I just finished going thru the wiki's 'recommended' workflow with one of my subjects, which is just great !
It tells me to use tkmedit to check my wm and pial surfaces after autorecon2. I did and they looked pretty good (nothing caught my eye to have to fix). However when autorecon3 completed, I could easily see a couple little defects on the spheres.
1) should i try to fix those defects and re-do autorecon3 ? I imagine so, especially if I want to cut and flatten my surfaces. 2) how do i link tksurfer and tkmedit so that I may click on the sphere and see the corresponding location on the wm volume? I can't find that explained yet in the wiki.
thx! Mark
ps- using the latest dev version
Hi Mark,
there won't be any defects on the sphere once you've run the topology correction, and in general you never need to look at the spheres (does the workflow tell you to??). If the white and pial surfaces look good you should be all set (although you should look at the auto parcellation to make sure that the registration worked). To go to corresponding points click on save point and goto point in tksurfer/tkmedit.
Bruce
On Sun, 6 Nov 2005, Mark Pinsk wrote:
I just finished going thru the wiki's 'recommended' workflow with one of my subjects, which is just great !
It tells me to use tkmedit to check my wm and pial surfaces after autorecon2. I did and they looked pretty good (nothing caught my eye to have to fix). However when autorecon3 completed, I could easily see a couple little defects on the spheres.
- should i try to fix those defects and re-do autorecon3 ? I imagine so,
especially if I want to cut and flatten my surfaces. 2) how do i link tksurfer and tkmedit so that I may click on the sphere and see the corresponding location on the wm volume? I can't find that explained yet in the wiki.
thx! Mark
ps- using the latest dev version
No the workflow doesn't tell me to, I just was looking thru all the surfaces to see how they look. Yes the topology correction was run. So here's a snapshot of what the sphere looks like, and you can see something on the sphere surface, so I just thought it was a defect. thanks! mark
On 11/6/05, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Mark,
there won't be any defects on the sphere once you've run the topology correction, and in general you never need to look at the spheres (does the workflow tell you to??). If the white and pial surfaces look good you should be all set (although you should look at the auto parcellation to make sure that the registration worked). To go to corresponding points click on save point and goto point in tksurfer/tkmedit.
No the workflow doesn't tell me to, I just was looking thru all the surfaces to see how they look. Yes the topology correction was run. So here's a snapshot of what the sphere looks like, and you can see something on the sphere surface, so I just thought it was a defect. thanks! mark (not sure if attachment went thru, seems not).
Hi Mark,
which surface was that? In any case, I wouldn't worry about it. In the final analysis, the only 3 things you care about for cortex are:
1. Accuracy of the ?h.white surfaces 2. Accuracy of the ?h.pial surfaces 3. Accuracy of the cortical parcellation (which also tests the morph)
cheers, Bruce
On Sun, 6 Nov 2005, Mark Pinsk wrote:
No the workflow doesn't tell me to, I just was looking thru all the surfaces to see how they look. Yes the topology correction was run. So here's a snapshot of what the sphere looks like, and you can see something on the sphere surface, so I just thought it was a defect. thanks! mark
On 11/6/05, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Mark,
there won't be any defects on the sphere once you've run the topology correction, and in general you never need to look at the spheres (does the workflow tell you to??). If the white and pial surfaces look good you should be all set (although you should look at the auto parcellation to make sure that the registration worked). To go to corresponding points click on save point and goto point in tksurfer/tkmedit.
That was the right hemisphere sphere (tksurfer subjid rh sphere).
Yep all three things you listed look great, so I think I'm good.
thanks mark
Hi, I was just after some more pointers on using control points. Some gyri are missed in some of my images, and using control points recovers them, but it also tends to make the wm surface eat into the grey matter in some regions. Just wandering if there's anyway to prevent this from happening? Thanks, Alex
Alex Fornito M.Psych/PhD (clin. neuro.) candidate Melbourne Neuropsychiatry Centre and Department of Psychology The University of Melbourne alexander.fornito@wh.org.au
Hi Alex,
it depends on your images. What kind of coil were they acquired with? In general if a volume coil then placing control points in the wm at the base of the strand at least 1-2mm from the gray/white junction will fix this without eating into the gray. If you have a surface coil/phased array acquisition it may not be fixable so easily.
cheers, Bruce
On Tue, 8 Nov 2005, Fornito, Alexander wrote:
Hi, I was just after some more pointers on using control points. Some gyri are missed in some of my images, and using control points recovers them, but it also tends to make the wm surface eat into the grey matter in some regions. Just wandering if there's anyway to prevent this from happening? Thanks, Alex
Alex Fornito M.Psych/PhD (clin. neuro.) candidate Melbourne Neuropsychiatry Centre and Department of Psychology The University of Melbourne alexander.fornito@wh.org.au
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
Fornito, Alexander -
Mark Pinsk