Hi,
We see this sometimes when the original images (as viewed in
FSLView) are 90 degrees or more out of alignment. Typically
any change in the scanner orientation (sagittal, coronal, axial)
will do this, as can other things. If this is the case then changing
the search option to "Full search" should hopefully fix the problems.
If this doesn't fix it, or doesn't seem to be the problem then it is
probably best to upload an example case to our server and we'll
see if we can find the problem.
All the best,
Mark
On 31 Aug 2009, at 22:23, Lisa Eyler wrote:
> Hi Folks,
>
> We are puzzled by some registration errors that we are getting within
> MELODIC. When we look at the registration summary, we see two
> different
> types of errors that affect some but not all subjects. In each
> case, there
> seems to be a problem in the registration of the example_func to the
> highres, but no problem with registration of the highres to
> standard. Some
> of the example_funcs are reversed in the A-P plane compared to the
> highres
> and some are reversed in the I-S plane. Others, as I've said, are
> fine.
> When we compare the headers (either in AFNI or using fslhd) there
> are no
> differences that we can see in terms of the stated orientation of the
> subjects that worked vs. those that didn't. Also, ones that worked
> and ones
> that didn't all look the same when brought up in FSLView. Any
> thoughts
> about where things might be going awry and why it might affect some
> subjects
> but not others?
>
> Thanks,
> Lisa and Sharron
>
|