Application of Neural Network and Genetic Algorithms in Computer Aided Gastric Diagnostic System
Z Swierczyński, Anna Zaganczyk · 2005
In the paper, computer aided stomach diagnosis problems are presented. The subject of the study is electrical signal generated by human stomach, called the electrogastrographic (EGG) signal. The non-invasively measured signals were subjected to parametrization and then neural network based classification. The parametrization was performed with one of the time series modelling methods, with linear autoregressive models (AR). A special feature of the presented methodology of classification is its hybrid approach. The idea of this specific combination is that a genetic algorithm is used as the evolutionary method of training of the neural network. The structure and parameters of the system (NEUROGEN v.02), used for classification of the parameterized EGG data, are described. The finally obtained effectiveness of the whole system (NEUROGEN v.02 with the parametrization method applied), amounting to 74%, is quite high and, according to the authors’ analysis, can be improved. A way of improvement of the effectiveness are also outlined in the summary.