On Tue, 12 Jul 2005 15:14:42 -0400, Sunil Patel <[log in to unmask]> wrote:
>Hello all,
>
>We had a question regarding the correct specification of parameters for
>slice timing correction. Looking at the some of the posted messages
>and the comments in the code, it seems that the convention for
>numbering slices is "1=bottom". Does "bottom" refer to the bottom of
>the brain or the bottom of the FOV (the bottom of the SPM
>window/bounding box).
It refers to what SPM _thinks_ is the bottom of the brain. In your words,
this is the "bottom of the FOV."
> For example, we have 24 axial slices acquired in
>interleaved order from the bottom of the brain to the top on the
>scanner. After I convert the GE DICOM files to Analyze format using
>MRIcro (no reorientation done at this step) and then view them in SPM2
>(using the Display button), I see an upside down brain when looking at
>the coronal and saggital views. So, according to the SPM convention
>does slice number 1 in our case refer to the slice near the top of the
>brain?
Yes.
>Also, is the convention the same for SPM99 and SPM2?
Yes. It's not a convention for the slice timing module _per se_,
however. Rather, it stems from the Analyze format (as interpreted by SPM)
and the fact that SPM is a computer program and as such has limitations.
Meaning, when you hand SPM an "upside-down brain," SPM will (reasonably)
interpret the bottom of the FOV to be the bottom of the brain. While it
might be obvious to us humans that this isn't really right, it's not
obvious to a computer program.
>Sorry if this is a stupid question and thanks in advance for any advice.
It's actually a very good question.
Caveat: reorienting images using the "Reorient" button (after using
the "Display" button) will _not_ affect how SPM interprets these things
for the purpose of the slice timing correction module. "Reorient" doesn't
alter the *.hdr and *.img files; rather, it creates or modifies an
accompanying *.mat file. From examples, it appears that this kind of
reorientation is "ignored" by the slice-timing module.
>=========================================================================
|