Hi Anastasia,
I saved out some of the completed trac-all.cmds following each TRACULA
'step', and I see where aparc+aseg is mapped to diffusion space. Thank you
again for all your help - my project is starting to look a lot better!
Sincerely,
Susie Kuo
NIH
On Wed, Mar 27, 2013 at 12:02 PM, Anas
Actually the brain mask used on the diffusion data *is* the aparc+aseg
mapped to diffusion space, so the reason you see that extra non-brain
stuff in the FA map is because the aparc+aseg was not aligned well to it.
So you just need to switch on bbregister and switch off flirt in your
configu
Hi Anastasia,
Thank you! I can see that the aparc+aseg spilled over badly. I am going
to try bbregister, and then re-betting (Brain Extraction Tool-ing) the
original image, since the dtifit_FA looks to include non-brain near the
frontal lobe. I'll keep you and the FS community updated on the resu
Hi Susie - The -tv (tract volume) option of freeview is designed
specifically to display the merged*.mgz output files of tracula. For a
regular volume, use -v or no option at all. For best displaying the
aparc+aseg, select it in freeview and choose "Lookup Table" from the Color
map menu.
Ho
Hi Anastasia,
I'm attaching the graphic I arrived at with the command:
> freeview -tv /dlabel/diff/aparc+aseg.flt.nii.gz -v
/dmri/dtifit_FA.nii.gz
Here, I used the subject I send you (Sa69845.5) for the subjectID. I could
not detect spill over in these images, mostly because the frontal lobe is
Hi Susie - Everything under dlabel/diff is mapped to DWI space (either
with flirt or with bbregister, whichever you use). The one in my
screenshot was dlabel/diff/aparc+aseg.flt.nii.gz.
a.y
On Wed, 27 Mar 2013, Susan Kuo wrote:
Hi Anastasia,
I see what you mean. I had previously used th
Hi Anastasia,
I see what you mean. I had previously used the dtifit_FA.nii.gz image and
overlaid that with the aparc+aseg+2mm.nii.gz image in /dlabel/diff, using
freeview. I didn't end up with the same images as you, however. Can you
tell me which images you used to obtain this overlay?
I am
Hi Anastasia,
I did as you recommended and checked the diffusion-to-anatomical
registration, and the overlay of aparc+aseg_mask on FA, and these views
seem to be good. Upon closer inspection, what I find is that there are
incipient 'bits' of all the tracts, but they seem to not have 'grown',
thou
Hi Susan - Good to hear that you get good results for most of your
subjects. Have you checked the aparc+aseg and the diffusion-to-anatomical
registration for the subjects that are failing? I'd check the
aparc+aseg_mask (in the dlabel/diff directory) over the FA map to see if
there are any hol
