Nema data preparation

[KrisThielemans]

don't merge!

Likely this code will go into SIRF

[KrisThielemans]

thanks. this will have to be converted to a function, such that we can run it on other data that we get (i.e. most of it should not be NEMA specific). It would make most sense if there is some diagnostics written to a logfile (although that could be handled by just redirecting output). I think we need output files in 2 directories:

• one with only the final output (prompts, multfactors, additive term)
• another one with other intermediate files (ac/acfs, norm, randoms, scatter, whatever) and log of conversion

we also need a "quick reconstruction OSEM" at the end, with output images save in the "final" dir as Interfile, and 3 pngs (transverse, sagittal, coronal, all 3 middle slices) in the "other" directory for a quick visual check that all went fine.

Unfortunately, OSEM settings will be scanner dependent (also for the scatter by the way), but ok to hardwire 21 subsets, 21 updates, for the mMR for now.

[KrisThielemans]

create separate script download_Siemens_mMR_NEMA_IQ to download/mMR_NEMA_IQ

• download from zenodo
• copy mMR_template_span11.hs from sirf_data_path = os.path.join(examples_data_path('PET'), 'mMR')

this wasn't done yet.