Classification Atrial Fibrillation Using Stacked Autoencoders Neural Networks

Abstract

In this paper, a combination of deep learning method called stacked auto encoder with the aim of classifying atrial fibrillation (AF) is utilized. An electrocardiogram (ECG) signals from MIT-BIH database are used and spectral, time and non-linear features are extracted from this signal. First extracted features were evaluated using statistical test, analysis of variance (ANOVA) and selected significant features then used for stacked auto encoder as parallel form to classify AF and normal samples. Then, final decision performed using the ensemble averaging method. The average accuracy for classifying AF and normal samples were 95.5%.