Brain tissue classification of alzheimer disease using partial volume possibilistic modeling: Application to ADNI phantom images

Abstract

This paper describes an automatic segmentation approach for PET and T1-weighted MR images using a possibilistic clustering algorithm for deriving fuzzy tissue maps of white matter, gray matter and cerebrospinal fluid volumes, and using the fuzzy C-means algorithm for the centers initialization process; this hybrid technique allows to compute the degree of membership of each voxel to different brain tissues. The fuzzy process is illustrated for Alzheimer's disease using phantom images from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Our method, inspired from the conventional possibilistic algorithm, is less sensitive to noise while taking into consideration the effect of partial volume.