Thank you Mark.
The colin_tlrc_1x1x1.nii.gz T1 appears to match Talairach space well:
the coordinates for the ventral thalamic nucleus in question seem to
spatially match the coordinates of same nucleus as indicated at
http://talairach.org/applet.html.
In addition , after the (FLIRT, linear) registration of the
colin_tlrc_1x1x1.nii.gz T1 with the MNI152_T1_1mm_brain, the
coordinates of the ventral coordinates are also spatially consistent.
The coordinates of the unregistered Talairach mask, viewed in fslview,
also correspond well to the points in Talairach space taken from
http://talairach.org/applet.html
The unregistered Talairach mask can not be overlaid on any image.
I will revisit this registration process again.
Best,
Charles
On Fri, Aug 12, 2016 at 4:41 AM, Mark Jenkinson
<[log in to unmask]> wrote:
> Hi,
>
>
>
> I think the quality here will depend on how well colin_tlrc_1x1x1.nii.gz
> actually matches the Talairach space. You have only done a linear
> registration, so this might not be that good, as the mapping between the
> spaces is not quite that simple.
>
>
>
> Have you looked at the thalamic nucleus mask on top of this image (prior to
> registration)? If that looks good and the registration of that image to the
> MNI looks good then things should be OK. So I suspect that one or other of
> these things is at fault.
>
>
>
> Alternatively, you might like to look at transformations that other people
> have investigated for mapping Talairach to MNI coordinates. For instance:
>
> http://www.brainmap.org/icbm2tal/
>
> or
>
> http://imaging.mrc-cbu.cam.ac.uk/imaging/MniTalairach
>
> though I'm sure there are others too.
>
> These have been fairly carefully looked at, so should probably work well.
>
>
>
> All the best,
>
> Mark
>
>
>
>
>
> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
> Charles Leger <[log in to unmask]>
> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
> Date: Thursday, 11 August 2016 at 22:12
> To: "[log in to unmask]" <[log in to unmask]>
> Subject: Re: [FSL] Talairach to MNI transform
>
>
>
> Recently, and following much appreciated input, I used the below call
>
> in FLIRT to calculate a Talairach to MNI transform; the initial step
>
> in transferring a thalamic nucleus mask created in Talairach space to
>
> MNI space
>
>
>
> flirt -ref $FSLDIR/data/standard/MNI152_T1_1mm_brain -in
>
> colin_tlrc_1x1x1.nii.gz -out higresTal_T12standard -omat
>
> higresTal_T12standard.mat
>
>
>
> However, my attempts to apply the transform matrix to the Talairach
>
> mask have resulted in inaccurate results; the coordinate results
>
> differ too widely from known coordinates from Talairach coordinates
>
> either converted to MNI coordinates with tal2icbm_fsl or to Talairach
>
> label coordinates in fslview MNI space.
>
>
>
> I have tried to follow directions outlined on the FLIRT FAQ
>
> http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FLIRT/FAQ#How_do_I_transform_a_mask_with_FLIRT_from_one_space_to_another.3F
>
> but must be making some errors.
>
>
>
> Any recommendations on how to apply the transform to the Talairach
>
> space mask would be much appreciated.
>
>
>
> Best,
>
> Charles
>
>
>
> On Tue, Aug 9, 2016 at 2:09 PM, Charles Leger <[log in to unmask]> wrote:
>
> Thank you Colin. This is very helpful...
>
>
>
> Best,
>
> Charles
>
>
>
> On Tue, Aug 9, 2016 at 1:04 PM, Charles <[log in to unmask]> wrote:
>
> FSL experts,
>
>
>
> Recently I posted a question asking how to transform a mask created in
> Talairach space to MNI space - all data in a current project will ultimately
> be in MNI space for initial analysis in FSL. Taliairach coordinates can be
> converted to MNI space and subsequently a sphere mask can be made in FSL but
> it would be preferable to convert to MNI space an actual Talairach space
> mask made for the structure in question (a thalamic nucleus); the thalamic
> nucleus in question is not currently available in MNI space.
>
>
>
> While the FLIRT (FNIRT) FAQ instructions
> (http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FLIRT/FAQ#How_do_I_transform_a_mask_with_FLIRT_from_one_space_to_another.3F
> ) work flawlessly for transforming masks between spaces, converting a mask
> from a different standard (Talairach) appears to involve more steps
> (FNIRT).
>
>
>
> Again, any suggestions would be much appreciated.
>
>
>
> Best,
>
> Charles
>
>
|