Cortical thickness measurements in BrainVisa is restricted to a sulcus, does this mean that sulcal _basins_ are labelled in concordance with the sulcal labelling of BrainVisa and then the average Laplacian-based thickness is computed over all points of each labelled basin? Or is the measurement restricted to the projection points of a given median mesh over the cortical surface? Any information on how the cortical thickness measurement is restricted to a given sulcus is appreciated (or a reference explaining it).
Merci
cortical thickness along a sulcus
- riviere
- Site Admin
- Posts: 1361
- Joined: Tue Jan 06, 2004 12:21 pm
- Location: CEA NeuroSpin, Saint Aubin, France
- Contact:
Re: cortical thickness along a sulcus
Hi,
The cortical thickness in sulci graphs is currently not computed using a Laplacian, but a distance map on cortical segmentations (in voxels). Thus it has a very low precision on a given point and has to be averaged. In sulci thickness is averaged in sulcal regions (Voronoi from the bottom lines of sulci). You cannot expect a high precision thus because sulcal regions include portions of cortex of different nature, different curvature... These measurements are meant to distinguish large trends, not much more. We have to replace this method by a more precise one (Laplacian would be a good candidate), and limit the averaging to "reasonable" regions (typically not including folds bottoms with the highest curvature). This has been in the TODO list for long, but up to now we have not taken the time to actually do so.
Otherwise in the cortical surface toolbox, there are some experimental processes to process vertex-wise thickness maps and median meshes. They are still "experimental" because they have never been really validated and compared carefully to what Freesufer (typically) does...
Denis
The cortical thickness in sulci graphs is currently not computed using a Laplacian, but a distance map on cortical segmentations (in voxels). Thus it has a very low precision on a given point and has to be averaged. In sulci thickness is averaged in sulcal regions (Voronoi from the bottom lines of sulci). You cannot expect a high precision thus because sulcal regions include portions of cortex of different nature, different curvature... These measurements are meant to distinguish large trends, not much more. We have to replace this method by a more precise one (Laplacian would be a good candidate), and limit the averaging to "reasonable" regions (typically not including folds bottoms with the highest curvature). This has been in the TODO list for long, but up to now we have not taken the time to actually do so.
Otherwise in the cortical surface toolbox, there are some experimental processes to process vertex-wise thickness maps and median meshes. They are still "experimental" because they have never been really validated and compared carefully to what Freesufer (typically) does...
Denis