[Eeglablist] pre-processing for sLORETA and connectivity analysis

[Makoto Miyakoshi]

Dear Mina,

> I wonder if the filter I will choose for the pre-processing of my EEG
data will affect significantly the result of sLORETA source localization
and connectivity analysis afterwards.

I've never used sLORETA but I guess you need to choose a single frame from
which inverse problem is solved... if this is the case, filtering as well
as averaging window length (and timing of course) would significantly
affect the result.

> So far I use IIR filter as EEGLAB plugin, with a 0.5-40Ηz band-pass and
all the default parameters. Do you think this should be OK or should I use
another filter?

I don't use IIR myself but unless your data are horribly screwed up using
IIR should be fine.

> I've seen Makoto's notes on filtering and connectivity analysis but it
addresses Granger causality in particular, and in my case we are interested
in functional connectivity analysis so I am not so sure if it applies.

Unless you do multivariate Granger, I recommend you perform our standard
FIR 1-50 (or 100, 250 etc) filter. Note that EEGLAB's '1-Hz filter' is 0.5
Hz cutoff at -6dB.

> Also I would like to ask you for the same analysis framework, whether a
Surface Laplacian filtering, as a final step in the pre-processing, works
well with sLORETA or it is something I should avoid.

The assumption in using surface laplacian is that EEG sources are all
radial, which I don't understand. I've seen many tangential scalp
topography in ICA results. But, that being said, it isn't a terribly bad
idea to apply it since there are much more radial sources.

Good luck!

Makoto


On Fri, Jan 8, 2016 at 8:44 AM, Mina Marmpena <mina.marmpena at gmail.com>
wrote:

> Dear EEGLAB list,
>
> I wonder if the filter I will choose for the pre-processing of my EEG data
> will affect significantly the result of sLORETA source localization and
> connectivity analysis afterwards.
>
> So far I use IIR filter as EEGLAB plugin, with a 0.5-40Ηz band-pass and
> all the default parameters. Do you think this should be OK or should I use
> another filter? I've seen Makoto's notes on filtering and connectivity
> analysis but it addresses Granger causality in particular, and in my case
> we are interested in functional connectivity analysis so I am not so sure
> if it applies.
>
> Also I would like to ask you for the same analysis framework, whether a
> Surface Laplacian filtering, as a final step in the pre-processing, works
> well with sLORETA or it is something I should avoid.
>
> Finally, I was wondering if you could suggest some papers or resources
> with sLORETA and connectivity analysis that could help me to better
> understand how to deal with this kind of analysis, especially how to derive
> a connectivity measure from the current densities.
>
>
> Thank you in advance,
> Mina
>
> _______________________________________________
> 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
>



-- 
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/20160111/bef7052f/attachment.html>