Print

Print


Hello. I still have a problem with FSL. My problem is converting my tracts into standard space. I tried flirt with both the GUI and the command line.
I typed the line:
flirt -in /Users/pspc5f/Desktop/sub2/STG2STG.nii.gz   -ref /Applications/fsl/data/standard/MNI152_T1_2mm_brain.nii.gz -out /Users/pspc5f/Desktop/sub2/STG2STG_MNI.nii.gz -applyxfm  -init /Users/pspc5f/Desktop/sub2/sub2.bedpostX/xfms/diff2standard.mat 

and Im getting a different tract than the one I made. Im not sure but it might be rotated in 90degrees on the brain. Sometimes it just look different. I don't find a way to convert successfully the tract to MNI space. When I try to draw the masks in MNI space and run probtractx, I get very weird results (I think because fsl still assumes that the masks were drawn on diffusion space). Please help. Oren

Date: Tue, 20 Sep 2011 11:56:41 +0100From: [log in to unmask]
Subject: Re: [FSL] registration of bedpostx to MNI space
To: [log in to unmask]

Hi, 

You need to specify the probtrackx input masks (seed, inclusion/exclusion) in the desired space (MNI in your case) and then results will be produced in this space, no need to register bedpostx results or data.
You further need to specify transformations between the desired and diffusion spaces. You can obtain those using the FDT registration tool. I would have a look at the FDT online documentation:

http://www.fmrib.ox.ac.uk/fsl/fdt/fdt_reg.html
http://www.fmrib.ox.ac.uk/fsl/fdt/fdt_probtrackx.html

Cheers,
Stam



On 20 Sep 2011, at 11:31, oren poliva wrote:

Hello. I have a problem with the registration to MNI space. I registered my no_diff_brain.nii.gz file (DTI) to MNI space, but when I use probractx it takes the data from the bedpostx directory, which wasn't registered in MNI space and create a tract that I can't overlay on my brain image. How do I convert the files in the bedpostx directory so they will create the tract in MNI space.
Just a side note - Im a new mac user (just came from windows) and everything is new to me, so explanation for beginners will be welcome.
Thank you for your time.
Oren