Dissimilarity application for medical imaging classification

In the traditional way of learning from examples of objects the classifiers are built in a feature space. However, alternative ways can be found by constructing decision rules on dissimilarity (distance) representations. In such a recognition process a new object is described by its distances to (a subset of) die training samples. The use of the dissimilarities is especially of interest when features are difficult to obtain or when they have a little discriminative power. Purpose of this work is the development of an automatic classification system which could be useful for radiologists in the investigation of breast cancer. The software has been designed in the framework of the MAGIC-5 collaboration. In the automatic classification system the suspicious regions with high probability to include a lesion are extracted from the image as regions of interest (ROIs). Each ROI is characterized by some features extracted from co-occurrence matrix containing spatial statistics information on ROI pixel gray tones. A dissimilarity representation of diese features is made before me classification. A feed-forward neural network is employed to distinguish pathological records, from non-pathological ones by the new features. The results obtained in terms of sensitivity (percentage of pathological ROIs correctly classified) and specificity (percentage of non-pathological ROIs correcdy classified) will be presented.