[Eeglablist] Convincing neural dipole dipfit localized to white matter

Tarik S Bel-Bahar tarikbelbahar at gmail.com
Mon Nov 13 11:56:38 PST 2017


Hello Davida, you probably want to just refer to the nearest grey matter.
If you are working with multiple subjects, you should get a cloud of
dipoles (using study) that are near some grey matter.

There is a lot of error that is integrated into the final dipole solution
you are looking at, don't expect it to be perfect or spot on.
There is also plenty of error (usually) in source estimates. They are not
perfect solutions.
Thus overall, even a 10 MM (or more) could be expected, especially if you
have low channel count and are using an average head that is not
subject-specific.

Note that the validity of dipole localizations is dependent in part on
A. whether you have an individualized single-subject head model
B. whether you have integrated single-subject 3D head locations
C. which specific head model you are using (3-shell, 4-shell, etc..)
D. whether your channels are closely/correctly registered to the head
E. how many channels you have (64+ is better, 128+ is best).
F. I believe that dipole localizations are often usually more valid for ERP
than resting data.

You may also want to review chapters on dipole fitting and source
estimation. You amy also want to review other software that uses source
estimation (e.g., Brainstorm, MNE python).

You may also want to check out Ojeda's distributed source estimator for
eeglab that uses openMEEG. You can google that on eeglablist for recent
version.







On Sun, Nov 12, 2017 at 12:14 PM, Davida Streett <davida_s2000 at yahoo.com>
wrote:

> EEGLAB community especially dipole fitting gurus,
>     AMICA applied to my 128 channel data produced a component that
> certainly seems to correspond to a neural dipole, with a pronounced alpha
> peak, low RV%, passes all automated tests, convincing topography etc.
> (There is no left-right symmetry so I believe it should be fit by one not
> two dipoles.)  And several connectivity/causality measures in SIFT suggest
> that this component/dipole is absolutely integral to what I am studying in
> the brain.  But dipfit localizes it to white matter (and its Tailarach
> coordinates -5 -12 32 are not within 5 mm of gray matter).  I don't have
> MRI or Polhemus data, but would still like to take a stab at determining
> its most likely true location.  Any help would be very appreciated.
>     Thanks,
> Davida
>
> _______________________________________________
> 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
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20171113/fc4b7aeb/attachment-0001.html>


More information about the eeglablist mailing list