The non-binary values are doubtlessly due to using some sort of
interpolation. Use nearest-neighbor instead or threshold at 8.5. This is
probably the reason that it is too wide (which should probably also be
fixed by the thresholding). Alternatively, you can map the segmentation
directly into the 152 space instead of going thru the low-res functional
space. To do this, create a registration matrix between the standard 152
brain and your subject's anatomical, something like:
fsl_rigid_register -r subject/mri/brain.mgz -i
$FSLDIR/etc/standard/avg152T1_brain.img -o junk.mgh -regmat register.dat
-dof 12
This is a freesurfer script wrapped around flirt (-dof 12 makes it
non-rigid) You can delete junk.mgh
Edit the first line of the register.dat to be that of the name of the
subject.
Check/adjust the registration and save as an fsl matrix:
tkregister2 --s subject --mov $FSLDIR/etc/standard/avg152T1_brain.img
--reg register.dat --fslregout fsl.mtx
Use fsl.mat to resample your hippo mask to standard space (note: you may
need to invert fsl.mtx)
doug
Paymon Hosseini wrote:
> Dear Steve and Ged,
>
> this is the solution that settled on for making masks for the
> hippocampus for my subjects. i am wondering if you can tell me what
> you think of it. also, i have run into a new problem. i will greatly
> appreciate any advice on how to solve it.
>
> here is what i did:
>
> i ran the structural runs of a subject through Freesurfer to get the
> automated probabilistic cortical and subcortical segmentation maps for
> this subject.
> Then i followed the steps under "CORTICAL AND SUB-CORTICAL
> SEGMENTATIONS" (http://www.fmrib.ox.ac.uk/fsl/freesurfer/index.html)
> to generate binary masks for this subject's hippocampus to use in FSL.
>
> running
>
> avwmaths featdir/reg/freesurfer/aseg.nii.gz \
> -thr 17 -uthr 17 \
> featdir/reg/freesurfer/lh.hippocampus.nii.gz
>
> will give me a mask in the space of my functional runs. this mask is
> either 0 or 17 (0 being background and 17 being hippocampus). all is
> good so far.
>
> next, i tried to re-sample this mask to the highres space using flirt
> and example_func2highres.mat. the re-sampled mask has two problems: 1)
> it becomes too wide in x, y and z and 2) it has continuous values from
> 0 to 17. i cannot just threshold the new mask at 17, because i will
> miss major parts of the hippocampus, and i cannot use it because it is
> too wide. any suggestions on how i can solve these two problems?
> thanks
> paymon
>
> Ged Ridgway wrote:
>
>> Hi Paymon,
>>
>> As Steve implies, I don't think maps yet exist for avg152, however,
>> you can get some probabilistic maps for ICBM452, which may include
>> some structures of interest to you...
>>
>> http://www.loni.ucla.edu/ICBM/ICBM_Probabilistic.html
>>
>> Also, I believe people have got approximate masks for e.g. the ICBM
>> single subject template (AKA Colin27), though these aren't the same
>> as probabilistic population templates, they may be all you need for a
>> basic and rough ROI analysis. See e.g. Tzourio-Mazoyer et al's AAL
>> stuff for SPM (should be pretty easily converted to separate NIfTI ROIs)
>>
>> http://www.cyceron.fr/freeware/
>>
>> Hope that helps,
>> Ged.
>>
>> Steve Smith wrote:
>>
>>> Hi - we will be generating such probabilistic maps in the near
>>> future, watch this space.
>>> Cheers.
>>>
>>> On 22 Nov 2006, at 22:25, Paymon Hosseini wrote:
>>>
>>>> Dear FSL users,
>>>>
>>>> i am wondering if there are structural masks for the standard brain
>>>> (avg152T1_brain) in FSL. For example, is there a mask for the
>>>> hippocampus or the amygdala?
>>>>
>>>> thanks a lot,
>>>> paymon
>>>
>>>
>>>
>>> ---------------------------------------------------------------------------
>>>
>>> Stephen M. Smith, Professor of Biomedical Engineering
>>> Associate Director, Oxford University FMRIB Centre
>>>
>>> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
>>> +44 (0) 1865 222726 (fax 222717)
>>> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve
>>> ---------------------------------------------------------------------------
>>>
>>>
>
>
--
Douglas N. Greve, Ph.D.
MGH-NMR Center
[log in to unmask]
Phone Number: 617-724-2358
Fax: 617-726-7422
In order to help us help you, please follow the steps in:
surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
|