Use of artificial neural networks for clinical decision-making (Maldescensus testis)

The application of artificial neural networks (ANNs) to medical diagnosis and in particular for the Maldescensus testis domain is presented. Various architectures and schemes using the traditional backpropagation are considered. The input data to the neural networks were encoded in two different ways, that is, using real values and gray code representation. These different architectures and schemes were evaluated and compared in terms of classification accuracy and speed