This paper describes the application of Machine Learning (ML) techniques to a real world problem: the automatic diagnosis (classification) of mammary biopsy images. the starting point consistsin a set of data (solved cases) provided by the signal theory research group of our university. The techniques applied are genetic algorithms (GA) and case-based reasoning (CBR). The paper compares our results with previous ones obtained using neural networks (NN). The main goals are: to efficiently solve classification problems of such a type, to compare different alternatives for ML and to study hybrid systems. The paper also introduces the systems we developed for solving this kind of classification problems: GeB-Cs (Genetic Based Classifier System) fo a GA approach, and CaB-CS (Case-Based Classifier System) for a CBR approach