That is predominantly signal loss, from either air in the mouth or metal
as your T1w image is a gradient echo sequence. While there likely is some
distortion along the readout direction, it is much smaller than typical
phase encoding direction distortion as in EPI. You can still correct for
this with a field map (or a field map computed from topup) as the HCP
Pipelines do. Theoretically you could even reverse the readout direction
and run topup on the opposing high resolution images, however in practice
this is very slow as topup is not optimized for high resolution images and
also we found issues with fat shifting into the brain with a reversed
readout direction when we played around with this in the HCP Piloting. If
you want images of the mouth without dropout you would be better off using
a spin echo sequence like the Siemens T2w SPACE sequence, though you might
also need to use something like a bite bar to keep people from moving
their mouth during imaging.
On 8/7/17, 12:59 PM, "FSL - FMRIB's Software Library on behalf of John
anderson" <[log in to unmask] on behalf of
[log in to unmask]> wrote:
>Dear FSL experts,
>I have T1 images with a "susceptibility-induced distortions" in the mouth
>area (attached). I am wondering if "topup" works on T1 images. If yes
>what is the detain file in this case?
>I appreciate your feedback!