| Original-Sender: [log in to unmask]
| Date: Tue, 26 Oct 1999 12:36:33 -0400 (EDT)
| X-Authentication: IMSP
| MIME-Version: 1.0
| Subject: Analysis of multislice fMRI data.
| From: [log in to unmask]
| To: [log in to unmask]
| X-List: [log in to unmask]
| X-Unsub: To leave, send text 'leave spm' to [log in to unmask]
| X-List-Unsubscribe: <mailto:[log in to unmask]>
|
|
| Hi,
|
| I have been recently trying to analysis multi-slice
| fMRI data using SPM96, and I was just wondering if my
| procedure has been correct.
|
| I have 5 adjacent slices which I first combine into
| volumes. Then, within SPM96, I realign and smooth the
| data but I don't normalise it (the slices I acquire
| are at an angle and so I cannot give an accurate
| origin for any normalisation). I model the data on a
| simple box-car function with no global normalisation.
| Is this sufficient analysis for such data??
You do not necessarily have to spatially normalize the data.
Note that in SPM99b, that it would be possible to provide better
starting estimates for spatial norm, and also that the realignment
may be slightly more accurate.
| Also, I have been attempting to overlay any activation
| I obtain on the subject's relevant anatomical slices
| (which have also been previously combined into a
| volume) using the 'slices' option. A problem which has
| arose for every dataset is that, when I do this, I
| only appear to have four anatomical slices regardless
| of whether there is activation or not. I'm not sure
| about the fifth slice and I was wondering if someone
| could explain why this has been happening. The 5
| slices I acquire (both the anatomical and the
| functional) have a 0.5mm gap in-between. Is this the
| problem or is something inherent happening when I
| reslice the data to remove the final slice??
Sorry, I can't explain this one.
Regards,
-John
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|