[Eeglablist] How to use ICA

Stephen Politzer-Ahles spa268 at nyu.edu
Thu Mar 12 10:59:54 PDT 2015

Hello Aditya,

See http://sccn.ucsd.edu/wiki/Chapter_09:_Decomposing_Data_Using_ICA .
Basically you just need to run an ICA decomposition, use some method to
decide which ICs represent noise that you would like to remove, and then
remove those ICs from the data. If you search through the literature and
the EEGLAB mailing list archives, you will find lots of discussion of more
details about how to make ICA work better (regarding topics such as what
reference to use, what baseline to use, whether or not to reject certain
other artifacts before ICA, whether to run ICA on continuous or epoched
data, etc.).

Good luck,

Stephen Politzer-Ahles
New York University, Abu Dhabi
Neuroscience of Language Lab

On Thu, Mar 12, 2015 at 5:29 PM, Aditya Bagusmulya <
aditya.bagusmulya at gmail.com> wrote:

> Hello,
> In my thesis i need to use ICA for separating between real data and noise.
> My data is EEG signal and the case is epilepsy. So I use EEGLAB for do it.
> But, I still cant understand how to use ICA. I'd read the step and tried to
> do but i didn't know, whether it was true steps or not.
> This is the tutorial that i read http://sccn.ucsd.edu/eeglab/quickrej.html
> Do I need to reject the artifact first? Can I go to run ICA step without
> reject artifact?
> Please help me.
> I picked the data from Bonn University.
> http://epileptologie-bonn.de/cms/front_content.php?idcat=193&lang=3
> Hopefully someone can help me.
> Thank you.
> _______________________________________________
> 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/20150312/145b378b/attachment.html>

More information about the eeglablist mailing list