Image registration by local histogram matching
Pattern Recognition
Diffusion Tensor Image Registration Using Tensor Geometry and Orientation Features
MICCAI '08 Proceedings of the 11th International Conference on Medical Image Computing and Computer-Assisted Intervention, Part II
ISBI'10 Proceedings of the 2010 IEEE international conference on Biomedical imaging: from nano to Macro
| Hi-index | 0.00 |
Quantitative mapping of the relationship between brain structure and function has emerged as an active research field in order to answer the question of whether brain structure can predict its function. In the literature, it is widely believed that a brain region's structural connectivity pattern largely determines the function it performs. Based on this premise, we used multimodal diffusion tensor imaging (DTI) and task-based fMRI data to perform fMRI-guided DTI image registration and built structural fiber atlases for functional brain Regions of Interests (ROIs). First, we used workingmemorytask-based fMRI-derived ROIs as the functional constraint to register DTI images into a template via an energy minimization procedure. Then, the regularity and variability of the warped white matter fibers for each ROI was quantitatively assessed, and it turns out that structural connection patterns for corresponding functional ROIs across different subjects are quite consistent. Therefore, we constructed the white matter fiber atlases for these functional ROIs, which can be used as intrinsic attributes of those ROIs for quantitative representation of cortical architecture. Our results provided evidence that there is deep-rooted regularity of the common human brain architecture and that structural connectivity is strongly correlated with brain function.