Abstract:

Cardiac activity is one of the most important determinants of a patient's condition. It results in the appearance of several waves on the course of the electrocardiograph: it is the cardiac signal, the electrocardiogram: ECG. The analysis of the ECG signal and the identification of its parameters constitute an essential step for the diagnosis. However, a set of methods and algorithms are developed in view of the importance of this signal and its use in clinical routine in the diagnosis of cardiac pathological cases. This paper fits into this problem and proposes a classifier of cardiac arrhythmias by application of neural networks. The results were validated by ECG signals from the different patients in the MIT-BIH Arrhythmias database, and given a recognition rate of 94%, this rate of classification exceeds the results obtained in the literature.

Keywords: