Update (new version) of MEF3 format (MED).

[MEDFormat]

Hello,

I am Matt Stead, the creator of all the MEF formats (1-3). The new version MEF version (MED) is faster & produces better compression than its predecessors. It continues to be HIPAA & IRB compliant. A fuller description of MED is available on https://MEDFormat.org.

I currently have 4 groups asking me about official BIDS compatibility.

MED data can be read in open source multiple tools (Matlab, FieldTrip, EEGLab, Python , Neo).

These tools are freely available in compiled form for all major platforms (MacOS, Linux, Windows) as well as having source code and specification documents available on https://github.com/MEDFormat.

There are also multiple free data converters (MEF, EDF, DAT, Neuralynx (NCS & NRD), & NATUS) available.

I would be happy to make the pull request for this.

@bids-standard/raw-electrophys-ieeg

[effigies]

FWIW I'm +1 on this, but I don't work with these data.

I believe the places to update in the schema would be:

• https://github.com/bids-standard/bids-specification/blob/master/src/schema/objects/extensions.yaml
• https://github.com/bids-standard/bids-specification/blob/master/src/schema/rules/files/raw/ieeg.yaml

I did open MEDFormat/MEDPython#6, though. The usability of a GPL'd Python library is questionable if users do not want to GPL their own code. I am having trouble digging up references, but I recall seeing some lack of clarity whether Python's import mechanism triggered LGPL or not. IIRC the most similar license that was definitely safe for use in non-GPL Python projects was the Mozilla Public License v2.

---

Edit: Got a reply at MEDFormat/MEDPython#6 (comment). import dhn_med_py is not considered to trigger GPL restrictions by the authors.

[dorahermes]

I am +1 on this as well, as long as we can make sure that the old MEF3 format remains valid.

[cimbi]

Totally agree with this we already started using MED in BIDS.

[EmmanuelBarbeau]

Hi all, we fully support this as we would also like to use MED in BIDS in our Epilepsy Research Group.

[michaelpereira73]

Hello everyone, we are also a few research groups in Grenoble, France to use the MED format and are hoping for it to be included in the BIDS format! Thank you.

[KLehongre]

Hello, I confirm that having MED in BIDs would be really useful for the management and sharing of the data of our database. Thank you!

[tjr1]

We'd be very glad to have both MED and MEF3 in BIDS!

[dorahermes]

I think @effigies described the next steps above. Are there any examples you are interested in adding as well?

[MEDFormat]

Hi Dora, I’ll try to get into this tomorrow. I’ve been busy adding parity capabilities to MED library (optional damage recovery mechanism). think it will be finished today. I think I have the info I need to do this. It looked pretty easy. Cheers, Matt  Matt Stead ***@***.***

…

On Oct 10, 2024, at 11:10 AM, dorahermes ***@***.***> wrote: I think @effigies <https://github.com/effigies> described the next steps above. Are there any examples you are interested in adding as well? — Reply to this email directly, view it on GitHub <#1931 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATNKTHUUODFJMRVQ2JGKIQLZ22YHNAVCNFSM6AAAAABO5W2BXKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMBVGY2DCNZSGM>. You are receiving this because you authored the thread.

[MEDFormat]

Hi Chris, I did the changes & pull requests that you recommended, but on one it failed a check: #1957 Update ieeg to include MED.yaml by MEDFormat · Pull Request #1957 · bids-standard/bids-specification github.com The check error is in "schemacode_ci / Validate schema” Which gives this for the test details: =========================== short test summary info ============================ FAILED tools/schemacode/bidsschematools/data/tests/test_rules.py::test_rule_objects - AssertionError: Undefined objects found in rules: files.deriv.preprocessed_data.ieeg_ieeg_common.extensions[0] == .medd/ files.raw.ieeg.ieeg.extensions[0] == .medd/ ============ 1 failed, 1 passed, 66 deselected, 2 warnings in 7.59s ============ I looked around for that file, but I could not find it. Any chance you can point me to it? Cheers, Matt  Matt Stead ***@***.***

…

On Sep 26, 2024, at 2:05 PM, Chris Markiewicz ***@***.***> wrote: FWIW I'm +1 on this, but I don't work with these data. I believe the places to update in the schema would be: https://github.com/bids-standard/bids-specification/blob/master/src/schema/objects/extensions.yaml https://github.com/bids-standard/bids-specification/blob/master/src/schema/rules/files/raw/ieeg.yaml I did open MEDFormat/MEDPython#6 <MEDFormat/MEDPython#6>, though. The usability of a GPL'd Python library is questionable if users do not want to GPL their own code. I am having trouble digging up references, but I recall seeing some lack of clarity whether Python's import mechanism triggered LGPL or not. IIRC the most similar license that was definitely safe for use in non-GPL Python projects was the Mozilla Public License v2. — Reply to this email directly, view it on GitHub <#1931 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATNKTHVXMNZAGHNNILQ7RI3ZYRSIPAVCNFSM6AAAAABO5W2BXKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNZXHAZTGMRUGM>. You are receiving this because you authored the thread.

[effigies]

I'll merge those into a single PR...

[effigies]

Hi all. If iEEG users (ping @dorahermes @cimbi @EmmanuelBarbeau @michaelpereira73 @KLehongre @tjr1, who have chimed in already) could have a review of https://github.com/bids-standard/bids-examples/pull/469/files to make sure that it looks correct, that would be greatly appreciated.

@MEDFormat provided an emptied data directory, and I updated it along the lines of the already existing ieeg_epilepsy example. Validation is passing on the schema validator using the schema changes proposed in #1956. I have not made a plan to update the legacy validator, though it would not be difficult if someone wants this to be available in the legacy validator (you would just update these lines).

The main thing that needs review is that this dataset would be understandable to iEEG researchers and work with BIDS tools.

[MEDFormat]

Hi Chris, From what I can see it looks like MED is BIDS approved. If I’m wrong about that, please let me know. Cheers, Matt  Matt Stead ***@***.***

…

On Oct 18, 2024, at 7:48 AM, Chris Markiewicz ***@***.***> wrote: Hi all. If iEEG users (ping @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>, who have chimed in already) could have a review of https://github.com/bids-standard/bids-examples/pull/469/files to make sure that it looks correct, that would be greatly appreciated. @MEDFormat <https://github.com/MEDFormat> provided an emptied data directory, and I updated it along the lines of the already existing ieeg_epilepsy example. Validation is passing on the schema validator using the schema changes proposed in #1956 <#1956>. I have not made a plan to update the legacy validator, though it would not be difficult if someone wants this to be available in the legacy validator (you would just update these lines <https://github.com/bids-standard/bids-validator/blob/bec5fea5318953073dc8b97908f11a77ca00845e/bids-validator/bids_validator/rules/file_level_rules.json#L454-L463>). The main thing that needs review is that this dataset would be understandable to iEEG researchers and work with BIDS tools. — Reply to this email directly, view it on GitHub <#1931 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATNKTHSUBRF2HTFB24PABILZ4EGSTAVCNFSM6AAAAABO5W2BXKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMRSGUZDEMBYGE>. You are receiving this because you were mentioned.