Dear fsl users,
I have used FNIRT to register an MRI image (IBSR data base) to the MNI space.
More specific
1) flirt -in IBSR_brain -ref /usr/share/fsl/4.1/data/standard/MNI152_T1_2mm_brain -out IBSR_brain_reg2MNIbrain_2mm.nii.gz -omat IBSR_brain_reg2MNIbrain_2mm.mat -bins 256 -cost corratio -searchrx -90 90 -searchry -90 90 -searchrz -90 90 -dof 12 -interp trilinear
2) fnirt --in=IBSR --aff=IBSR_brain_reg2MNIbrain_2mm.mat --config=T1_2_MNI152_2mm.cnf --cout=coef_IBSR --iout=fnirted_IBSR
The registered image looked fine when loaded in fslview.
Then I calculated the invert transformation matrix and applied that to the MNI152_T1_2mm_brain in order to bring it to the native space of the IBSR image
invwarp --ref=IBSR_01_ana.nii.gz --warp=coef_IBSR_01_ana.nii.gz --out=rev_coef_IBSR_01_ana
applywarp --ref=IBSR_01_ana.nii.gz --in=$FSLDIR/data/standard/MNI152_T1_2mm_brain --warp=rev_coef_IBSR_01_ana --out=MNI152_T1_2mm_brain_reg2_IBSR
In the later case though, when I loaded the MNI152_T1_2mm_brain_reg2_IBSR in fslview I saw a lot of errors including brain tissue far outside the scalp as you can see in the attached screenshot.
Are there issues related to the inverse consistency of FNIRT or it's just that I am doing something wrong here?
Thank you
Antonios Konstantinos
|