Identifying white-matter fiber bundles in DTI data using an automated proximity-based fiber-clustering method.
Identifying white-matter fiber bundles in DTI data using an automated proximity-based fiber-clustering method.
复制标题
使用基于自动接近的纤维群集方法在DTI数据中识别白色纤维束。
DOI：
10.1109/tvcg.2008.52
复制
发表时间：
2008-09
影响因子：
5.2
通讯作者：
Laidlaw DH
中科院分区：
文献类型：
--
作者：
Zhang S;Correia S;Laidlaw DH
We present a method for clustering diffusion-tensor imaging (DTI) integral curves into anatomically plausible bundles. An expert rater evaluated the anatomical accuracy of the bundles. We also evaluated the method by applying an experimental cross-subject labeling method to the clustering results. Our approach is guided by assumptions about the proximity of fibers comprising discrete white-matter bundles and proceeds as follows: We first employ a sampling and culling strategy for generating DTI integral curves and then constrain the curves so that they terminate in gray matter. This approach seems likely to retain anatomically plausible fibers. We then employ a clustering method based on a proximity measure calculated between every pair of curves. We interactively selected a proximity threshold to achieve visually optimal clustering in models from four DTI data sets. An expert rater then assigned a confidence rating about bundle presence and accuracy for each of the 12 target fiber bundles of varying calibers and types (i.e., commissural, association, and projection) in each data set. The interactive clustering and evaluation information was incorporated to create a fiber-bundle template. We then used the template to cluster and label the fiber bundles automatically in new data sets. According to expert evaluation, the automated proximity-based clustering and labeling algorithm consistently yields anatomically plausible fiber bundles, although fiber bundles with smaller calibers and those that are not highly directionally coherent are identified with lower confidence. This work has the potential to provide an automatic and robust way to find and study neural fiber bundles within DTI.
