Besides just looking at the realignment parameters you should probably look at the images themselves. If this is a continuous movement then compare the first to the last image before and after realignment. You could also make a movie clip. Look at AnalyzeMovie -- I wrote it for spm99, but you can easily modify it to work with other spm versions.
But for it to be the scanner would require something to be wrong with the gradient system (this could be hardware and/or software). Run a phantom that you know does not move.
Move likely it's the participant.
Good luck,
Robert
-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Robert C. Welsh, PhD
Research Assistant Professor
Departments of Radiology and Psychiatry
University of Michigan
(734) - 764 - 8541 (fax)
(734) - 647 - 6781 (Ofc)
rcwelsh @ med.umich.edu
>>> Torben Ellegaard Lund <[log in to unmask]> 01/03/10 7:35 AM >>>
Dear Wei
This could be caused by a number of factors, e.g. it could be a problem with the sequence or the positioning of the subject. In order to find out about the origin, a first thing to do could be to find out if the behavior is specific to this one subject, or if it happens to other images using the same sequence/scanner.
Best
Torben
Den 03/01/2010 kl. 03.09 skrev Huang, Wei (Psychiatry):
> Dear All,
>
> I've done realignment (estimate and reslice) on 246 EPIs (taken in one
> continuous scan), and the result showed a total 7mm continuous movement
> in the y direction (see attached 1st.fig), which led me doubt the
> quality of images and/or realignment. I went ahead to do another
> realignment based on the already realigned 246 scans and got the result
> of very small movement in all directions (see attached 2nd.fig), which
> made me believe the 1st realignment job was fine. However, I have never
> got such big continuous translation in one direction. Is this set of
> images reliable? Is there another way to check if anything is wrong
> with my images?
>
> Thanks,
> Wei
> <1st.fig><2nd.fig>
**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues
|