[Eeglablist] ICA and dipfit questions

[Makoto Miyakoshi]

Dear Paul,

Thank you for your encouraging words.

> Per Makoto’s (very very useful) pipeline page it states to do so “as long
as they share the initial reference electrode with scalp electrodes.” I’m
not clear if this is with respect to how the raw data are collected or
whether you apply a common reference upon importing into EEGLAB (such as I
have to do since I use a Biosemi system).

If one electrode serves a reference and the rest of others are all
referenced to it, such referencing style is called monopolar referencing.
Your recording montage could be a combination of monopolar-reference scalp
electrodes plus bipolar-referenced EOG electrodes etc. If the latter is the
case, you may want to remove EOG channels before ICA (Is it true though?
Now I'm not that confident... I feel that including bipolar referenced
recordings are still useful for ICA.) That's what I mean.

Biosemi data needs to be referenced by users. At this point you will
loose data rank by one.

> I ask because I utilize ICA to remove, among other things, cardiac field
artifact (CFA) and pulse artifacts in order to get a look at heartbeat
evoked potentials. ICA obviously models the EKG traces very well as
individual components, but I am not confident that removing these helps
remove confounds like the CFA.

What you do sounds fine to me.

> Makoto’s pipeline recommends after ICA running DIPFIT and then fitting
bilateral dipoles with their own respective plugins. However, I see that
the dipfit GUI has a selection box for fitting bilateral dipoles. Is this
using the “fit bilateral dipoles” plugin, but within the context of DIPFIT?

If you turn on that option, ALL the dipoles fitted will be symmetrical
bilateral dipoles.

> This would be really helpful because, when one runs the “fit bilateral
dipoles” plugin EEGLAB closes out and does not reopen all datasets that
were held in memory (pretty annoying).

What? I think that's a bug, sorry. I can fix it if you can tell me how to
reproduce the problem.
Meanwhile, if you type 'eeglab redraw' in the command line, you should be
able to recover GUI you have been working on.

Makoto



On Tue, May 5, 2020 at 12:17 PM Beach, Paul Anthony <
paul.anthony.beach at emory.edu> wrote:

> EEGLAB gurus,
>
> I have two general questions that have been nagging at me for a while.
>
> **First
> When it comes to setting up ICA, I have seen many recommend including all
> channels (including auxiliary, nonEEG channels like EKG) in the prompt. Per
> Makoto’s (very very useful) pipeline page it states to do so “as long as
> they share the initial reference electrode with scalp electrodes.” I’m not
> clear if this is with respect to how the raw data are collected or whether
> you apply a common reference upon importing into EEGLAB (such as I have to
> do since I use a Biosemi system).
>
> I ask because I utilize ICA to remove, among other things, cardiac field
> artifact (CFA) and pulse artifacts in order to get a look at heartbeat
> evoked potentials. ICA obviously models the EKG traces very well as
> individual components, but I am not confident that removing these helps
> remove confounds like the CFA.
>
> **Second
> Makoto’s pipeline recommends after ICA running DIPFIT and then fitting
> bilateral dipoles with their own respective plugins. However, I see that
> the dipfit GUI has a selection box for fitting bilateral dipoles. Is this
> using the “fit bilateral dipoles” plugin, but within the context of DIPFIT?
> This would be really helpful because, when one runs the “fit bilateral
> dipoles” plugin EEGLAB closes out and does not reopen all datasets that
> were held in memory (pretty annoying).
>
>
> Thanks!
> Paul
> --
> Paul Beach DO, PhD
> PGY4, Chief Resident Physician
> Department of Neurology
> Emory University School of Medicine
>
>
>
> ________________________________
>
> This e-mail message (including any attachments) is for the sole use of
> the intended recipient(s) and may contain confidential and privileged
> information. If the reader of this message is not the intended
> recipient, you are hereby notified that any dissemination, distribution
> or copying of this message (including any attachments) is strictly
> prohibited.
>
> If you have received this message in error, please contact
> the sender by reply e-mail message and destroy all copies of the
> original message (including attachments).
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu