[ENH] Allow MED format for iEEG data (*_ieeg.medd/)

[MEDFormat]

Added MED format

• closes Update (new version) of MEF3 format (MED). #1931

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 86.87%. Comparing base (7ddb7be) to head (4fa72dc).

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1956   +/-   ##
=======================================
  Coverage   86.87%   86.87%           
=======================================
  Files          16       16           
  Lines        1410     1410           
=======================================
  Hits         1225     1225           
  Misses        185      185           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[Remi-Gau]

This should not be considered a blocker but I would REALLY like to see an example dataset with this type of data added to the BIDS examples repository.

[Remi-Gau]

This should not be considered a blocker but I would REALLY like to see an example dataset with this type of data added to the BIDS examples repository.

Will ping all the people who reacted on the associated issue to see if anyone is up for this.

@dorahermes @cimbi @EmmanuelBarbeau @michaelpereira73 @KLehongre @tjr1

[MEDFormat]

That's no problem. I just need permission. Cheers, Matt  Matt Stead ***@***.***

…

On Oct 16, 2024, at 2:03 AM, Remi Gau ***@***.***> wrote: This should not be considered a blocker but I would REALLY like to see an example dataset with this type of data added to the BIDS examples repository <https://github.com/bids-standard/bids-examples>. Will ping all the people who reacted on the associated issue to see if anyone is up for this. @dorahermes <https://github.com/dorahermes> @cimbi <https://github.com/cimbi> @EmmanuelBarbeau <https://github.com/EmmanuelBarbeau> @michaelpereira73 <https://github.com/michaelpereira73> @KLehongre <https://github.com/KLehongre> @tjr1 <https://github.com/tjr1> — Reply to this email directly, view it on GitHub <#1956 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATNKTHTEU327M35HMUZANZ3Z3YMULAVCNFSM6AAAAABP7LB376VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMJWGAZDIMJQGQ>. You are receiving this because you authored the thread.

[Remi-Gau]

I just need permission

Note that the example datasets are 'emptied' of any data: https://github.com/bids-standard/bids-examples/blob/master/CONTRIBUTING.md#why-do-we-only-host-truncated-data-with-0kb-size

This should also make it easier to 'share' them, though we prefer when possible to make sure that the full 'source' dataset we are using is also openly accessible, but this is not a requirement.

[MEDFormat]

It is not possible to have a MED dataset with zero kb size (how is this possible at all in any format?). I can create a dataset with no samples, but it will still have a finite size (headers). Is this what you would like? I have small datasets that contain no protected info, which would be far more useful, if you prefer. Cheers, Matt  Matt Stead ***@***.***

…

On Oct 16, 2024, at 5:55 AM, Remi Gau ***@***.***> wrote: I just need permission Note that the example datasets are 'emptied' of any data: https://github.com/bids-standard/bids-examples/blob/master/CONTRIBUTING.md#why-do-we-only-host-truncated-data-with-0kb-size This should also make it easier to 'share' them, though we prefer when possible to make sure that the full 'source' dataset we are using is also openly accessible, but this is not a requirement. — Reply to this email directly, view it on GitHub <#1956 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATNKTHSELYP3IUIHEF34DYDZ3ZHZ5AVCNFSM6AAAAABP7LB376VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMJWGU4TQMBTGE>. You are receiving this because you authored the thread.

[Remi-Gau]

I meant the 'large' datafiles in the example datasets are truncated:

see here for how we do it.

https://github.com/bids-standard/bids-examples/blob/master/CONTRIBUTING.md#how-to-truncate-data-files-to-0kb

I guess header only files must be pretty light, so I think this could do as well. @effigies what do you think?

[MEDFormat]

I think header only would not be too useful to new users. Small anonymous data set seems like a better choice to me. Cheers, Matt  Matt Stead ***@***.***

…

On Oct 16, 2024, at 8:23 AM, Remi Gau ***@***.***> wrote: I meant the 'large' datafiles in the example datasets are truncated: see here for how we do it. https://github.com/bids-standard/bids-examples/blob/master/CONTRIBUTING.md#how-to-truncate-data-files-to-0kb I guess header only files must be pretty light, so I think this could do as well. @effigies <https://github.com/effigies> what do you think? — Reply to this email directly, view it on GitHub <#1956 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATNKTHVIDVK6M772L3Q4KGTZ3ZZHDAVCNFSM6AAAAABP7LB376VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMJWHE4TOMZUGM>. You are receiving this because you authored the thread.

[effigies]

BIDS examples are strictly about directory structure, file naming and BIDS-specified metadata. This allows us to validate the examples to exercise the rules.

If we intend to duplicate format-internal metadata in BIDS metadata and validate that they are the same, it's worth including enough header information to check that. We have one example with NIfTI files truncated to the header to do this, but otherwise the NIfTI files are truncated to 0 bytes and the header checks skipped.

This is not to say that there shouldn't be a place with a working MED example, just that that's not what bids-examples is for. What's ideal is if there is a canonical example, we can strip it down with a script for inclusion in bids-examples and link back to the original.

