The identification of ventricular escape beat based on wavelet transform and BP Neural Network

Ventricular escape beat is one of the most serious diseases of the arrhythmia, which usually happen in the patients that suffer coronary heart disease, myocarditis and myocardial infaction. It has essential clinical significance to find ventricular escape beat promptly and exactly. This paper proposes a method of identification, which based on wavelet transform and BP neural network, to benefit the intelligent systems of ECG monitoring with an efficacious solution. After selecting the characteristics of ECG signals of the MIT-BIH ECG database using wavelet transform, this paper constructs feature vectors by the results to train BP Neural Network, which then is used to identify the ventricular escape beat. The practical value of this method is substantiated by the final ideal rate of identification.