Dear Max,
there are two possibilities:
1) The first scan should be treated as dummy scan - unless you run extra dummy scans prior to the experiment
2) You or someone did a coregistration, using the first images a source and not as reference image.
Good luck,
Karsten
--
-------------------------------------------------------------
Karsten Specht, PhD
Department of Biological and Medical Psychology
Bergen fMRI group
University of Bergen
Jonas Lies vei 91
5009 Bergen
Norway
Tel.: +47-555-86279
Fax: +47-555-89872
[log in to unmask]
http://fmri.uib.no/
-----Ursprüngliche Nachricht-----
Von: "Max Hilger" <[log in to unmask]>
Gesendet: 03.jun.2011 18:21:11
An: [log in to unmask]
Betreff: [SPM] SPM8: Artificial movement at the beginning of a session?
>Dear SPM-Users,
>
>we are currently reanalyzing a fMRI-dataset in which all subjects did multiple conditions in the same succession.
>
>The first step of the analysis was realigning and unwarping the scans. The graphs showed heavy movement (up to 15mm) of all the subjects, but ONLY between the first and second scan of the first condition. According to the graphs, all subjects moved in negative-X-direction. There were also rotations in different directions, but again only between the first and second scan. It seems implausible that all the subjects really moved in the same direction at the same point of time, so we do assume that there is an error in our data set/ analysis that causes this pattern of results.
>
>Since the data was previously analyzed with spm2 (using the same parameters), we compared the results with the allready existing ones. In the image attatched you can see the results for two conditions of the same subject which are characteristic for all the subjects. The graphs to the left are from the spm8 analysis, the ones to the left from the older spm2 analysis.
>As you can see, the graphs don't differ for condition 004, but for condition 002 there is heavy movement between the first and second scan. After this, the graphs of the spm2 and spm8 analysis are very similar again.
>
>We tried to work around the problem by simply excluding the first image from analysis (and reducing the onsets of the stimuli etc by TR), but we are not sure if this is an adequate solution.
>
>The parameters we used for the analysis were:
>
>Estimation Options
>. Quality: 1
>. Interpolation: 7th Degree B-Spline
>. Wrapping: Wrap Y
>Unwarp Reslicing Options
>. Interpolation: 7th Degree B-Spline
>. Wrapping: Wrap Y
>All other parameters were set to default.
>
>So I would very thankfull if you could provide insight in the origin of this artifact and help me to solve this problem. Also I would like to know if our attempt to work around it is appropriate. I can of course also provide you with additional information, if needed.
>
>Thanks in advance for any input!
> Max Hilger
>
|