On Wed, 9 Feb 2005, Yves Roy wrote:
> First question:
> ---------------
>
> If after having performed a coregister (which creates f*.mat) ) + reslice
> (which apply the f*.mat computed earlier to create the rf*.[img, hdr] to
> correct for the movement, and THEN did a coregister *only* just to access
> the extent of the residual movement (after the correction), i.e. to
> appreciate the quality of the corrected images, THEN THE .mat THAT WERE
> CREATED BY THIS second coregister (only) SHOULD BE REMOVED, otherwise those
> rf*.mat will *corrupt* the subsequent steps of my analysis (which are what
> exactly...) where the .mat will be taken into account.
>
> Am I right? (if they were not removed, then the above error could occur)
You would perhaps better not remove them, but replace the .mat files of
image2...imageN with the one from image1. The reason to do so might be
answer to your 2nd question.
> Second question:
> -----------------
>
> I have converted a functional images from DICOM to ANALYSE using the DICOM
> toolbox in SPM2 and I noticed the presence of .mat files.
>
> My question is:
>
> Should I get rid of the .mat BEFORE starting an analysis (preprocessing +
> model estimation, ...) or can I leave them there?
You may safely leave them in place. These .mat files just tell SPM about
the position of your 3D volume with respect to the MR scanner origin
(which is for SIEMENS exactly the centre of the magnet). If you have
acquired images with different slice orientation (e.g. structural and
functional) during one session, then the .mat file will give you a good
starting estimate for coregistering those images.
Hope this helps,
--
Volkmar Glauche
-
Department of Neurology E-Mail [log in to unmask]
UKE Hamburg Phone 49(0)40-42803-5781
Martinistr. 52 Fax 49(0)40-42803-9955
20246 Hamburg
|