Pet template file names issue in bep009 1.6.0-dev rendered draft

bids-standard / bids-specification

Brain Imaging Data Structure (BIDS) Specification

https://bids-specification.readthedocs.io/

Creative Commons Attribution 4.0 International

278 stars 162 forks source link

Pet template file names issue in bep009 1.6.0-dev rendered draft #768

Closed chrisrj closed 3 years ago

chrisrj commented 3 years ago

Hello, If I am not mistaken, in early 2021, the PET template file name changed from sub-[_ses-][_task-][_acq-][_rec-][_run-]_pet.json to sub-[_ses-][_task-][_trc-][_rec-][_run-]_pet.json that is the '_acq' key label was renamed '_trc'. Currently, the PET template file name in 09-positron-emission-tomography.md is generated by a macro {{ MACROS___make_filename_template(datatypes=["pet"], suffixes=["pet"]) On the draft https://bids-specification.readthedocs.io/en/bep-009/04-modality-specific-files/09-positron-emission-tomography.html, this is rendered as sub-[_ses-][_task-][_acq-][_trc-][_rec-][_run-]pet.json with both a'acq' and a '_trc' key label. Best regards, Christian

tsalo commented 3 years ago

Thanks for reporting this. Two questions:

Should this extend to blood.tsv files as well? Currently both acq and trc are allowed.
Can you think of any cases where a generic label (i.e., acq) could be useful for either pet.nii.gz or blood.tsv files? We allow acq for most files in BIDS because it serves as a freeform label, but if trc and other entities cover all of reasonable cases for PET I can remove it.

effigies commented 3 years ago

I would vote to drop acq unless someone can identify an acquisition parameter variation that isn't already covered by trc- and rec- for PET or recording- for blood. We can re-add entities in later releases if the example comes later.

Come to think of it, for passive recording techniques, including MEG and iEEG, there's limited scope for varying acquisition parameters. Would need to be BIDS 2, but we might want to consider removing that. EEG might have more need because of the variability of electrode placement.

mnoergaard commented 3 years ago

Thanks for reporting this @chrisrj. Based on previous discussions regarding this the trc was meant to replace the acq. This is also how the validator has been made. This applies both to pet files and blood files.

tsalo commented 3 years ago

I can open a PR to drop acq from both pet.nii.gz and blood.tsv files. We can also double-check the other entities in use in the pull request review.

tsalo commented 3 years ago

Not sure why #769 didn't close this, but I'm going to close now that the PR has been merged. @chrisrj please reopen if the problem wasn't resolve, or open a new issue if you identify any other issues with the schema/rendering in BEP-009.