Elsevier, NeuroImage, 3(26), p. 839-851, 2005
DOI: 10.1016/j.neuroimage.2005.02.018
Full text: Download
A probabilistic framework is presented that enables image registration, tissue classification, and bias correction to be combined within the same generative model. A derivation of a log-likelihood objective function for the unified model is provided. The model is based on a mixture of Gaussians and is extended to incorporate a smooth intensity variation and nonlinear registration with tissue probability maps. A strategy for optimising the model parameters is described, along with the requisite partial derivatives of the objective function. (c) 2005 Elsevier Inc. All rights reserved.