information is obtained from the intervals and amplitudes of the cardiac waves. In an ECG, the non-stationary signal commonly changed its statistical property with time. In the proposed paper an algorithm based on wavelet packet tree classifier (for detection of QRS complex) has been implemented for the comparative study of automatic real-time ECG data. The amplitude and duration of the characteristic waves of the ECG can be more accurately obtained using Wavelet Packet Tree (WPT) analysis. WPT techniques have been employed to extract a set of linear (time and frequency domain) characteristics. Neuro-fuzzy techniques have been employed to extract a set of non-linear characteristic features from the transformed ECG signals. The real-time signals are obtained from various diagnostic centers. The hybrid model of Wavelet Packet Tree and Neuro-fuzzy network is proposed for the analysis and comparative study of an ECG signal.