I would like to know the exact difference between combination methods OC and tedana_OC when specified via the -combine_method tag in afni_proc.py. From what I understand, they both seem to be doing the same 'optimal combination' based on T2* weighting as outlined in Posse et al. (1999). Is the only difference then that OC is done using the AFNI command @compute_OC_weights whereas tedana_OC is done using tedana.py (as part of Prantik's meica v2.5)?
Yes, they are just different implementations of the same thing, and so both called OC.
Potential bias aside, I prefer AFNI's OC. While the results are very similar, the tedana.py version leaves some voxels out of the results where their OC computed weighting seemed out of an expected range. AFNI's OC just limits the weighting and does not remove any results.
I don't recall whether tedana's OC ends up masking the data (as the tedana projection step does). If so that is another aspect that seems better to avoid. But I am not recalling with certainty. Of course, it's simple enough to check.
rick
The
National Institute of Mental Health (NIMH) is part of the National Institutes of
Health (NIH), a component of the U.S. Department of Health and Human
Services.