Dear colleagues,
I'm wondering about the wrap option in spm_realign and spm_reslice: In
message no. 018099 John Ashburner explains that this option is meant to
correct the common wrap-around artefacts in y-direction by setting it to [0
1 0].
We made the observation that with setting wrap to [1 1 1] in the GUI (in a
dataset without any visible wrap-around effect), the activation in the
resulting glass-brain images looks more focused and three more significant
voxels are reported.
1. I would be grateful if someone could point what calculations are
performed when a wrap value is set to 1.
2. When working with batches, is it necessary to pass the same wrap
parameters to spm_realign AND to spm_reslice?
3. Does it make any sense to use this option at all when there's no visible
wrap-around artefact? What potential harm can it do?
Thanks!
Joerg
|