I’m trying to figure out how to get a fast dump out (text) of the MNI coordinates and value of the max voxels within a spherical ROI (e.g. 2mm radius, centered on a meta-analytic maxima coordinate.) This is because I want to address individual differences in precise spatial localization of what is essentially a canonical activation. 3dExtrema and 3dMaxima seem candidates, but I’m not seeing anything on a spatial mask. Would the mask just BE the data-set invoked? What would be a good quick script/strategy?
From how you described your question, I was assuming that you have the list of “x y z” coordinates in a certain orientation defined already (your “meta-analytic maxima coordinate(s)”). I call this file “coor_centers.txt”, below. And you can put in the name of the file whose contents you are probing in this variable (in whatever space):
“set vmast = DSET_OF_INTEREST”
One important assumption is that the spheres are not contiguous/overlapping/touching. Here, only one max would be found for such a “chain” of spheres.
#!/bin/tcsh
set coor_sph = coor_centers.txt
set vmast = DSET_OF_INTEREST
# make binary mask of *isolated* spheres centered around ${coor_sph} locations
3dUndump \
-overwrite \
-xyz \
-orient RAI \
-prefix f00_roi_sph.nii.gz \
-master "${vmast}" \
-datum byte \
-srad 2.1 \
${coor_sph}
# multiple that mask of spheres times the dset of interest-- i.e., punch out sphere regions from that dset of interest
3dcalc \
-overwrite \
-a "${vmast}" \
-b f00_roi_sph.nii.gz \
-expr 'step(b)*a' \
-prefix f01_dset_sph.nii.gz
# for each isolated sphere, make a map of single maximum value
3dROIMaker \
-overwrite \
-inset f01_dset_sph.nii.gz \
-prefix f02_dset_sph_max \
-only_some_top 1 \
-nifti
# dump the max locations+values to a text file
3dmaskdump \
-xyz \
-noijk \
-mask f02_dset_sph_max_GM.nii.gz \
"${vmast}" \
> f03_sph_max_loc.txt
Thanks folks! No shortage of approaches here. -Jim
The
National Institute of Mental Health (NIMH) is part of the National Institutes of
Health (NIH), a component of the U.S. Department of Health and Human
Services.