Hi,
I don't think that this work-around that you are doing is likely to give very good results.
I think you need to sort out why the standard registration procedure is not working.
Did you brain extract your structural image?
Does it have a lot of neck (in which case using robustfov or fslroi can help)?
I think once you've got the standard registration to work then everything else will be better.
All the best,
Mark
On 3 Dec 2013, at 09:41, Jenny Mercolini <[log in to unmask]> wrote:
>
>
> Dear experts,
>
> I have run melodic ICA with temporal concatenation for the first time.
> I have had several problems with the registration part. To solve this I chose to only register the epi to the bg_image, and skip the initial structural image and the main structural image registration. This worked well. Is there any major problems with doing it this way?
>
> There were also a couple of subjects that that I never managed to register. I excluded these just to see if the group analysis would run. ( and it did). But I would like to registerthese subjects' epi's to the template and add them back into my Melodic group analysis.
> Is there any way to do that?
> So far I have only used the GUI.
>
> I am thankful for any tips or replies :)
>
> Kind regards,
> Jenny
>
|