[Eeglablist] Changing the order of the electrodes

Makoto Miyakoshi mmiyakoshi at ucsd.edu
Wed Mar 31 15:29:10 PDT 2021


Dear Ivano,

> since it's not correct to use the ICA with interpolated channels. Am I
wrong?

Wrong. As long as you can specify the effective data rank correctly, it is
not a problem.
See the following two cases

1. If the electrode interpolation is linear -> Your scalp EEG data is now
exactly rank deficient.
2. If the electrode interpolation is 'spline' (EEGLAB default) -> Your
scalp EEG data is effectively rank deficient.

The case 1 is easy because rank(EEG.data') shows the exact data rank which
can be used for 'pca' option for ICA.
The case 2 is a bit complicated: see my Wiki page
https://sccn.ucsd.edu/wiki/Makoto%27s_useful_EEGLAB_code#How_to_turn_on_the_automatic_protection_against_rank-deficient_ICA_in_pop_runica_.283.2F31.2F2021_added.29

In any case, it is recommended that you check the effective data rank
following my recommendation above and use it for 'pca' option when calling
pop_runica (or amica)

pop_runica() has a safety against this rank issue, but the implemented
safety has been disabled.

Makoto


On Wed, Mar 31, 2021 at 12:07 PM ivano triggiani via eeglablist <
eeglablist at sccn.ucsd.edu> wrote:

> Hi Clement,
>
> thank you so much for the answer. Actually, I am not sure I got the point
> in Makoto's answer to that message. I sometimes have some bad channels, I
> remove them, I go on with the pre-proc (following basically the steps
> Makoto suggests), and I perform the ICA for the artifacts. Only later, I
> interpolate the bad channels since it's not correct to use the ICA with
> interpolated channels. Am I wrong?
>
> The function reorder is only for 19 channels, but, of course, I can modify
> it for my entire set, thank you (and Makoto) very much. It is very useful.
>
> Ivano
> ----
>
>
>
> On Wed, Mar 31, 2021 at 2:31 PM Clement Lee <cll008 at eng.ucsd.edu> wrote:
>
> > Hi Ivano,
> >
> > Hopefully you have not run ICA yet since there are some complications
> (for
> > details: https://sccn.ucsd.edu/pipermail/eeglablist/2015/010610.html).
> > I don't think there is a built in function for that in EEGLAB, but Makoto
> > has written a `reorder19Channels` plugin (downloadable through the plugin
> > manager) if you want a reference script.
> >
> > Best,
> > Clement Lee
> > Applications Programmer
> > Swartz Center for Computational Neuroscience
> > Institute for Neural Computation, UC San Diego
> > 858-822-7535
> >
> >
> > On Sun, Mar 28, 2021 at 9:42 AM ivano triggiani via eeglablist <
> > eeglablist at sccn.ucsd.edu> wrote:
> >
> >> Hello,
> >>
> >> After the pre-processing and the interpolation of the bad channels, I
> have
> >> for every subject a different order for the electrodes. This appears not
> >> to
> >> be an issue for the Study functions (despite a warning appearing, does
> it
> >> mean something?).
> >> But if I want to use some external function or just import the data
> >> elsewhere, this can be a big deal. Do I have to write a script to change
> >> the order of the channels, or is it possible to do it in EEGlab in a
> >> simpler way?
> >>
> >> Thanks in advance,
> >>
> >> Ivano
> >>
> >> ________________________________________________
> >> _______________________________________________
> >> 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
> >>
> >
> _______________________________________________
> 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
>


More information about the eeglablist mailing list