[FieldTrip] reconstructing EEG using leadfield (Beamformer)

[Eelke Spaak]

Hi Tyler,

Beamforming is a scanning method, which means that the power/current
density at each grid point is estimated independently from all other
grid points. In general one would not expect sum_i(leadfield_ix *
pow_i), where i indexes grid points, to faithfully reproduce the
potential measured at sensor x. This as opposed to something like e.g.
minimum-norm estimate, where the estimation is done in one
regression-like step, rather than by scanning grid points.

Best,
Eelke

On 20 August 2014 04:47, Tyler Grummett <tyler.grummett at flinders.edu.au> wrote:
> Hello fieldtrip experts,
>
> I was just wondering how you would reconstruct EEG using the leadfield, to try and verify how well it is modelling the data.
>
> I have tried multiplying the source.avg.mom by the source.leadfield.
> I wont clutter the email with code, however I can once requested.
>
> I had a look at the spectra of the EEG and the spectra of the reconstructed EEG and they look   quite different.
>
> Any help would be greatly appreciated.
>
> Tyler
>
> Sent from my iPad
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip