An Accurate and Fully Automatic Segmentation of Normal Brain Guided by Deformable Model

S. Bourouis and K. Hamrouni (Tunisia)


Medical imaging; Brain; MRI; 3D segmentation; Level-setapproach.


The automatic 3D segmentation of MR images has been an area of intense study. However, this task has proven problematic, due to the non homogeneous intensities. Here we investigate this line of research by introducing a new automatic algorithm for segmenting brain magnetic resonance (MR) images into anatomical structures. A novel algorithm driven by level set model has been developed. Our model integrates both local and global information into the same formalism. This formalism offers signi´Čücant advantages over many other methods such as the work of Bourouis et al. [1]. It is able to give good estimation and segmentation of tissue volume. Qualitative and quantitative evaluations of the automatic segmentation against a ground truth are presented. The obtained results are validated on many real data and against a ground truth.

Important Links:

Go Back