I'm presuming you have 3 image spaces
1. Structural space where your DTI seeds/output are
2. Diffusion space
3. Freesurfer space with aparc2009 image
In freesurfer, convert .mgz to .nii:
mri_convert aparc2009.mgz parcellatedimage.nii
mri_convert orig.mgz structuralimage.nii
In SPM, use imcalc to created binarised image of your parcellated cortex
input = parcellatedimage.nii
Term: i1>0
Coregisister and reslice structuralimage to your defining seed space,
and take binarised parcellated cortex as extra image.
Now everything is in the same space just go back to imcalc and multiply
(i1.*i2) your coreistered binarised image by your tractography results.
An alternative approach would be to segment your seed structural image
and sum the white and grey matter images, or create binarised
thresholded images of your tractography results and move them to
freesurfer space (similar approach to above) which then gives you the
option of playing with meshes.
- Chris
>
>
> ---------- Forwarded message ----------
> From: *Saad Jbabdi* <[log in to unmask] <mailto:[log in to unmask]>>
> Date: Tue, May 18, 2010 at 8:39 AM
> Subject: Re: [FSL] white matter mask - Freesurfer
> To: [log in to unmask] <mailto:[log in to unmask]>
>
>
> Hi
> You might want to have a look at the latest FDT documentation under
> "surface tracking". You will find some useful instructions on how to
> take a volume from the freesurfer conformed space into diffusion space
> (1. Registering FreeSurfer conformed structural space to diffusion space)
>
> Cheers,
> Saad.
>
> On 18 May 2010, at 03:06, Matt Glasser wrote:
>
> > As more recent discussions have covered, using a white matter mask as
> > nodif_brain_mask was a hack that is not optimal. It is better to
> use the
> > inverse of the white matter as the stop mask (use fslmaths <wmmask>
> -sub 1
> > -mul -1 <invwmmask> to generate, use the flag --stop= in probtrackx).
> >
> > As for creating a NIFTI file white matter mask with the same
> dimensions and
> > orientation as your original structural image with FreeSurfer, this
> should
> > be easy to do. You might try asking that on the FreeSurfer list, as I
> > convert my FreeSurfer data to Caret formats before doing stuff like this
> > (which is hardly necessary in your case) so I don't actually know the
> > command to do this in FreeSurfer. As far as FSL goes, once you have the
> > white matter mask in structural space, you will need to provide a
> > transformation matrix between structural and diffusion space, and
> your seeds
> > will need to be in structural space to use this method. Otherwise,
> if your
> > seeds are in diffusion space, you will need to transform the mask into
> > diffusion space first, because the --stop mask needs to be in the
> same space
> > as the seed mask.
> >
> > Peace,
> >
> > Matt.
> >
> > -----Original Message-----
> > From: FSL - FMRIB's Software Library [mailto:[log in to unmask]
> <mailto:[log in to unmask]>] On Behalf
> > Of Theodor R?ber
> > Sent: Monday, May 17, 2010 6:15 PM
> > To: [log in to unmask] <mailto:[log in to unmask]>
> > Subject: [FSL] white matter mask - Freesurfer
> >
> > Dear FSL experts,
> >
> > I am trying to track the Pyramidal Tract in healhty
> > subjects, and found fibers escaping the brain as well as
> > fibers in areas with very low FA values. Going through
> > several discussions in the FSL archive, I surmised that
> > the best way to get rid of these unwanted fibers would be
> > a Freesurfer-generated white matter mask to be used as
> > nodif_brain_mask. Consequently, I generated a few white
> > matter masks with Freesurfer and found that these masks
> > were not only in a different space, but also has a
> > different x-y-z orientation from the normalized dti_FA.
> > What steps are necessary to turn the Freesurfer output
> > into a FSL-FDT compatible white matter mask?
> >
> > Thank you very much for in advance. Your help will be
> > greatly appreciated.
> >
> > Best,
> >
> > Theodor
> >
>
> --
> Saad Jbabdi
> University of Oxford, FMRIB Centre
>
> JR Hospital, Headington, OX3 9DU, UK
> (+44)1865-222466 (fax 717)
> www.fmrib.ox.ac.uk/~saad <http://www.fmrib.ox.ac.uk/%7Esaad>
>
>
|