Hello Kevin.
I have a bit of experience with this issue, I suggest that you use the -f
option ( -f <f> fractional intensity threshold (0->1); default=0.5; smaller
values give larger brain outline estimates) and I would set that to about
0.3 with dti images, you should also use the -m option which produces a bet
brain mask and also -n (to not generate an actual BETted dti images, you
could also use the -r option.
I hope this helps.
All the best,
Ahmed M. Radwan
On 25 Mar 2015 18:38, "Kevin Japardi" <[log in to unmask]> wrote:
> Hello everyone,
>
> I'm preprocessing my DTI data before feeding into DTIFIT & TBSS, and I'm
> coming across a small conundrum.
>
> Current EDDY pipeline (post-topup):
> 1) eddy --imain=${INPUT} --mask=hifi_b0_Tmean_brain_mask
> --acqp=acqparams.txt --index=index.txt --bvecs=dti.bvec --bvals=dti.bval
> --flm=quadratic --fwhm=0 --topup=topup_results --out=${INPUT}_eddy.nii.gz
>
> 2) bet ${INPUT}_eddy.nii.gz ${INPUT}_eddy_brain.nii.gz -R -F -m
>
> When I eddy-correct my DTI, it seems to work, and looks less distorted.
>
> The problem is: When I bet my dti_eddy results, it cuts off the most
> rostral (more posterior) tip/bits of my brain. It seems that the rest of
> the brain is properly bet-ed when comparing the before and after, but I was
> wondering if anyone had any ideas how to correct this issue? (Or possibly
> if there are better bet-ing processes to run on my eddy-corrected data?)
>
> Thank you,
> Kevin Japardi
>
