Dear list,
I am trying to use fugue to unwarp my EPI volumes to align them as good as
possible with a high-resolution MPRAGE structural volume that I use
reconstruct the corical surface with freesurfer.
I am pretty confident that my field maps are correct (the two sets of
volumes are aquired with a TE of 7ms and 9.24ms, so the asymmetric spin echo
time should be 2.24 ms. And I've used both fugue with the --asym option, and
avwmaths according to the fugue tutorial to get the same field map).
The dwell time that I am using is from the DICOM header field 0043,102C,
and is 552 microseconds in our case. That gives dwell-to-asym-time ratio
of .2464, which is in the middle of the range according to the
website of .2 to .3.
While the unwarping direction that I am using is correct for each data set
(I have some volumes with the phase encoding direction L-R, and some with
A-P), the unwarping seems to be just a bit too strong i.e. the results seem
better if I use a dwell time that is a bit shorter. The effect is pretty
subtle, but clearly visible.
We are using ramp sampling in our EPI sequence, and I am suspecting that
perhaps the calculation of the dwell time displayed in the header does not
take the ramp sampling into account, and thus gives a value that is larger
than the effective dwell time. Is this possible, and has anyone else
experienced this? Or has anyone else experienced that FUGUE gives unwarping
results that are just a bit too strong? Our scanner is a GE Signa 3.0T.
Thank you very much for your help.
Sincerely,
Valentin
|