Conditional Random Fields For Brain Tissue Segmentation
Magnano, Ameet B. Soni, S. Natarajan, Gautam Kunapuli · 2013
Current atlas-based methods for MRI analysis assume brain images map to a ormal template. This assumption, however, does not hold when analyzing abnormal brain shapes or disease states. We pro- pose a discriminative-graphical model framework based on conditional random elds (CRFs) to mine MRI brain images. As a proof-of-concept, we apply CRFs to the problem of brain tissue segmentation. Experi- mental results show robust and accurate performance on tissue segmen- tation comparable to other state-of-the-art segmentation methods. Our algorithm generalizes well across data sets and is less susceptible to out- liers, while relying on minimal prior knowledge relative to atlas-based techniques. These results provide a promising framework for future ap- plication on disease classication and atlas-free anatomical segmentation.