Mapping Structural Connectivity Using Diffusion <scp>MRI</scp> : Challenges and Opportunities

The effects of the misalignment between image-intensity-based tissue segmentation and brain parcellation on assignment streamlines to network nodes: Upper panel: (a) With application anatomical constraints (e.g., Ref. 35), streamline endpoints (colored in purple) occur at GM-WM interface or within subcortical GM. (b) Due factors such as discretization structural labels, many these (purple points) do not locate inside the, for example, FreeSurfer image (blue ribbon)48 thus are assigned a label. (c) T1 shown (a,b) is replaced by GM partial volume maps derived from intensity-based segmentation.80 (d) A zoom region illustrates discrepancy (pointed arrows) parcellation, revealing that cannot be purely based voxels where reside. Bottom discrepancies also present GM, fact could even crucial when images prepared transforming an atlas individual's space (i.e., left branch Fig. 5). This because degree segmentations parcellations increased registration error. As example: (e) AAL coregistered individuals' data via linear nonlinear transformations. background shows T1-weighted slice subject. (f) map obtained segmentation. (g) red arrows point considerable two thalami regions Chun-Hung 1666–1682.