[Eeglablist] not removing the last ICA artifacts

[Makoto Miyakoshi]

Dear Xi Zh,

There is no hard line of good and bad here, which has been one of the
problems in practical EEG applications.

My suggestion is to use icLabel(), which is a plugin, with which you can
obtain probabilistic label of ICs.
https://sccn.ucsd.edu/wiki/ICLabel
You can set arbitrary threshold, such as 85% or 90%, to reject/include the
EEG/artifact. ICLabel is a published method.

Makoto

On Fri, Jul 26, 2019 at 9:17 AM Xi Zh <timsasah at gmail.com> wrote:

> Hi all,
> I'm using ICA (independent component analysis) with "runica" to remove
> artifacts in my EEG data. I saw the video of Mike X Cohen
> https://www.youtube.com/watch?v=AKCK7DXa0gY. He explained that he prefers
> to remove the first IC components which contain artifacts and that he
> prefers not to remove the other components. My question is If I have 32
> electrodes and I run ICA to give me 32 components. Then I check and find
> that the first 4 components contain artifacts and I remove them. Then i
> find that the 31st and 19th component contain artifact but I don't remove
> them. Is this correct? If it's correct, does anyone have a reference
> (paper) for this.
>
> many thanks for your help in advance
> _______________________________________________
> 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
>


-- 
Makoto Miyakoshi
Assistant Project Scientist, Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego