Dear Ben,
> I am trying to run feat on the first run of a five run event-
> related design. I am getting a consistent error in the Registration
> step when the script attempts to register the high_res (my
> structural image) to the template. I've attached a picture from the
> feat output. My structural image is converted from dicom to ANALYZE
> format with an internally developed script at the imaging center
> where I collect data. Based on previous suggestions, I have
> converted the image to NIFTI format using SPM, fslchfiletype
> command, and AFNI's 3dAFNItoNIFTI command. In all cases I am able
> to convert the file to NIFTI, swap dimensions so that the structural
> image and my EPI images are in the same orientation upon display,
> but when I run feat I get this same poor registration. Any
> suggestions would be much appreciated.
if, as you say, your highres->standard_space registration fails, that
is not affected by the relative orientation of your EPI and your struct.
When that fails it is usually because
1. Your struct is in a very different orientation for the standard
space images.
or
2. Bet has failed badly, which might happen if e.g. you have lots of
neck in your structural image and the brain is only in the top half of
your volume.
So as a first step I would check those things.
Good luck Jesper
|