Interpretation of multimodal medical images using connectionist and variational methods

Our medical objective is to match multimodal 3D medical images into a coherent model of the patient, from which diagnosis can be assessed and therapeutics guided. 3D image segmentation is absolutely necessary to reach this objective. We investigate two complementary approaches for segmenting 3D medical images. First we present some definitions, basic properties and recent theoretical results about formal neural networks, and show that these results can be applied to brain tumour segmentation. A variational approach (called the 'snake spline' method) is then detailed. We finally show how segmented 3D images can be used for multimodal image matching.