 |
 |
What pathways are you looking at, and have you tried finding them anatomically (i.e. placing ROIs on the color map where you expect the pathways to lie)?
Peace,
Matt.
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Cherif Sahyoun
Sent: Monday, August 04, 2008 9:56 AM
To: [log in to unmask]
Subject: Re: [FSL] Streamline numbers across subjects in probtrackx
Hey Tim,
Thanks for the tip.
The pathways of the problematic subjects are indeed thinner and noisier, as expected.
I played around with the mask size (made it crawl more into white matter), but although I get more streamlines, they don't increase nearly enough to catch up with other subjects (we're talking several thousands samples of difference).
Do you think it could be something like more motion problems in the "bad" subjects?
Looking at the FA map and the dtifit_V1 doesn't give me anything strikingly wrong.
I also loaded up the raw data in fslview and turned on the movie loop, but again, there doesn't seem to be more motion than in others.
I really doubt that in the pathways i'm looking at, the difference could be real, as most subjects have the same ballpark of numbers, except a couple of very low numbers and a couple of very high numbers.
Anything you can think of that i can troubleshoot to understand better what's going on?
Thanks,
Cherif
------------------------------------------------------------------------------------------
Cherif P. Sahyoun HST-MEMP
Developmental Neuroimaging of Cognitive Functions
C: 617 688 8048
H: 617 424 6956
[log in to unmask]
"Live as if this were your last day. Learn as if you'll live forever"
Ghandi
-------------------------------------------------------------------------------------------
On Tue, Jul 29, 2008 at 4:02 AM, Tim Behrens <[log in to unmask]> wrote:
Cherif - the best thing to do is to go and look carefully at the original pathways (fdt_paths).
This will give you a good feel for why your numbers are discrepant.
It may be casued by several different reasons (e.g. the pathway just misses the mask in some subjects; the pathway is much weaker in some subjects etc. etc. )
Cheers
T
On 28 Jul 2008, at 07:12, Cherif Sahyoun wrote:Hi,
Here's what I'm trying to do:
I resampled fMRI-based ROIs from a group omnibus contrast onto dti scans to
investigate the tracts underlying my functional networks. The DTI scans are
60 directions, b=700.
I ran BEDPOSTX and PROBTRACKX without errors, but it seems that between
equivalent pairs of ROIs (a seed and a target, registered onto all
individual DTI space), my waytotal varies between 18.00 and 30,000 depending
on subjects. What could cause such a discrepancy, given that the seed and
target ROIs seem to have registered fine and are approximately the same size
in all subjects?
The probtrack command I used for each subject looks like:
probtrackx --mode=seedmask -x $subject/dti/DTI_ROIs/lh.TPJ.label.nii -l -c
0.2 -S 2000 --steplength=0.5 -P 5000
--stop=$subject/dti/DTI_ROIs/lh_supPrecentralS.label.nii --forcedir --opd -s
$subject/dti/dti.bedpostX/merged -m
$subject/dti/dti.bedpostX/nodif_brain_mask
--dir=$subject/dti/dti.bedpostX/probtrackx_lhTPJ2lhsupPrecentral
--waypoints=$subject/dti/DTI_ROIs/lh_supPrecentralS.label.nii
Any help in troubleshooting or ideas welcome!
Thanks,
Cherif