Hello,
I am trying to unwarp EPI images of rats acquired on a Bruker 9.4T scanner. Bruker have a product field-map sequence which produces a plausible looking field-map scaled in Hertz.
I have discovered that I only get sensible results out of fugue if I resample the field-maps to the same space as the EPI I want to unwarp. If I use the field-map as is from the scanner, fugue does not report any errors, but does very odd things to the EPI (nothing in some regions, over-correcting others).
Having to resample the field-maps seems a reasonable step, but despite searching I cannot find any mention of this in the documentation or any mailing-list postings. Hence I’m concerned I’m doing something wrong? There is also the additional problem that resampling the field-maps is tricky - rats have large signal voids on either side of the brain, and the resampling can produce erroneous results there. Hence if I do have to resample, I’d appreciate any advice on how to do this well.
Currently I do:
fugue -v --loadfmap=$FMAP --mask=$FMAP_MASL --despike --unmaskfmap --savefmap=$FMAP_DESPIKE
*resample with antsApplyTransforms and simple linear interpolation*
fugue --verbose --in=$FILE --loadfmap=$FMAP_RESAMP --unwarp=$UNWARPED \
--dwell=$UNWARP_ESP --unwarpdir=$PEDIR
Best wishes,
Toby
