CHERRADI, Bouchaib Université Hassan II Mohammadia - E N S E T, Morocco , Bouattane, Omar Université Hassan II Mohammedia - Ecole Normale Supérieure de l Enseignement Technique (ENSET), Maroc , YOUSSFI, Mohamed Université Mohammed-V - Faculty of Science, Morocco , BENSALAH, Mohamed O. Université Mohammed-V - Faculty of Science, Morocco , RAIHANI, Abdelhadi Université Hassan II Mohammadia - Faculty of Science, Morocco

Parallel Algorithm for Brain Tissues Segmentation in T1-Weighted MR Images on 3D ReconfigurableMesh Computer

35328

In this paper, we propose a parallel algorithm for brain tissues segmentation from T1-weighted Magnetic Resonance Images (MRI) on Massively Parallel architecture named reconfigurable mesh computer (MCR), this brain tissues are already extracted using our method named Threshold Morphologic Brain Extraction method (TMBE). The use of this massively parallel architecture is introduced in order to improve the complexities of the corresponding algorithms. The image of size (M x N x K) to be processed must be stored on the RMC of the same size, one Voxel per Processing Element (PE). The proposed method consists in the brain tissues segmentation using parallel version of the modified fuzzy c-means MFCM, named PMFCM. This algorithm is directly applied on the extracted volume. The corresponding parallel program of the proposed algorithm is validated on a 3D Reconfigurable Mesh emulator.

142

MRI , Brain Extraction , Brain Tissues segmentation , Parallel computing , 3D , RMC

Mediterranean Telecommunications Journal

147