Hi
I suspect you directly registered the thresh_zstat* to the reference
image -- I wouldn't expect that to work well at all. Try registering an
example epi to highres first (use the -omat option to output the
matrix) and then use the pre-calculated registration to transform the
IC maps into high res space.
hope this helps
Christian
On Tuesday, Jun 3, 2003, at 20:57 Europe/London, Andrew Bleicher wrote:
> Hi,
>
> Thank you both very much for your quick replies. I have tried both
> methods, which provided similar results. However, I have come to
> realize that MELODIC does not include a registration step so that the
> functional data are not registered to standard space or to a high
> resolution image. As such, I tried to run the FLIRT GUI to register
> the thresh_zstat*.img to a high resolution image, but the results were
> not to my liking. Perhaps there is some other way that I ought to be
> approaching this problem.
>
> While on the subject of AFNI, I have unable to use certain tools such
> as calculating % change and using AFNI's Talairach coordinate system
> with my FEAT output. Also, when constructing timecourse graphs, AFNI
> uses time instead of volumes to label its x axis. This has caused my
> graphs of signal timecourse to be shifted in time to the left, so that
> responses "anticipate" the stimulus. I am not sure why this happens,
> and I have had difficulty quantifying the degree of "anticipation" to
> correct the labels I add to these graphs afterwards.
>
> I understand that these are AFNI questions as much as FSL questions,
> but I would appreciate any help that is offered. I had hoped that the
> newest version of AFNI might have made these steps easier, but it
> hasn't.
>
> Thank you again,
>
> Andrew Bleicher, MD
> Resident
> Department of Radiology
> Montefiore Medical Center
> Albert Einstein College of Medicine
> Bronx, New York
|