Improving an automatic arrhythmias recogniser based in ECG signals

Abstract

In the present work, we have developed and improved a tool for the automatic arrhythmias detection, based on neural network with the "more-voted" algorithm. Arrhythmia Database MIT has been used in the work in order to detect eight different states, seven are pathologies and one is normal. The unions of different blocks and its optimization have found an improvement of success rates. In particular, we have used wavelet transform in order to characterize the patron wave of electrocardiogram (ECG), and principal components analysis in order to improve the discrimination of the coefficients. Finally, a neural network with more-voted method has been applied.