[Eeglablist] Does EEG.icaact represent components in channel reference?

[Martin Simoneau]

Dear Makoto,

Thanks for your answer.

After I run ICA, does the structure EEG.data contains the components transformed in channel-by-times?

If not, how can I transform back the components in channel-by-time series?

Regards,

Martin


De : Makoto Miyakoshi [mailto:mmiyakoshi at ucsd.edu]
Envoyé : jeudi 9 février 2017 22:42
À : Martin Simoneau <Martin.Simoneau at kin.ulaval.ca>
Cc : eeglablist at sccn.ucsd.edu
Objet : Re: [Eeglablist] Does EEG.icaact represent components in channel reference?

Dear Martin,

> As a result, if row 20 in EEG.data represents C3, analysing row 20 in EEG.icaact is analysing C3, am I right?

No. Channels and ICs are related via weight matrix. It's not one to one!

> In other words, if I remove components, it is possible that C3 is no longer present.

No. When you remove IC20, then the backprojection of the remaining ICs will miss contribution by IC20.

Makoto


On Tue, Feb 7, 2017 at 1:08 PM, Martin Simoneau <Martin.Simoneau at kin.ulaval.ca<mailto:Martin.Simoneau at kin.ulaval.ca>> wrote:
Hi everyone,

I have questions regarding this sentence: “the components were transformed back to the channel-by-time series.” From my understanding, the authors ran ICA on their EEG channels. Thereafter, transformed back to the channel time-series meant that they used EEG.icaact to analyse their time-series in EEG channel reference. Therefore, the time-series in EEG.icaact are the components in the channel reference, am I right?

As a result, if row 20 in EEG.data represents C3, analysing row 20 in EEG.icaact is analysing C3, am I right?

However, if I remove components because some showed eye blink or other artefact, now EEG.icaact has less row than EEG.data. How can I find C3 after removing components?

In other words, if I remove components, it is possible that C3 is no longer present. Can I transform components back in channel-by-time series without losing some channels after removing components?

Thanks for your help,

Martin


_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu<mailto: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<mailto:eeglablist-request at sccn.ucsd.edu>



--
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20170210/3c05e34b/attachment.html>