Cardiac Arrhythmia Help – Diagnosis System Using Wavelets and Hidden Markov Models

Abstract:- This paper reports the development of a help-diagnosis system where the physician is required to analyze some ECG pulses that can not be accurately classified by the system. A confidence measure is estimated on the basis of massive experimental tests on data from MIT-BIH Arrhythmia Database, and was set on a threshold above which no classification errors were obtained. Cardiac arrhythmia detection and classification is performed by using Wavelets and Hidden Markov Models (HMMs). The types of beat being selected are normal (N), premature ventricular contraction (V) which is often precursor of ventricular arrhythmia, two of the most common class of supra-ventricular arrhythmia (S), named atrial fibrillation (AF), atrial flutter (AFL), and normal rhythm (N). Experimental results are obtained in real data from MIT-BIH Arrhythmia Database and a developed Data-Acquisition System