[FieldTrip] Help: sourceanalysis

[Eelke Spaak]

Hi Valentina,

The computation of the beamformer spatial filter only uses the
covariance matrix, and does not use the average data itself. Since you
have the covariance matrix computed "manually", you should be able to
wrap it in a FieldTrip-style timelock structure: tl = []; tl.cov =
cov; and put some zeroes of appropriate size in tl.avg. Then, make
sure to use cfg.lcmv.keepfilter = 'yes' in the call to
ft_sourceanalysis, and get the filters out from the resulting source
structure. (Note that the dipole moments in the source structure will
be meaningless since the avg data was fake; but the filters are still
meaningful.) After that, simply multiply your sensor-level resting
state data with the spatial filters to get the source time courses.

I hope that helps. (If you have no idea what I'm talking about, then
it's probably a good idea to go through some tutorials and examples
(including about so-called "virtual channels") anyway using
event-related fields before attempting this on resting state data.)

Cheers,
Eelke

On Sun, 13 Jan 2019 at 16:45, Valentina Nasato <vale.nasato at gmail.com> wrote:
>
> Good evening everyone,
>
> through the fieldtrip toolbox I solved the direct problem using the FEM method for EEG data. The EEG data are in resting state. Now I have to implement the indirect problem but on the tutorial there is no guide. How should I implement fieldtrip sourceanalysis? I have the sourcemodel, headmodel, leadfield and covariance matrix. I calculated the covariance matrix directly in Matlab.  My problem now is to understand how to calculate the average of the EEG data to be given as input to the sourceanalysis. How to calculate an average of resting state data ?
>
> thanks
>
>
>
>
>
> Inviato da Posta per Windows 10
>
>
>
> _______________________________________________
> fieldtrip mailing list
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> https://doi.org/10.1371/journal.pcbi.1002202