[MEDFormat]

OK I’ll make one with no data. Not sure all tools (Python, Matlab, etc.) will work with it - never tested that situation. This will take a little work because I’ll have to write some code to make a session like that. Probably by tomorrow, if that’s OK. Cheers, Matt  Matt Stead ***@***.***

…

On Oct 16, 2024, at 10:48 AM, Chris Markiewicz ***@***.***> wrote: BIDS examples are strictly about directory structure, file naming and BIDS-specified metadata. This allows us to validate the examples to exercise the rules. If we intend to duplicate format-internal metadata in BIDS metadata and validate that they are the same, it's worth including enough header information to check that. We have one example with NIfTI files truncated to the header to do this, but otherwise the NIfTI files are truncated to 0 bytes and the header checks skipped. This is not to say that there shouldn't be a place with a working MED example, just that that's not what bids-examples is for. What's ideal is if there is a canonical example, we can strip it down with a script for inclusion in bids-examples and link back to the original. — Reply to this email directly, view it on GitHub <#1956 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATNKTHS6ZO3M4AKJFOJFDCDZ32KGDAVCNFSM6AAAAABP7LB376VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMJXGM4DGMRQGY>. You are receiving this because you authored the thread.

[MEDFormat]

HI Chris, I made it. Where do I upload it? Cheers, Matt  Matt Stead ***@***.***

…

On Oct 16, 2024, at 11:13 AM, Contact MFO ***@***.***> wrote: OK I’ll make one with no data. Not sure all tools (Python, Matlab, etc.) will work with it - never tested that situation. This will take a little work because I’ll have to write some code to make a session like that. Probably by tomorrow, if that’s OK. Cheers, Matt <MFO-Logo-GIF-small.gif> Matt Stead ***@***.*** > On Oct 16, 2024, at 10:48 AM, Chris Markiewicz ***@***.***> wrote: > > > BIDS examples are strictly about directory structure, file naming and BIDS-specified metadata. This allows us to validate the examples to exercise the rules. > > If we intend to duplicate format-internal metadata in BIDS metadata and validate that they are the same, it's worth including enough header information to check that. We have one example with NIfTI files truncated to the header to do this, but otherwise the NIfTI files are truncated to 0 bytes and the header checks skipped. > > This is not to say that there shouldn't be a place with a working MED example, just that that's not what bids-examples is for. What's ideal is if there is a canonical example, we can strip it down with a script for inclusion in bids-examples and link back to the original. > > — > Reply to this email directly, view it on GitHub <#1956 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATNKTHS6ZO3M4AKJFOJFDCDZ32KGDAVCNFSM6AAAAABP7LB376VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMJXGM4DGMRQGY>. > You are receiving this because you authored the thread. >

[effigies]

Fork https://github.com/bids-standard/bids-examples/ and add it in a pull request.

[MEDFormat]

Hi Chris, I received this (forwarded below) yesterday after uploading the test data set. All I did to makes the other changes was copy the MEF3 entries in the other files & change them to MED versions. This did not include all MED directories & file types, as MEF3 did not. Is that what is causing this? If so, should I add those? Cheers, Matt  Matt Stead ***@***.***

…

Begin forwarded message: From: "MEDFormat" ***@***.***> Subject: [bids-standard/bids-examples] PR run failed: validate_datasets, Attempt #2 - MED example dataset (no data, only headers) (adc75fa) Date: October 17, 2024 at 1:07:10 PM MDT To: "bids-standard/bids-examples" ***@***.***> Cc: "Ci activity" ***@***.***> Reply-To: bids-standard/bids-examples ***@***.***> [bids-standard/bids-examples] validate_datasets workflow run, Attempt #2 validate_datasets, Attempt #2: All jobs have failed View workflow run <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> validate_datasets, Attempt #2 / build (ubuntu-latest, master) Failed in 2 minutes and 15 seconds 1 <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> validate_datasets, Attempt #2 / build (ubuntu-latest, stable) Failed in 1 minute and 34 seconds 1 <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> validate_datasets, Attempt #2 / build (ubuntu-latest, master-deno) Failed in 1 minute and 6 seconds 1 <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> validate_datasets, Attempt #2 / build (macos-latest, master) Failed in 1 minute and 57 seconds 1 <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> validate_datasets, Attempt #2 / build (macos-latest, stable) Failed in 1 minute and 44 seconds 1 <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> validate_datasets, Attempt #2 / build (macos-latest, master-deno) Failed in 58 seconds 1 <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> validate_datasets, Attempt #2 / build (windows-latest, master) Failed in 10 minutes and 9 seconds 1 <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> validate_datasets, Attempt #2 / build (windows-latest, stable) Failed in 9 minutes and 46 seconds 1 <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> validate_datasets, Attempt #2 / build (windows-latest, master-deno) Failed in 1 minute and 58 seconds 1 <https://github.com/bids-standard/bids-examples/actions/runs/11390195544> — You are receiving this because you are subscribed to this thread. Manage your GitHub Actions notifications <https://github.com/settings/notifications> GitHub, Inc. ・88 Colin P Kelly Jr Street ・San Francisco, CA 94107