Arrhythmia Classification Using Neural Netwrok

Abstract

This paper presents an effective application of artificial neural network for diagnosis of cardiac arrhythmia. Comparison of the result generated by Backpropagation and Conjugate gradient algorithm. The classifier was developed and tested with the MIT-BIH Arrhythmia Database. Automatic analysis is performed by using artificial neural network. Following three consecutive steps are required for automatic detection: 1) R-R interval detection 2) Heart Rate Calculation and 3) Classification.