Hi Ben
I'm sure someone will correct me if I'm wrong, but I don't think it
will matter. The difference in alignment for each scan relative to
the other ones in the series should be the same, even if the
difference relative to the first scan is different----i.e., when
looking at your realignment plots in SPM, the values along the y-axis
might be different, but the overall profile (and thus the effect of
the regressor) shouldn't change.
(However, if you want to save a copy of the original realignment
parameters, simply copy the rp_*.txt file from each directory to
another file so that it doesn't get overwritten when you re-run
realign.)
Jonathan
On Mon, Dec 15, 2008 at 4:35 PM, Xu, Ben (NIH/NINDS) [E]
<[log in to unmask]> wrote:
> Hi, Doug,
>
> If I pick a volume from the previous session, how do I preserve the movement
> parameters for the new session for use later as movement correction
> regressors? Wouldn't it over (or maybe under) correct for movement as the
> realignment parameters are based on a previous image, not the first image
> within the session?
>
> Thanks, again.
>
> Ben
> ________________________________
> From: Doug Burman [mailto:[log in to unmask]]
> Sent: Thursday, December 11, 2008 10:14 AM
> To: [log in to unmask]
> Subject: Re: [SPM] Realign images from separate study sessions
>
> Ben,
>
> The realignment procedure always moves the subsequent images into alignment
> with the first image selected. Simply select one of the previously-aligned
> images (or the mean image from that alignment) as the first image, followed
> by the set of new images.
>
> Doug
>
> ==============Original message text===============
> On Thu, 11 Dec 2008 8:18:23 am CST "Xu, Ben (NIH/NINDS) [E]" wrote:
>
> Hi,
>
> I have a set of EPI scans and would like to realign them to the scans
> from the same subject acquired previously. These previous scans have
> already been realigned with spm5. Is there a way to realign the new
> scans to the previous ones without re-doing the realignment of the
> previous scans? Is it right if I realign the new scans first and, then
> realign them again with the mean image of the previous scans?
>
> Thank you for your help!
>
> Ben
>
> ===========End of original message text===========
>
>
>
>
|