Request to de-identify more fields with highlevel.anonymize_edf()

[catvasily]

Would it be possible to have more fields, which can be de-identified.
Now it is Subject name and BirthDate.
Could you please add Subject code, Admin code, Additional Info, and Technician to the list?
You never know what to expect - EEG technicians might use different fields occasionally - there are no strict guidelines.
At least, Subject Code, please.
Thank you.

[skjerns]

you can de-identify any field you like by just passing it to the function

to_remove = ['patientname', 'birthdate', 'admin_code']
new_values = ["xxx", 0, 'test']
anonymize_edf(edf_file, to_remove=to_remove, new_values=new_values, verify=False, verbose=False)

[catvasily]

Thank you for letting me know.
I was confused with the help for this function - it specifies two fields.
Tested it for
to_remove=['patientname', 'patientcode', 'patient_additional', 'admincode', 'technician', 'recording_additional']
and it works fine.
Thanks!

[catvasily]

Also, I would like to report one peculiarity. Not an issue, strictly speaking.
When I try to de-identify the field 'admincode', say, with a sting like '4761fdc1-3219-4e68-b5a1-00d93159ad39', highlevel.anonymize_edf() generates a new EDF without any error or warning, and returns True.
At the same time, EDFbrower cannot open this file.
Just wanted to let you know.

[skjerns]

What error message is EDFBrowser giving? Can you send a sample file?

The local patient information field (which combines a couple of fields like admincode, name, etc) can only be 80 characters in total (see EDF+ specs). It might be that you exceed this limit. I'm not sure if pyedflib gives a warning in this case, I need to check.