BIDS for magneto- and electro- physiology

[CPernet]

Brain Imaging Data Structure for MEEG

Chair
Cyril Pernet

We wish to have a round table with contributors and users to discuss BIDS for MEEG, possible improvement for the current specification, and discuss/present extensions for derived data

Session
Monday AM 10.30

Agenda

• current specification usage for MEG and EEG any feedback welcome
• MEEG connectivity: should we have a single framework with BEP017
• derivatives, alignement with common derivatives and hot topics

[ChristophePhillips]

Count me in.

[TimVanMourik]

I added this meeting to the agenda for Tuesday 16:00-17:00 (June 11th). Would that work?

[CPernet]

if there are distinct room, could I have that in parallel to demos on Monday? I don't want to miss Thomas keynote on Tuesday

[TimVanMourik]

The morning demos only start on Tuesday so Monday morning is fine! Pick your time.

[CPernet]

thank you - can we do 10.30? is there a place I post a little agenda thing

[TimVanMourik]

I added the meeting to the calendar. In the description I am currently pointing to this issue. Let me know if you'd like me to have it point to an agenda instead.

[CPernet]

great thx

[jasmainak]

Is it just me but I see the meeting on the calendar on Tuesday not Monday

[CPernet]

just you :-)

[sappelhoff]

I see it on Monday as well :-)

[jasmainak]

Hah, okay I figured it out. There are two MEG/EEG events so I was confused between the two

[TimVanMourik]

Ahh, sorry @jasmainak, I was under the impression I moved the meeting, but apparently I copied it. Thanks for spotting the mistake :)

[sappelhoff]

Meeting notes

The meeting happened as planned. Below I summarize some points that were raised. Please add to this list in case I forgot/missed something (@CPernet also took some notes).

Also, please tag other people who attended this meeting and have a GitHub handle (but are not subscribed to this issue yet --> see in the top right of the Github website under "Participants").

@franklin-feingold @wmvanvliet

agenda 1: General Feedback

1. "In our lab we want to format our data according to BIDS for internal use only. It would be convenient for us to use our file formats without converting them to the "accepted" EEG BIDS data formats."
   • Answer: "We cannot extend the list of accepted data formats in BIDS"
   • Follow up: "Could we extend the bids-validator to accept an optional flag in the command line, that non-BIDS-accepted fle formats are validated (not ignored) ... but do not raise an "not-supported-file-format" error?"
   • Answer: "Sounds reasonable, find somebody to make a PR to bids-validator"
2. "Why is the scans.tsvplaced on the session level in the file hierachry as opposed to within each session firectory?"
   • Answer: "Probably historical reasons" (we don't know right now)

agenda 2: Derivatives

1. Should we join BEP017 (connectivity) to also define connectivity derivatives for MEEG?
   • Consensus: Yes ... let's try to specify together ... if it turns out to be impossible, we can split off
2. The derivatives for MEEG (BEP021) have been stagnating a bit because efforts were directed at making the raw specs for eeg/ieeg happen
   • this will change in the coming months
   • we want to make a survey aimed at finding out what most researchers need specified (e.g., annotations, epochs, ICA, ...)
   • based on this, we will adopt a "data-driven 80-20" rule, that is: based on the survey, try to implement in a brief amount of time the bulk of derivatives that most researchers need
   • We will make a call for contributions to BEP021 and distribute it ...
     • on Github
     • on the Google Mailing list
     • on neurostars.org ... and all other channels we can think of
   • this call for contributions will make it clear that BIDS is user-driven and that everyone can chime in and enrich the process of fleshing out the derivatives specification
   • regarding annotations: use JSON sidecards and the HED column in accompanying TSV files extensively ... and then see where it's not sufficient to capture all richness of annotations

---

@TimVanMourik can you please re-open this issue? I think it'd improve the visibility of it.