[FieldTrip] ICA artifact correction for continuous multimodal MEG and EEG data

[Schoffelen, J.M. (Jan Mathijs)]

Hi Erfan,

Welcome to FieldTrip! Yes, it is possible to use an ICA procedure to clean your continuous data. In theory, it is possible to unmix (with ICA) the EEG and MEG data in a single call to ft_componentanalysis. Yet, I would be reluctant to do this. Although not inhibited by practical insights, I think that such an approach could be suboptimal. The reason for this is that, when the signals are backprojected from components to channels, all backprojected channels will be mixtures of EEG and MEG signals. I’d rather run the cleaning procedure twice, i.e. unmix and backproject the EEG and MEG channels separately. 

Best wishes,
Jan-Mathijs
 

> On 30 Jan 2021, at 09:10, Erfan Vi <erfan1994vi at gmail.com> wrote:
> 
> Dear all,
> I am an SPM user and not familiar with the Fieldtrip toolbox and have a resting state multimodal MEG and EEG data without any trial definition for epoching.  What I want to know is that can I use ICA artifact correction for my multimodal continuous data (10 minutes recorded data) and get acceptable results? I have performed SVD artifact correction in SPM but the output wasn't good. I have used the EEGLAB toolbox just for EEG and the result wasn't acceptable again, because I think EEGLAB performance depends on epoching the data but it's crucial for me to maintain the continuous form. Does the same condition true about the Fieldtrip?
> Best regards,
> Erfan
> _______________________________________________
> fieldtrip mailing list
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> https://doi.org/10.1371/journal.pcbi.1002202