Skip to content. Skip to navigation
CIM Menus

A structural alternative to the deformable brain atlas paradigm

Jean-Francois Mangin
Commissariat Energie Atomique
Orsay France

December 6, 2001 at  3:00 PM
MC 437



The talk will describe a complete system allowing automatic recognition of the main sulci of the human cortex. Sulci are folds of the cortical surface that are the only macroscopic features that can be used to match individual cortices. These sulci, however, present very different shapes across individuals, which make their recognition a challenging problem. Our system relies on a complex preprocessing of MR images leading to abstract structural representations of the cortical folding. This preprocessing consists of a sequence of automatic algorithms mainly based on Mathematical Morphology. The representation nodes are cortical folds, which are given a sulcus name by a contextual pattern recognition method. This method can be interpreted as a graph matching approach, which is driven by the minimization of a global function made up of local potentials. Each potential is a measure of the likelihood of the labelling of a restricted area. This potential is given by a multi-layer perceptron trained on a learning database. A base of 26 brains manually labelled by a neuroanatomist is used to validate our approach. The system developed for the right hemisphere is made up of 265 neural networks. The whole system is a symbolic alternative to the usual deformable atlas principle. This alternative consists of using a higher level of representation of the data to overcome some of the difficulties induced by the complexity and the striking variability of the cortical folding.



D. Rivi\`ere, J.-F. Mangin, D. Papadopoulos, J.-M. Martinez, V. Frouin and J. R\'egis, {\em Automatic recognition of cortical sulci using a congregation of neural networks}, MICCAI'2000, Pittsburgh, LNCS-1935, Springer Verlag, pp 40-49

Speaker background:


{Jean-Fran\c{c}ois Mangin} received the engineering degree from \'{E}cole Centrale Paris in 1989, the {\bf M.Sc.} degree in numerical analysis from Pierre et Marie Curie University (Paris VI) in 1989, and the {\bf PhD} degree in signal and image processing from \'{E}cole Nationale Sup\'{e}rieure des T\'{e}l\'{e}communications of Paris in 1995. Since october 1991, he has been working with Service Hospitalier Fr\'{e}d\'{e}ric Joliot, Commissariat \`{a} l'\'{E}nergie Atomique, Orsay, France, on image analysis problems related to brain mapping. Since 1999, he has been leading a group, which project consists of the development of a brand-new set of brain mapping methods designed from a structural point of view. The goal is to get closer to the class of neuroscience approaches dealing with structural models. The underlying image analysis tools are mainly Mathematical Morphology, Markovian random fields and Graph based representations.