[Eeglablist] What to do with EOG channels when average re-referencing and ICA decomp.

[Cedric Cannard]

Hi,

ICA should be able to identify your eye artifacts without using your EOG channels in theory. Have you tried using ICLabel?

After preprocessings and importing your channel loations, you coult try this (you need to have the ICLabel plugin installed):

%Run ICA
EEG = pop_reref(EEG, [],'exclude', your_EOG_channels); %change input here with your EOG channel numbers
EEG = pop_runica(EEG, 'icatype', 'picard', 'pca', -1);
% EEG = pop_runica(EEG, 'icatype', 'runica', 'extended',1,'interrupt','on');
EEG = iclabel(EEG);
[~, ind] = max(EEG.etc.ic_classification.ICLabel.classifications, [], 2);
badComps = [find(ind == 2);find(ind == 3)];
EEG.reject.gcompreject(badComps) = 1;
EEG = pop_subcomp(EEG, badComps);
pop_eegplot(EEG, 1, 1, 1);

Hope this works,
Cedric


--
Cédric Cannard, MSc, PhD candidate in Neuroscience
Université Toulouse III, CerCo Lab, CNRS, France
ccannard at protonmail.com

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Monday, May 4, 2020 10:12 AM, Maitane Barrenetxea Carrasco <mbarrenetxea at mondragon.edu> wrote:

> Very dear EEGLab-list,
>
> I am currently working on a dataset with 128 EEG channels + 4 EOG channels
> (2 vertical + 2 horizontal) and a nose-tip reference for all channels.
> I have doubts regarding the EOG channels when re-referencing and
> decomposing the data with ICA.
>
> When re-referencing the data to average, I have originally excluded the EOG
> channels, as this would propagate the EOG artifact through the rest of the
> channels. Hence, the 128 EEG electrodes are average referenced, and the EOG
> channels are referenced to the original reference (nose-tip). When
> subsequently decomposing the data using ICA for artifact removal, I am
> unsure as to whether I should include the EOG channels in the decomposition
> as they don't share the same reference as the rest of the data. If I
> exclude them, will ICA still be able to isolate the EOG components?
>
> If I want to include them in the ICA decomposition, should I also include
> them in the average re-reference (even if this will propagate the EOG
> artifact through the whole dataset)?
>
> Thank you very much in advance,
>
> Maitane
>
>
> ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> [image: logo_berria] Maitane Barrenechea Carrasco (PhD)
> Biomedikoa - BIO
> Mondragon Unibertsitateko Goi Eskola Politeknikoa
> Loramendi, 4; 20500 Arrasate - Mondragón (Gipuzkoa)
> Tel. : +(34) 647504294 / +(34) 943794700 + Ext. 8162
> https://urldefense.com/v3/http://www.mondragon.edu;!!Mih3wA!R3YyH8oUN_McckyKcs2aVxBV5PnWKusJfcpxbAgsGVYH97m9KGzjBmwkqtogp4ca-W4-og$
>
> 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