A perceptron artificial neural network (PNN) model is proposed to discriminate zones of high mineral potential in the Eastern Desert of Egypt using remote sensing and airborne spectral gamma-ray data stored in a GIS database. A neural network model with one hidden unit was selected by means of a perceptron neuron, which uses the hard-limit transfer function. The trained network delineated a gold potential map efficiently, detected a previously known area as well as a suggested potentially mineralized one. These initial results suggest that PNN can be an effective tool for mineral exploration using spatial data modeling.