Les réseaux de neurones un outil de sélection de variables : Le cas des facteurs de risque de la maladie du cancer du sein

Abstract

This paper uses the neural networks with an incremental learning algorithm as tool of selection of the most relevant risk factors in the disease of the breast cancer diagnosis. The results testify of the relevance of the neuronal approach with an incremental algorithm in research. An experimental study is performed using simulated data are further validated on real clinical data for breast cancer diagnosis; it was possible to us to determine the optimal combination of factors allowing to affect a good predictive performance of the type of tumor.