Hi,
You need to apply the same transformations to the FA image you used in tractography. If you have a linear transformation then you need to use flirt, something like:
flirt -in dti_FA -ref $FSLDIR/data/standard/MNI152_T1_2mm_brain.nii.gz -applyxfm -init diff2standard.mat -out FA_2_MNI
If you have a nonlinear transformation then you need applywarp, have a look at the documentation of these tools.
Cheers
Stam
On 11 Sep 2013, at 00:38, Dana Wagshal wrote:
> Hi FSL group,
>
> I've run probtrackx2 in an example subject to get a track that goes from ROI_1 -> ROI_2. However, my ROIs were in MNI space and so I clicked on seed space is not diffusion and chose non-linear and inputted the diff2standard.mat and standard2diff.mat files that I got from using Fdt's registration tab. Now I would like to get the mean FA from the track. However, when I run fslstats (fslstats dti_FA.nii.gz -k fdt_paths.nii.gz -M) I get an error message that that the masks are not in the same space. I realize it's because fdt_paths.nii.gz is in MNI space and dti_FA.nii.gz is in diffusion space.
>
> How would I convert fdt_paths.nii.gz from MNI space to diffusion space? Would I use flirt or fnirt? And what would be my reference for diffusion space, the subject's dti_FA.nii.gz file or the standard2diff.mat file found in the bedpostX/xfms directory? And would it look something like this?
>
> flirt -in fdt_paths.nii.gz -ref dti_FA.nii.gz -out fdt_paths_to_FA.nii.gz (not sure what should come after this...)
>
> fnirt --in=fdt_paths.nii.gz --aff=bedpostX/xfms/standard2diff.mat --cout=fdt_paths_to_FA.nii.gz --config=T1_2_MNI152_2mm
>
> Best,
> Dana
>
|