Hi. I was trying to prepare my data for dtifit and have seemed to run into a
problem and was looking for some advice.
I have 4 averages consisting of 1 b0 and 20 gradient directions (b1000),
before averaging these together to make one large DWI 4D file, I was looking
to strip each of the 4 b0 images (using BET), align each b0 image to the first
b0 and then apply those parameters to the rest of the gradient directions
(using FLIRT and applyxfm4D). (See script below).
While BET seems to have correctly stripped each b0, after running flirt to align
the b0 and apply it to the rest of the gradient directions, my output file
(${subj}.${DTI}.FG.EC.CR.nii.gz) has large pieces of skull around the frontal
lobe and even the eyesocket/ball, while the superior, posterior, and inferior
brain areas seem to be fully stripped. Any ideas what I am doing wrong, or
how to fix it?
Thanks in advance!
-Megan
Below is the part of my script I seem to be having trouble with...
foreach DTI ( ${runnames} )
echo +_+ "BET ${DTI}"
bet ./${subj}.${DTI}.FG.EC.nii.gz ./
${subj}.${DTI}.FG.EC.STR.nii.gz -m -f 0.3
end
#Co-register stripped averaged 1st b0 to stripped 2nd-4th averaged b0s
foreach DTI ( ${DTIs[2]} ${DTIs[3]} ${DTIs[4]} )
echo +_+ "Coregister ${DTI} to ${DTIs[1]}"
flirt -in ${subj}.${DTI}.FG.EC.STR.nii \
-ref ./${subj}.${DTI[1]}.FG.EC.STR.nii.gz \
-omat ${DTI}.to.${DTIs[1]}
#Apply b0 co-registration parameters to the rest of the gradient directions
applyxfm4D ./${subj}.${DTI}.FG.EC.nii.gz \
./${subj}.${DTI[1]}.FG.EC.STR.nii.gz \
./${subj}.${DTI}.FG.EC.CR.nii.gz \
${DTI}.to.${DTIs[1]} -singlematrix
end
|