BIDS manuscript

[pold...@gmail.com]

Dear BIDS Community members,

I am writing to invite you all to join us in coauthoring a manuscript titled "The Past, Present, and Future of the Brain Imaging Data Structure (BIDS)". I am attaching the draft manuscript here.  If you are interested in joining as a coauthor, please complete the following form, which includes an opportunity to provide comments on the manuscript:

https://forms.gle/DtG89DR9eqL1aRUx9

The authorship policy is as follows:

• Authorship is open to anyone who has made demonstrable contributions to the BIDS ecosystem, including contributions via Github or Google Docs to the main specification or any BIDS Extension Proposals (merged or proposed). 

• Each author will denote their contributions according to the CRediT Taxonomy.  All authors are expected to at least contribute by reviewing and editing the manuscript, as well as at least one additional contribution category reflecting their role(s) in the BIDS community.

• Authorship order beyond the primary authors (RAP, CJM, KJG) will be determined alphabetically.

If you are interested in joining, please complete the form by August 1, 2023.

Looking forward to seeing many of you at OHBM!

Cheers,

Russ

[Yaroslav Halchenko]

Thank you Russ for leading this effort!

Don't you think it would be worthwhile already to push it out to some
preprint archive, e.g. bioarxiv?

Cheers,

On Wed, 05 Jul 2023, pold...@gmail.com wrote:

> Dear BIDS Community members,
> I am writing to invite you all to join us in coauthoring a manuscript
> titled "The Past, Present, and Future of the Brain Imaging Data Structure
> (BIDS)". I am attaching the draft manuscript here.  If you are interested
> in joining as a coauthor, please complete the following form, which
> includes an opportunity to provide comments on the manuscript:
> https://forms.gle/DtG89DR9eqL1aRUx9
> The authorship policy is as follows:

> * Authorship is open to anyone who has made demonstrable contributions

> to the BIDS ecosystem, including contributions via Github or Google
> Docs to the main specification or any BIDS Extension Proposals (merged
> or proposed). 

> * Each author will denote their contributions according to the CRediT

> Taxonomy.  All authors are expected to at least contribute by
> reviewing and editing the manuscript, as well as at least one
> additional contribution category reflecting their role(s) in the BIDS
> community.

> * Authorship order beyond the primary authors (RAP, CJM, KJG) will be

> determined alphabetically.

> If you are interested in joining, please complete the form by August 1,
> 2023.
> Looking forward to seeing many of you at OHBM!
> Cheers,
> Russ

--
Yaroslav O. Halchenko
Center for Open Neuroscience http://centerforopenneuroscience.org
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
WWW: http://www.linkedin.com/in/yarik

[Pradeep Reddy Raamana]

I enjoyed reading this paper, which would help educate folks who haven't been as involved over the years. The para on better coordination with DICOM is interesting: "As a result, many standardization efforts have been duplicated. As DICOM is the industry standard and more data will be arriving in DICOMs, coordination with developments in DICOM could help to ensure more rapid adoption of new imaging sequences".

it would help to provide an example of such duplication, and how coordination can be improved. It was a lot of pain for us (while developing mrQA) to understand DICOM and docs on BIDS website helped in some cases, although some confusion still remains in how to read some parts of the DICOM header.

[Yaroslav Halchenko]

Well -- consider

- the list of working groups present in DICOM:
https://www.dicomstandard.org/activity/wgs and you find many efforts
overlapping with domains/modalities for which there are BEPs, e.g.
"WG-32 Neurophysiology Data", "WG-30 Small Animal Imaging",
and some standards (e.g. NWB) BEPs we have.

So overall -- both standards work in the same domains.

- we use over 70 DICOM tags, see
https://github.com/bids-standard/bids-specification/pull/1450
so we are at large duplicating there

- DICOM started to use ontologies and standard dictionaries, e.g. here
are some pointers from David Clunie of DICOM (private correspondence):

> On the subject of Anatomy aka. Body Part, the encoding of this information (if
> present at all) in DICOM has evolved over time from being defined text strings
> to using standard ontologies (like SNOMED CT), e.g., see:
>
> https://dicom.nema.org/medical/dicom/current/output/chtml/part16/sect_CID_8134.html#table_CID_8134
> https://dicom.nema.org/medical/dicom/current/output/chtml/part16/chapter_L.html
>

so here we are slowly arriving to common ideas of reusing standard ontologies
etc, similarly to what NIDM efforts are trying to achieve.