[Eeglablist] Source localization with minimum norm

[Fernando R. Santos]

Dear Victoria,

The usual standard for estimating the localization of a single source in minimum
norm solutions is to look for the source maximum.
This is how their performance has usually been tested with simulated data, by
measuring the distance between the simulated dipole and the MN maximum (= dipole
localization error).

However, this may be seen as an unnatural way to evaluate distributed inverse
solutions, because one of their strengths should be the possibility of
identifying more than one source.
A method to to study multiple sources is the resolution matrix (see Michel et al
(2004). EEG source imaging. Clinical Neurophysiology, 115, 2195-2222. for some
info and references on this).

Hope this is helpful in any way!
Best,

-- 
Fernando R. Santos

Neuropsychophysiology Laboratory
Faculty of Psychology and Education Sciences
University of Porto


> Hello all,
>
>
> I'm using the minimumnormestimate() function from the Forwinv toolbox with a
> four-shell spherical head model, and assuming orthogonal dipoles orientation.
> When applying this function, I obtain the strengths and power at each source
> location. If my objective is choosing a single source, which procedure should I
> use after estimation of these current densities?
>
> Thanks in advance,
>
>
> Victoria Eugenia Montes
> Universidad Nacional de Colombia - Manizales




-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.