Autoregressive modeling and classification of cardiac arrhythmias

Abstract

Computer assisted analyses of cardiovascular signals facilitate timely diagnosis and treatment in critically ill patients. Different methods have been employed for the analysis and diagnosis, yet there is scope for enhancement of classification/diagnosis accuracy. Autoregressive modeling (AR) has been applied to ECG signals and the AR coefficients have been used for classification into arrhythmias such as atrial premature contraction, premature ventricular contraction, ventricular tachycardia, and ventricular fibrillation. Two classification algorithms including the generalized linear model and a multi-layer feed forward neural network using back propagation are proposed to classify the beats into one of the five classes. The results show that autoregressive coefficients can be an effective tool for modeling and classification of ECG signals.