Diagnosis of asthma severity using artificial neural networks

Abstract

During the last years, neural networks have become a very important method in the field of medical diagnostic. In this work, a technique is proposed that involves training a Multi-Layer Perceptron with back-propagation learning algorithm, in order to recognize three classes of asthma severity, through the results of breathing tests. The breathing test parameters and the diagnosis of physicians for 200 cases of children- patients, aged 10-12 years from Alexandroupolis Hospital in Greece, are used in the supervised training method to update the network parameters. This method was implemented to diagnose three asthma cases according to their severity: mild, moderate and severe asthma. Results obtained by using Neural Network Toolbox of Matlab, show that the proposed ANN can be used in asthma diagnosis with 98% success. This research work improves the asthma diagnosis accuracy with higher consistency in order to specify the seriousness of the condition of a patient and the appropriate course of medical treatment. © 2010 International Federation for Medical and Biological Engineering.