Neural network models for studying and for managing in real-time schistosomiasis control programs

Introduces a novel application of neural network (NN) methodology for processing infectious disease epidemiological data. Specifically it presents a neural network model that is trained, on the basis of annual schistosomiasis epidemiological data collected from 251 school children in Egypt. The trained model is then used to predict population infection rates from the following two years. Actual 2nd and 3rd year data are used in the evaluation. The performance of the model compares favorably with traditional epidemiological results based on logistic regression. Unique aspects of NN modelling are explored to aid the management of wide-scale disease control programs and to increase overall cost-effectiveness.