On 10/14/2010 10:05 AM, Bruno L. Giordano wrote: > Hi, > > I have a few unrelated questions: > > 1. I have a surface representation of subject A, and want to project a > functional. Does it help or is it pointless to set the AC origin in the > functional before projecting it onto the surface? > The surface's origin should match the functional's origin. If the surface's origin is the brainstem, setting the functional's origin to the AC will hurt. If the surface's origin is the AC, then setting the functional's origin to match will help. > 2. I would like to use a few more landmarks to guide the spherical > registration process: is it enough to specify them in both the reference > (atlas) and target (subject) border files? Yep. Make sure they're named identically across borderproj files. > Will the registration > procedure automatically take into account the additional landmarks > loaded in the border files of both the reference and target? > Yep. > 3. I would need a few guidelines on how to proceed when creating group > representations from those of the (registered) subjects. For example, > what can I do with the average coordinates of the registered fiducial of > my subjects? Surface: Create Average Coordinate File > Is it necessary to create an average topology to properly > visualize them in caret? No, use whatever was the target topo file during registration. > If so, how? Is it possible (and does it make > sense) to generate the sulcal depth from a fiducial surface only (e.g., > average coordinates)? Possible, yes. Does it make sense? Not really. Better to average the deformed shape files from the contributing subjects. > If yes, how? Can I calculate some "confidence" > borders for the average of the borders of the subjects? You can unproject the borders on the aligned spheres of the contributing subjects, and view them on a target sphere, such as the PALS-B12 atlas target. You can even hide the surface, so all you see is the spray of borders -- no surface. > In sum, is there > some tutorial for creating atlases with caret? > Not to my knowledge. We avoid making separate atlases wherever possible, but do so when necessary. For example, we found it necessary for studying babies, because their brains were too different from adults' brains. But we found the brains of seven year old kids to be similar enough that we didn't need to create a separate atlas.
Now we distinguish creating, say, average fiducial surfaces from creating atlases. One can generate the former without generating the latter, which is more involved. It would help if I knew more about what you're trying to do. > Thank you! > > Bruno > _______________________________________________ > caret-users mailing list > [email protected] > http://brainvis.wustl.edu/mailman/listinfo/caret-users > _______________________________________________ caret-users mailing list [email protected] http://brainvis.wustl.edu/mailman/listinfo/caret-users
