[Eeglablist] ICA after PREP pipeline incl channel interpolation

[Tarik S Bel-Bahar]

Greetings Alexandra, some responses below, best wishes.

0. What precisely is used as the input data depends on a specific lab's or
tribe's processing preferences :)

1. If you haven't had a chance to yet, and are unfamiliar with ICA and
eeglab, it is very important to go through the whole eeglab tutorial using
the various eeglab tutorial datasets. These will help you make more sense
of everything, including the mechanics of ICA and it's pragmatics in
eeglab.

2. If you haven't had a chance to yet, Google&review Makoto's pipeline for
ICA, as well as google eeglablist + your topic.

3. It is suggested by some experts to  NOT interpolate channels before ICA,
though there are several views about that. My understanding is that
interpolation adds "fake dimensions" back into the data and might confuse
or bother ICA. However, I'm not exactly sure how interpolation might mess
with the rank or ICA. Further,  ICA should be picking up "patches of
activities" that don't belong to single channels, but rather to sources
that show up across groups of channels.

4. If you average reference before ICA, then the recommendation is to drop
one channel before ICA, or, alternatively, run ICA with Channels-Minus-One
as the number of requested ICs (see the pca flag in the runica function).

5. the PREP pipeline is pretty cool! Try to be sure about exactly what a
pipeline you're using is doing, so trust but verify. You may want to turn
off the interpolation before ICA in the PREP settings, and or compare to
alternative method of bad channel detection.

6. Yes eeglab is detecting the low rank via some computation it does (you
could find it you know how to read the code of the functions that are being
run, such as runica).
If it results in 115 channels, then it is doing the reduction
automatically, though you should check and see how many ICs result after
that warning note.

7. Review the output of your ICA for one or two subjects using different
settings. Things should make more sense when ICA is computing with correct
adjustments for














On Tue, Aug 16, 2016 at 12:18 PM, Alexandra Yvonne Vossen <
a.vossen.1 at research.gla.ac.uk> wrote:

>
> Dear all,
>
> I realise this is a popular question but I am really uncertain what
> happens to my data in my preprocessing pipeline. I am quite new to ICA/PCA
> decomposition.
>
> First, the raw data are run through the PREP pipeline (in EEGLAB 13.5.4b),
> including the (spherical) interpolation of noisy channels.
>
> Then after rejection of very noisy epochs I run ICA (for identification of
> artefact components) with EEG = pop_runica(EEG,'extended',1) %this is
> somebody else's code that I am re-using.
> (Minor question: At this point I do not baseline correct each epoch, is
> this recommended?)
>
> I get the info: "Data rank (115) is smaller than the number of channels
> (128).  Input data size [115,525723] = 115 channels, 525723 frames/nAfter
> PCA dimension reduction,
> finding 115 ICA components using extended ICA."
>
> Here is where I am losing track. For this example, there are 116
> non-interpolated channels and the data set was re-referenced to common
> reference before ICA.
> So I guess this might be behind the rank reduction, although I am not sure
> how runica knows about this (where is such information retrieved?)
>
> What precisely is used as the input data now? Any first 115 channels or
> only the non-interpolated channels?
> What happens when the resulting (clean) components are projected back onto
> channel space if some of the channels are not actually included in the data
> set used to calculate the components?
>
> Many thanks in advance,
>
>
> Alexandra Vossen
> PhD student
> School of Psychology
> College of Science & Engineering
> University of Glasgow
> _______________________________________________
> 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
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20160817/406ba40b/attachment.html>