Combining fuzzy logic and level set methods for 3D MRI brain segmentation

We propose to segment volumetric brain structures with a level set method including a fuzzy decision in the design of the evolution force. The role of fuzzy logic is to fuse gradient-based and region-based information into a single force term, to take advantage of their properties. The gradient-based approach increases the level set speed in high-contrasted areas, whereas the region-based approach is useful to treat nonhomogeneous tissues and areas of complex shapes, on which borders do not appear clearly. This fusion does not require any manually-tuned parameter to balance the respective influence of the gradient and region terms. We integrated this segmentation algorithm in a fully automatic succession of operations involving a registration step from known data to decrease the computation time. Experimental results on the MNI Brainweb dataset and on a database of real MRI volumes are presented and discussed.