I've heard from a number of researchers that this is an issue,
particularly for SPECT and PET. MR doesnt seem to be as much of a
problem. There is some catch-up that is necessary in the the nuclear
medicine industry to allow for seamless multi-modality coregistration.
I've noticed some discontinuity in how the nuclear image is acquired
on the actual machine that is backwards for some of the older
machines. The CT for attenuation correction may be oriented "head to
foot", but the nuclear images are acquired "foot to head". I'm told
that any attempt to change this occurrence seemed to lead to more
problems than fixing them.
If anyone can help with that, I'm all ears! :)