Diagnosing Chronic Obstructive Pulmonary Disease with Artificial Neural Networks using Health Expert Guidelines

Maria Angélica de Oliveira Camargo-Brunetto, André R. Gonçalves

2013

Abstract

Chronic Obstructive Pulmonary Disease (COPD) is characterized by airflow limitation and the spirometry is one of the tests that can be used to detect such disease. However there is a great problem related to the different ways of interpreting the values provided by spirometric devices, regarding different guidelines and reference values. Artificial Neural Networks (ANN) can be used to help with tasks of diagnosis as that. This work presents the modeling and analysis of three ANN models to classify subjects with COPD, based on different sets of variables: a set of observed measures from spirometry and a set of interpreted values according to the guideline proposed by the American Thoracic Society. The results shown that it is possible to achieve a good accuracy in the diagnosis of COPD using ANNs, besides these features set conducted the COPD identification problem to a nearly linearly separable classification problem.

Paper Citation

in Harvard Style

de Oliveira Camargo-Brunetto M. and R. Gonçalves A. (2013). Diagnosing Chronic Obstructive Pulmonary Disease with Artificial Neural Networks using Health Expert Guidelines . In Proceedings of the International Conference on Health Informatics - Volume 1: HEALTHINF, (BIOSTEC 2013) ISBN 978-989-8565-37-2, pages 207-214. DOI: 10.5220/0004234102070214

in Bibtex Style

@conference{healthinf13,
author={Maria Angélica de Oliveira Camargo-Brunetto and André R. Gonçalves},
title={Diagnosing Chronic Obstructive Pulmonary Disease with Artificial Neural Networks using Health Expert Guidelines},
booktitle={Proceedings of the International Conference on Health Informatics - Volume 1: HEALTHINF, (BIOSTEC 2013)},
year={2013},
pages={207-214},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004234102070214},
isbn={978-989-8565-37-2},
}

in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Health Informatics - Volume 1: HEALTHINF, (BIOSTEC 2013)
TI - Diagnosing Chronic Obstructive Pulmonary Disease with Artificial Neural Networks using Health Expert Guidelines
SN - 978-989-8565-37-2
AU - de Oliveira Camargo-Brunetto M.
AU - R. Gonçalves A.
PY - 2013
SP - 207
EP - 214
DO - 10.5220/0004234102070214