Tractography segmentation using a hierarchical Dirichlet processes mixture model
Author(s)
Wang, Xiaogang; Grimson, W. Eric L.; Westin, Carl-Fredrik
DownloadAccepted version (2.454Mb)
Terms of use
Metadata
Show full item recordAbstract
In this paper, we propose a new nonparametric Bayesian framework to cluster white matter fiber tracts into bundles using a hierarchical Dirichlet processes mixture (HDPM) model. The number of clusters is automatically learned driven by data with a Dirichlet process (DP) prior instead of being manually specified. After the models of bundles have been learned from training data without supervision, they can be used as priors to cluster/classify fibers of new subjects for comparison across subjects. When clustering fibers of new subjects, new clusters can be created for structures not observed in the training data. Our approach does not require computing pairwise distances between fibers and can cluster a huge set of fibers across multiple subjects. We present results on several data sets, the largest of which has more than 120,000 fibers. ©2010 Elsevier Inc.
Date issued
2010-08Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence LaboratoryJournal
NeuroImage
Publisher
Elsevier BV
Citation
Wang, Xiaogang et al., "Tractography segmentation using a hierarchical Dirichlet processes mixture model." NeuroImage 54, 1 (January 2011): 290-302 doi. 10.1016/j.neuroimage.2010.07.050 ©2010 Authors
Version: Author's final manuscript
ISSN
1095-9572