Thank you Steve.
Looking at my DTI processing scripts, I'm using the "hifi_b0_Tmean_brain_mask" output from topup (topup >> fslmaths >> bet) as the --mask input when running Eddy (and dtifit).
What I noticed is that these masks look blown up in comparison to the primary dti.nii.gz input for Eddy/dtifit. Do you think this bloated mask is what's causing the larger-than-what-should-be bet results?
Additionally, do you know if the actual 4D image file with unwarped images (--iout) is used in any of the processing functions?
Best,
Kevin Japardi
|