To do list

[mszinte]

BEP-020 reference file

This is a list of things we identified with @Remi-Gau for the BEP-020 of BIDS.
Previous meeting planned Tuesday December 13th
Next meeting planned Tuesday January 10th

• Pull all the changes locally
• Add change that were not push in my repo
• Modify the template to get links as here => not possible as not MACROS #2
• Modify the template to refer to matching data structure as here
• Describe the table for TSV of the future converted eyetracking "real" eyetracking bids data, see tabular files here.
• Verify that "The OPTIONAL... " text is gone (@Remi-Gau has supposedly delete it already).
• Case of "SamplingFrequency_eyetrack": check that "Sampling_frequency" information cannot take an addition sufficient for eye tracking, if not creating it will be fine.
• Mention in the text that for valid eye tracking data Screen settings are mandatory.
• Take out key name that thar are not necessary as _eyetrack is an addition recoding method (InstitutionName, Institution Address, TaskDescription, Instructions, CogAtlasID, CogPOID)
• Change AOI definition to SAOI
• Check that all remarks from the pull request are taken care of with @Remi-Gau (some are for the maintainers).
• Find a pupil dataset
• Take into account comments of @yarikoptic
• Write comment about branch: /(?!^master$)/ in response to @yarikoptic
• Deal with comment on ScreenAOI of @yarikoptic
• discuss with @Remi-Gau to fix error of mkdocs server
• figure out why local mkserver not reloading after changes
• take out element of .asc file from metadata
• review all documentation
• fix tabular input to keep alphabetical order
• deal with review of tsalo
• test without modality in each file
• Discuss with BIDS Steering group whether I should have a converter ready before submitting the proposal.
• last version with tabular data
• Make a pull request of the Task events description to add as RECOMMENDED the Screen settings (size, resolution, distance, refresh rate)
• read discussion about syncing and [fieldtrip] (https://www.fieldtriptoolbox.org/example/) and fieldtrip-meg
• See back my notes on this paper to see if we want to change anything
• @Remi-Gau Take out these parameters when PR about screen accepted, to avoid conflict, merge latest version to my branch.
• @mszinte end conversation about data format in github
• @Remi-Gau see red crosses on continuous integration
• @Remi-Gau @mszinte figure out how to have two format for data .tsv and .tsv.gz => solution put brackets
• @Remi-Gau make PR about HeadStabilization and VisualAids (waiting feedback)
• @mszinte Modify the resting state dataset
• @mszinte write converter from .edf to .tsv.gz
• @mszinte change the resting state dataset
• @mszinte update the text to point towards the version 2 of the dataset
• Add data to (bids-example)[https://github.com/bids-standard/bids-examples]
• @Remi-Gau include required screen settings in eyetrack
• @julia-pfarr Work a early version of the converter
• Modify Jan Brascamp dataset (pupil tracking)
• Modify with @Remi-Gau an openneuro dataset

[Remi-Gau]

Other points:

• support for EDF format
• handle compatibility with EEG / MEG / iEEG files that do contain eyetracking data (channels.tsv list PUPIL and EYEGAZE as possible channel type)

[Remi-Gau]

So we do not forget.

Are still needed to finalize the BEP:

1. make sure everything is fine on the PR for BEP (mostly fine)
2. add example datasets to the bids example repo

• see listing of examples: add them there if needed
• see for example this PR in the bids example repo

3. update validator:

• Ross started on it in an old PR based on the google doc: [WIP] BEP020 Eyetracking bids-standard/bids-validator#1423

[oesteban]

I've had a look into the openneuro dataset proposed above, and it seems to have been acquired with an Arrington Research Viewpoint Eyetracker. Documentation - https://www.biopac.com/wp-content/uploads/viewpoint_eyetracker_manual.pdf

It writes out text files, so relatively easy to parse. I may take a stab at this next week and we could present it as an example.

Once the BEP is accepted, I'll ask Gustav to update the openneuro dataset (or give me/us permission to move the eye tracking data into raw).

[oesteban]

I can also create a bids-example from my data (the repo I shared privately with you). Since I will not publish any data (i.e., the tsv.gz files will be empty files, as other examples), I can share the metadata without issues.

[Remi-Gau]

Pretty sure I already took care of it. Will. Just need to update the code to the latest change of the BEP.

[oesteban]

I have updated the HackMD to remind myself of adding columns/metadata recommendations for this eye tracker (once we pass the page of new vs old specs).

[Remi-Gau]

In case someone else asks for it. Like future me for example.

https://github.com/Remi-Gau/ds000201-eyetrack-conversion

[oesteban]

Pretty sure I already took care of it. Will. Just need to update the code to the latest change of the BEP.

BTW, how did you get around the fact that this eyetracker does not give you monotonic sampling (i.e., there is a sampling frequency but it is not "normalized")?

[Remi-Gau]

Don't remember. Not sure that I actually checked for that. Feel free to open an issue on that repo so we don't forget.

[julia-pfarr]

I also prepared some example datasets, just didn’t share them yet. Will do next week! Am 01.03.2024 um 16:03 schrieb Remi Gau ***@***.***>: Don't remember. Not sure that I actually checked for that. Feel free to open an issue on that repo so we don't forget. —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>