A mathematical framework for incorporating anatomical knowledge in DT-MRI analysis

被引：11

作者：

Maddah, Mahnaz ^{[1
]}

Zollei, Lilla ^{[2
]}

Grimson, W. Eric L. ^{[1
]}

Westin, Carl-Fredrik ^{[3
]}

Wells, William M. ^{[3
]}

机构：

[1] MIT, Comp Sci & Artificial Intelligence Lab, 77 Massachusetts Ave, Cambridge, MA 02139 USA

[2] Massachusetts Gen Hosp, Martinos Ctr Biomed Imaging, Boston, MA 02129 USA

[3] Brigham & Womens Hosp, Surg Planning Lab, Boston, MA 02115 USA

来源：

2008 IEEE INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING: FROM NANO TO MACRO, VOLS 1-4 | 2008年

关键词：

diffusion tensor MRI; clustering; anatomical information; tract-oriented quantitative analysis;

D O I：

10.1109/ISBI.2008.4540943

中图分类号：

R318 [生物医学工程];

学科分类号：

0831 ;

摘要：

We propose a Bayesian approach to incorporate anatomical information in the clustering of fiber trajectories. An expectation-maximization (EM) algorithm is used to cluster the trajectories, in which an atlas serves as the prior on the labels. The atlas guides the clustering algorithm and makes the resulting bundles anatomically meaningful. In addition, it provides the seed points for the tractography and initial settings of the EM algorithm. The proposed approach provides a robust and automated tool for tract-oriented analysis both in a single subject and over a population.

引用

页码：105 / +

页数：2