Two neural networks architectures for detecting AVB

Authors:
Salama Meghriche;Mohammed Boulemden;Amer Draa
Affiliations:
Research Laboratory in signal processing, University of Batna, Batna, Algeria;Research Laboratory in signal processing, University of Batna, Batna, Algeria;Computer Science Department, University of Constantine, Constantine, Algeria
Venue:
MUSP'08 Proceedings of the 8th WSEAS International Conference on Multimedia systems and signal processing
Year:
2008

Citing 5
Cited 0

Bioinformatics: the machine learning approach

Bioinformatics: the machine learning approach
Neural Networks and Artificial Intelligence for Biomedical Engineering

Neural Networks and Artificial Intelligence for Biomedical Engineering
Bootstrapping confidence intervals for clinical input variable effects in a network trained to identify the presence of acute myocardial infarction

Neural Computation
An integrative and interactive framework for improving biomedical pattern discovery and visualization

IEEE Transactions on Information Technology in Biomedicine
Detecting acute myocardial infarction in the 12-lead ECG using Hermite expansions and neural networks

Artificial Intelligence in Medicine

Quantified Score

Hi-index	0.00

Visualization

Abstract

The purpose of the paper is to present two networks. These networks were fed with same measurements from one lead of the electrocardiogram (ECG) but in different architectures. The first one based on a compound neural network (CNN) composed of three different multilayer neural networks of the feed forward type, and the second one based on only a multi-layer perceptron (MLP). Such both of them have the capability to classify ECGs as carrying atrioventricular blocks (AVB) or not. For each test case in the test set the neural networks classifier present an output value between 0 and 1. A threshold in this interval was used above which all values were regarded as consistent with AVB. The difference in performance between the two neural networks classifiers was measured as the difference in area under the receiver operating characteristic (ROC) curves.