[Mne_analysis] ICA failing to exclude bad channels

[Lyam Bailey]

Dear MNE users,


I'm trying to analyse some EEG data which contains a few very noisy channels (amplitude is often to the order of 1V). This seems to be causing problems with ICA, even after bad channels are excluded


I begin EEG preprocessing by excluding the bad channels, and then re-referencing the data to the average of all remaining channels with:

raw.info['bads'] = ['CP1',etc...]
raw, ref_data = set_eeg_reference(raw, ref_channels=None, copy=False)

After filtering and trial-by-trial artifact rejection, I run ICA with:

ica = mne.preprocessing.ICA(n_components=.99, method='fastica',
                            max_iter=500, random_state=ica_random_state)
picks = mne.pick_types(epochs.info, meg=False,
                       eeg=True, eog=False, stim=False, exclude='bads')
ica.fit(epochs)

This usually outputs a single IC component, and does nothing to address blinks/saccades etc that are clearly present in the raw data. My feeling is that ICA is somehow failing to exclude the bad channels, meaning that (in the presence of much higher variance, introduced by the noisy channels) it is relatively blind to 'normal' artifacts in the EEG.

Does anyone know why this might be happening? Any advice on the problem would be greatly appreciated!

Regards
Lyam


---------------------------------------------------------

Lyam Bailey, BSc.

Graduate Student
Department of Psychology & Neuroscience
Dalhousie University

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20170529/4d2ef46e/attachment.html