[Eeglablist] Pseudonymization of data (for protection of participant identity)

[Kasper Eskelund]

Dear EEGLABers,

My group has recently finished the collection of a large dataset on various
reward tasks. We are now ready to proceed to preprocessing, analyses and
further steps. Some of this will be done in collaboration with researchers
external to our institution.

The data are collected in a clinical setting and identified by the patient
ID of our clinic (in filename and setname), so that we internally can
combine EEG data with many other modalities of data (mostly symptom scales
and history).

However, due to data protection regulations, we cannot share our data with
external collaborators as long as it is marked in a way, where individual
patients can be traced and identified in other data (given access to
databases in our clinic). Thus we need to pseudonymize the EEG data and
delete all instances of patient ID in the EEG data.

Looking into the EEG structure, I have found three places where the
original ID is present: in EEG.setname, EEG.filename and EEG.history. These
are easily altered.

I would like to ask, if there are further, known ways that the filename and
setname are stored in the .set and .fdt files, possibly making patients
identifiable.

Also, if anyone has experience with pseudonymization of EEGLAB files, I
would be happy to learn from it.

Best,
Kasper