Anna-- there's two ways to do this... low tech and high-er tech.
Honestly for 6 masks--- I'd load each mask... go to the PEN edit tool
and set it to ERASE, then increase the size of the "erase" box from 1
voxel to like 10-15... basically you can erase quite quickly using
this... and if you use pgup/pagedown you can really quickly just LOAD
ORIGINAl---> delete one side... save it as a new file name... and do
that 12 times.
The slightly higher tech version is below.
1) Start fslview, then load the MNI_152_2mm mask...
2) Create an ROI ( create mask)
3) Turn on the drawing tool... and do the same trick..... you should
see two "1"'s next to each other on the far right of the screen.. one
controls the "painted value" and the one of the farteher right
controls the brush size... make this 10-20 voxels.. then you can
basically quite quickly (on the sagittal view) just "paint" half the
brain (the entire LEFT half for example)... probably takes 2-3 minutes
to just mask out the entire Left hemisphere... then save this as
L_HEMI_MASK
If you want you can do the same thing for the right... or do
4) fslswapdim LEFT_MASK -x y z RIGHT_MASK
Since the MNI_152_2mm brain is symmetrical, this trick can "flip"
your mask for you
5) Now for your 6 or 8 ROI's in (noncorrect) bash syntax/pseudocode
fslmaths ${ROI} -thr-.5 -uthr+.5 -mas -mas L_HEMISPHERE ${ROI}_extracted_L
fslmaths ${ROI} -thr-.5 -uthr+.5 -mas -mas R_HEMISPHERE ${ROI}_extracted_R
On Thu, Oct 6, 2011 at 12:38 PM, Anna Zamm <[log in to unmask]> wrote:
> I would like to extract VOIs from the Juelich Atlas for use in whole-brain network analysis. I have written a very simple shell script that extracts each ROI on the basis of intensity value (fslmaths ${ROI} -thr-.5 -uthr+.5 ${ROI}_extracted). This works for every ROI, with the exception of six pairs of regions for which each pair shares an intensity value. These region-pairs are:
>
> IntensityV=9: GM Amygdala_centromedial group L AND GM Amygdala laterobasal group L
> IntensityV=19: GM Hippocampus entorhinal cortex L AND GM Hippocampus hippocampal-amygdaloid transition area L
> IntensityV=37: GM Inferior parietal lobule Pfm L AND GM Inferior parietal lobule Pga L
> IntensityV=47: GM Primary somatosensory cortex BA3b L AND GM superior parietal lobule 5M L
> IntensityV=54: GM Primary somatosensory cortex BA2 R AND GM Superior parietal lobule 7P R
> IntensityV=99: GM Primary motor cortex BA4a L AND WM corticospinal tract L
>
> Is there a command/set of commands I could use to separate these paired regions post-extraction so that I don't have to manually edit them? FSLsplit maybe?
>
> Any help would be much appreciated, and thanks in advance!
> Anna
>
--
David A Gutman, M.D. Ph.D.
Assistant Professor of Biomedical Informatics
Senior Research Scientist, Center for Comprehensive Informatics
Emory University School of Medicine
|