This paper aims to develop a support system for seroprevalence prediction of hepatitis A. Logistic regression and artificial neural network models were considered. The accuracy of these models was measured based on the misclassification rate in a sample from the city of Duque de Caxias, Rio de Janeiro, where there is a high incidence of this disease. The results of the evaluation show that the neural model achieves an overall classification efficiency of 88%, when it uses relevant information extracted from the logistic model.
Neural Networks; Logistic Regression; Hepatitis A; Supporting Systems for Medical Diagnosis