I develop computational methods to automatically segment Magnetic Resonance (MR) images of the human brain. This is a difficult problem because of the complex anatomy of the brain and its enormous variability across individuals, the large overlap in intensity characteristics between brain structures of interest, and the dependency of MRI intensities on the acquisition sequences used. In order to overcome these difficulties, I develop Bayesian methods in which computational models of the MR acquisition process are combined with models of the expected location and shape of the various brain structures under investigation. My work has resulted in several successful methods for sequence-adaptive brain tissue classification, MR eld inhomogeneity correction, automated brain lesion delineation, partial volume segmentation, and computational modeling of brain anatomy. My current research focuses on applying these techniques to fully-automated segmentation of the subfields of the hippocampus in ultra-high resolution MRI.