If you use SPM2 to look at SPM5 NIFTI output, the origin will not be
usable due to the switch away from SPM2 Analyze format. This seems
unrelated to WFU_PickAtlas. However, you can look at SPM2 output in
MRIcro ignores the .mat file generated by SPM2. Depending on your
spm.defaults.analyze.flip setting, you may have to manually flip
left/right in MRIcro to match the SPM2 display of the same Analyze file.
This should not be necessary for SPM5 NIFTI output. MRIcron will also
read SPM5 NIFTI format if image data is not stored as double.
Please be sure that you are running the latest version 2.4 of
WFU_PickAtlas. It is available here from the Software Downloads link:
I'll write you a separate e-mail message with details about forwarding
an example to me for testing.
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]]
On Behalf Of zewang
Sent: Monday, October 29, 2007 3:29 PM
To: [log in to unmask]
Subject: [SPM] spm5 +pickatlas orientation problem. totally opposite to
that of spm2
I met the orientation nightmare again in spm5. When I use pickatlas
generate a ROI (left hippocampus) file, I got output images with totally
different orientations. The one from spm2 is correct (left is left using
spm2 spm5 display or checking with mricro), but that from spm5 is
(left is on the right using spm display function or checking with
Why could this happen? Is it a problem of pickatlas?
Another question is that, without touching any settings and changing the
orientation of the data in spm5, what is the right orientation in spm5's
Thank you very much!
Ze Wang, Ph. D
Center for functional neuroimaging,
Dept of Neurology,
University of Pennsylvania
3400 Spruce Stree,
3W Gates building,
Philadelphia, PA 19104