Population intensity outliers or a new model for brain WM abnormalities
Xavier Tomas-Fernandez, Simon Keith Warfield · 2012
We present a new automatic method for segmentation of Multiple Sclerosis (MS) lesions in Magnetic Resonance Images. The algorithm performs tissue classification combining a within subject global tissue intensity model and a local tissue intensity model derived from an aligned set of healthy reference subjects. MS lesions are detected as outliers towards the proposed coupled global/local intensity model. Evaluation using BrainWeb synthetic, show our new coupled local/global intensity GMM model to be sensitive towards MS lesions, as well to be robust to noise and intensity inhomogeneity artifacts found MRI scans.