Specify REF Electrode for EEG/iEEG for HM computation

[tmedani]

Hi team,

While in most cases the ref electrode is converted into an average during the inverse processing,
I believe it is important to specify the REF electrode for the HM computation, perform the computation, re-reference the final Gain to the REF electrode, and then do the inverse as is.

We can discuss further which option is best and review how this is handled across the different methods.

A+

[Edouard2laire]

hi.

(Naive question) How does the reference electrode change the computation of the forward model ?

[tmedani]

Hi @Edouard2laire
This is an important question.
It does not change the computation, but if not specified, it may use of the available electrode as the reference for the computation, which is the case for the FEM.
For OpenMEEG, we still need to check how they handle this; the final output is re-referenced to the average ref.
For the analytical solution, I believe the infinite ref is used in the formulation.
We will update some sections and make this clearer for the users and us.

[Edouard2laire]

Thx. How can one import the location of the ref electrode in Brainstorm? In our case, we also digitalize where we put the ref (FCz, often), but since it's not a channel in Brainstorm, the position is not imported.

[rcassani]

This question has been answered in the forum:
https://neuroimage.usc.edu/forums/t/retrieval-of-reference-channel/38501/10

The idea is apply a add-back-ref-channel montage to your data to add a new channel, the reference.
Then you can add its location as with any other channel.