Hi,
I wanted to create ROI based on fmri co-ordinates. I have four ROI's. Below is the script that I made to make those. I am having some issues that relate to the final FA value selection. By the end, the ROI's are no more spherical as they are mapped to the structural white matter (which is fine till they are constrained to anatomy of the white matter).
a) When I pickup the values, for some ROI's I get the output as zero. Also how can I implement it further to select FA values >0.2.
b) It would be great if I can get a feedback on the steps and if any of them is unnecessary or any additional steps are needed.
while read subj; do
cd ${subj_roi}/FFA
3dUndump -srad 8 -prefix OUTPUTMASK.nii.gz -master ${standard}/resampled_MNI_template.nii.gz -orient LPI -xyz
${rois}/${subj}/${subj}_FFA_R_fs
# The fmri template is the one where i got the coordiantes from. So I did the first step within that dimension.
fslregister --s ${subj} --mov ${standard}/resampled_MNI_template.nii.gz --reg OUTPUTMASK_Native.dat
mri_vol2vol --mov OUTPUTMASK.nii.gz --o ROI_subj.nii.gz --reg OUTPUTMASK_Native.dat --fstarg --interp nearest
# above I registered the template to each participants structrual data
mri_convert wm.mgz wm.nii.gz
fslmaths wm.nii.gz -bin wmbin.nii.gz
# above created a white matter binarized mask for that participant
fslmaths ROI_subj.nii.gz -mul wmbin.nii.gz ROI_wm.nii.gz
# above i tried to anatomically constrain the mask to individual participants wm
bbregister --mov ${subj}_FA_to_target.nii.gz --s ${subj} --reg lowb2orig.dat --dti --init-fsl
mri_vol2vol --mov ${subj}_FA_to_target.nii.gz --targ ROI_wm.nii.gz --inv --interp nearest --o ROI_wm_faspace.nii.gz --reg lowb2orig.dat --no-
save-reg
# now registering the mask from the structural to diffusion space of the participant. the diffusion space is in the standard FMRIB (an output file from tbss step)
#fslmaths ROI_wm_faspace.nii.gz -mul ${standard}/fusiform.nii.gz roi_fusiform.nii.gz
# In the above code (which I did not implement as i was not sure if I am double dipping) I visually checked if my ROI is within the labelled area of say fusiform gyrus where I want my mask to be located. The labelled atlas was teh JHU_MNI white matter atlas. First I thought of getting only the overlay voxels as my final mask so that it is within the anatomical fusiform gyri. But then, it is quite possible to loose the individual anatomy map if I do that step. So I did not implement it.
fslmaths ROI_wm_faspace.nii.gz -bin roi_bin_ffa.nii.gz
fslmeants -i ${subj}_FA_to_target.nii.gz -m roi_bin_ffa.nii.gz -o rFFA_${subj}.txt
# binarized the ROI and the picked the mean values of all the voxels within the ROI.
done<$subj
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|