PARTIAL VOLUME TISSUE CLASSIFICATION OF MULTICHANNEL MAGNETIC-RESONANCE IMAGES - A MIXEL MODEL

A single volume element (voxel) in a medical image may be composed of a mixture of multiple tissue types. We call voxels which contain multiple tissue classes mixels. A statistical mixel image model based on Markov random field (MRF) theory and an algorithm for the classification of mixels are presented in this paper. We concentrate on the classification of multichannel magnetic resonance (MR) images of the brain although the algorithm has other applications. We also present a method for compensating for the gray-level variation of MR images between different slices, which is primarily caused by the inhomogeneity of the RF field produced by the imaging coli.