questions on ICA to remove EOG artifacts from MEG data

Thu Mar 18 19:26:20 CET 2010

Dear Elisabeth,

I am a person who uses the ICA to Neuromag data. As far as I understand your data, at least one blink component should be found with the typical blink-topography. My first impression is that you might separate latitude and longitude gradiometers incorrectly in plotting topographies. Have you already obtained reasonable topographies with your script? If it is not the case, it will be difficult for me to give you appropriate suggestions.

>    * Is it correct to do the ICA on both horizontal and vertical
>      gradiometers together and then plot them sepparately to identify
>      artifacts?

I don't see any theoretical reason why we must do separately. Indeed, I apply ICA both gradiometers (GMs) and magnetometers (MMs) together. As far as I examined details for few subjects' results, its performance becomes more stable than ICA to GMs alone. 
To apply ICA together, GMs and MMs must be I normalised separately before ICA, since the scale of signals is different between these types of sensors and Neuromag data often need PCA before ICA (Please see my response to the next question). After the ICA, I inversely normalized to recover the original scales. 

>    * Should I do the PCA before the ICA and if so, how many principal
>      components should I retain?

I would say YES to users of 306ch Neuromag system. Without reducing the number of decomposing components, ICA results often contain complex values. Probably it is because the number of components to be decomposed (306) is too large relative to the number of actual components in MEG data. I always use PCA to reduce the components from 306 to around 80. If you applied SST in addition to SSS using Maxfilter (Neuromag preprocessing software), you may need to reduce more. 

>    * Should I do the ICA on both conditions together or sepparately
>      (since I expect different sources to be activated in both
>      conditions, e.g. one of the two primary somatsensory cortices
>      activated according to stimulation of the right or left hand)?

Arguments on this issue will be very similar to those on the optimization of spatial filter in beamformer (common across conditions <--> separately). You will find out advantages/disadvantages on the fieldtrip website. I use common ICA, since it avoids a risk of removing different noise components from different conditions.      

>    * Is it accurate enough to use complete trials for the component
>      analysis or do I have to extract and use just the parts containing
>      EOG artifacts?

I would recommend doing the standard method explained on the fieldtrip website: Decompose MEG data of whole trials and get topography components -> Decompose the segmented data using the topography components. 

>    * And finally, are the component topographies usually sufficient to
>      identify eye artifacts?

I think the usage of topographies alone does not highlight advantages of ICA. ICA decomposes data so that temporal independency between components becomes maximum, whereas PCA decomposes so that topographies of components are orthogonal to each other. Therefore, the timing of EOG noises is much helpful to identify in ICA. 

Sincerely yours,
Masaki Maruyama

>-----Message d'origine-----
>De : FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] De la part
>de Elisabeth May
>Envoyé : jeudi 18 mars 2010 09:51
>À : FIELDTRIP at NIC.SURFNET.NL
>Objet : [FIELDTRIP] questions on ICA to remove EOG artifacts from MEG data
>
>Dear Fieldtrip users,
>
>I am trying to use independent component analysis (ft_componentanalysis)
>to remove eye artifacts from my MEG data. The dataset was recorded with
>the Neuromag 306 system and contains two conditions with 40 trials each
>(one where a painful laser stimulation was applied to the subject's
>right hand and one where the same stimulation was applied to the left
>hand). So far, I am trying to do the ICA just for the gradiometers on
>preprocessed data (after removal of bad channels) and then plotting the
>topographies of the components to identify and then reject components
>representing the eye artifacts (as suggested in the example script on
>the Fieldtrip webpage). I have used different settings with the default
>option cfg.method = 'runica' (no PCA before ICA, different numbers of
>principal components to be retained (cfg.runica.pca), both conditions
>sepparately and together...). However, I have never gotten component
>topographies that I can clearly identify as components representing eye
>artifacts (showing two sources right above the eyes), although there are
>almost 200 eye artifacts in the 80 trials of this data set (i.e. there
>was usually more than one EOG artifact in every single trial). So I am
>still wondering about a few questions on the exact settings for the ICA
>in my case and if I am missing something really obvious:
>
>    * Is it correct to do the ICA on both horizontal and vertical
>      gradiometers together and then plot them sepparately to identify
>      artifacts?
>    * Should I do the PCA before the ICA and if so, how many principal
>      components should I retain?
>    * Should I do the ICA on both conditions together or sepparately
>      (since I expect different sources to be activated in both
>      conditions, e.g. one of the two primary somatsensory cortices
>      activated according to stimulation of the right or left hand)?
>    * Is it accurate enough to use complete trials for the component
>      analysis or do I have to extract and use just the parts containing
>      EOG artifacts?
>    * And finally, are the component topographies usually sufficient to
>      identify eye artifacts?
>
>Any help or advise on this would be greatly appreciated! Thanks in advance,
>Elisabeth
>
>
>--
>Dipl.-Psych. Elisabeth May
>
>Universitätsklinikum Düsseldorf
>Institut für Klinische Neurowissenschaften und Medizinische Psychologie
>Universitätsstr. 1
>40225 Düsseldorf
>
>Tel: +49 211 81-18075
>
>http://www.uniklinik-duesseldorf.de/med-psychologie
>
>----------------------------------
>The aim of this list is to facilitate the discussion between users of the
>FieldTrip  toolbox, to share experiences and to discuss new ideas for MEG
>and EEG analysis. See also
>http://listserv.surfnet.nl/archives/fieldtrip.html and
>http://www.ru.nl/neuroimaging/fieldtrip.

----------------------------------
The aim of this list is to facilitate the discussion between users of the FieldTrip  toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/neuroimaging/fieldtrip.