Our FAQ is a little out of date on this issue.
Currently FSL (4.1) should work fine with any
mix of neurological and radiological images.
So you should be fine as long as everything was
run with FSL 4.1. If you used 4.0 then it may have
worked but we would not be as confident and
it would be safer to run again with 4.1.
All the best,
Chris Davey wrote:
> Hi All
> I am doing some fMRI data analysis with the latest version of Feat. I thought today to check
> the orientation of my images, and was alarmed to find that the structural images are in
> neurological orientation, while the rest (fieldmaps and EPIs) are in radiological orientation.
> The DICOMs come from a Siemens Magnetom Trio, and I have used the dcm2nii program on
> a Mac to convert them (using the default settings) to 4d nifti files.
> I can see in the FAQ that it is strongly recommended that all images be in radiological
> order. Should I use the commands listed in the FAQ to switch the orientation of the
> structural images? Or does Feat/FNIRT recognise the orientation of the images and deal with
> them appropriately? I have done first level analyses on about 30 participants now, and am
> loath to do them again since each analysis takes about 2.5 hours. But obviously, if it is
> recommended I should . . .