Thank you Doug.
I also just got help from the AFNI message board which solved the problem: so here is a note for AFNI/freesurfer users if they come across this too - if you run AFNI's @auto_tlrc command use the '-rmode NN' so that it does nearest neighbor resampling.
----- Original Message ----- From: Doug Greve greve@nmr.mgh.harvard.edu Date: Monday, July 28, 2008 1:37 pm Subject: Re: [Freesurfer] AFNI /freesurfer aseg files problem
Sorry, don't know anything about those commands. You're probably better off registering the functional data directly to the anatomical (fslregister or spmregister) rather than going thru the talairach transform. You can then use mri_vol2vol to map the functional to the anatomical, then mri_segstats to exctract the average functional value (or you can probably use the afni command at that point).
doug
Liat Levita wrote:
Hello Doug, I transformed the aseg_m file to be 3x3x3. I will try and trouble shot the possible error in that transform-
The AFNI steps I used to transform the freesurfer files to AFNI
format, and than to be 3x3x3 are pasted below.
Liat
foreach subj (xxx)
#step1 @SUMA_Make_Spec_FS -sid ${subj}
#step2
@SUMA_AlignToExperiment \ -exp_anat
/Volumes/cbgb/Liat/freesurfer_subjects/${subj}/SUMA/${subj}_anat+orig \
-surf_anat
/Volumes/cbgb/Liat/freesurfer_subjects/${subj}/SUMA/${subj}_SurfVol+orig \
-wd \ -prefix ${subj}_AlignExp
#step 3: To align a given parcellation file
3dWarp -NN -matparent ${subj}_AlignExp+orig -prefix ${subj}_aseg_m
aseg.nii>
#step 4: I transformed the functional data set by the talairch
transform obtained for the anatomical data, and then which was resampled to a 3x3x3 space during this step.
@auto_tlrc -apar ${subj}_anat_at+tlrc -input
${subj}_FC_Seed_aseg_fs_AMY_stats_ni_pi_Final+orig -dxyz 3 -suffix _at3>
end
----- Original Message ----- From: Doug Greve greve@nmr.mgh.harvard.edu Date: Monday, July 28, 2008 1:07 pm Subject: Re: [Freesurfer] AFNI /freesurfer aseg files problem
I would suspect that something happened when you converted to
afni
format. The aseg starts out at 1mm3, 256^3, but it appears to be 3mm3, 54x64x50. These conversions are quit tricky and easy to get
wrong.
Also, did you check the aseg.stats file to assure that the volumes of
the
amyg are more-or-less symetrical?
doug
R-to-L extent: -79.500 [R] -to- 79.500 [L] -step- 3.000
mm
[ 54 voxels] A-to-P extent: -79.500 [A] -to- 109.500 [P] -step- 3.000
mm
[ 64 voxels] I-to-S extent: -63.500 [I] -to- 83.500 [S] -step- 3.000
mm
[ 50 voxels]
Liat Levita wrote:
Hi Doug, I just checked, yes they are.
Liat
PS. I attached the Info files to this message.
----- Original Message ----- From: Doug Greve greve@nmr.mgh.harvard.edu Date: Monday, July 28, 2008 12:16 pm Subject: Re: [Freesurfer] AFNI /freesurfer aseg files problem
are you sure that t1rc and pscA are the same dimension?
Remember,
freesurfer resamples everything into 1mm3, 256^3
doug
Liat Levita wrote:
Hello, I wonder if anyone using AFNI converted freesurfer files has
come
across this problem (or can tell me if I making an error
somewhere
that I am just not seeing right now). I posted the same
question
on
the AFNI board but haven't gotten feedback as yet.
Here is the problem:
After using freesurfer to get subcortical segmentations, I
converted the resulting freesurfer files to AFNI format, which
was
fine - I had no problems doing that. However, when I try use
these
subcortical segmentation data files (subjectID_aseg_m_at3+tlrc)
as
masks to pull put PSC signal from each region of interest, I am getting an odd effect of right versus left; whereby left
clusters
output is fine ( voxel numbers are correct after 3dmaskave),
but
for the left it is very wrong. I have checked that the labels
of
the ROIs I am using are correct on AFNI GUI and they are. (code
and
output below)
I am a bit at a loss; any help will be much appreciated.
Liat
AFNI Code used
Foreach subj (xxx) foreach cluster (18 54) 3dmaskave -q -mask ${subj}_aseg_m_at3+tlrc -mrange ${cluster}
${cluster} \
"${subj}_all_runs_pscA_at3+tlrc" >
${subj}_fsROI_PSC_cluster_${cluster}.1D
end end
Output (get the same odd voxel output from all right hand side
of
the brain freesurfer generated ROIs)
*For cluster 18 Left Amygdala ++ 3dmaskave: AFNI version=AFNI_2008_02_01_1144 (Jul 3 2008)
[32-
bit]>>>+++ 37 voxels survive the mask
*For cluster 54 Right Amygdala ++ 3dmaskave: AFNI version=AFNI_2008_02_01_1144 (Jul 3 2008)
[32-
bit]>>>+++ 200 voxels survive the mask
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu 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
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu 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
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu 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