[Eeglablist] dipole fit position outside head map

[Makoto Miyakoshi]

Dear Valentina,

Ok I see one dipole ahead of the nose and another one a few centimeters
posterior to the occipital scalp.

If I were you, I would obtain the indices of the ICs in question to
investigate what's wrong with them. This is not an issue you can solve by
thinking but investigating what's wrong.
That said, I think the majority of your ICs seem sound. My initial guess is
that those minor outliers are due to sporadic poor-quality data.

> Then why only a few dipoles do lie outside the headmap?

Each bad case usually has an understandable explanation (because computers
follow rules), but there is no general explanation for these random
outliers. You look into each of these cases and sometimes you can find
systematic cause, but not always.

> Most of the dipoles correctly positioned in the right lobe, but some of
them do lie in other areas (although they are the ones with a bigger
residual variance i.e. the bigger the size of the dots the higher the
residual variance is). How can I interpret the dipoles lying outside the
ideal area of interest?

With this visualization alone, I can't say for sure.
Do you want to show me these ICs via Zoom meeting?

Makoto

On Thu, Sep 1, 2022 at 2:50 PM Barone, Valentina (UT-TNW) <
v.barone at utwente.nl> wrote:

> Dear Makoto,
>
> Thanks a lot for your answer.
>
> Here at this link you find an image of the obtained dipoles I mentioned in
> my previous post:
> https://urldefense.com/v3/__https://www.researchgate.net/post/Dipole_fit_outside_scalp_map__;!!Mih3wA!HOVY3Kh7xAHUux2XHJkZ8o3dx3ulGsZns5uu58PdrfXP9bg9190KV4vMZgjdURuyFxCXSZxYXk3ld7v4yZNM3NE$
>
> As you mentioned some dipoles are close to the left mastoid, why is so?
> I have been told the outside dipoles may be due mostly to noisy EEG, even
> though given the nature of seizures I would guess that the entire seizure
> may be considered noise by the function? Then why only a few dipoles do lie
> outside the headmap?
>
> Furthermore, to test the method I use and determine if the reconstructed
> dipoles I find were actually positioned where expected, I tried to perform
> dipole fitting also during a focal seizure in the right temporal lobe. You
> can find the result here:
> https://urldefense.com/v3/__https://ibb.co/TvpJFHp__;!!Mih3wA!HOVY3Kh7xAHUux2XHJkZ8o3dx3ulGsZns5uu58PdrfXP9bg9190KV4vMZgjdURuyFxCXSZxYXk3ld7v44dQBsSI$
>
> Most of the dipoles correctly positioned in the right lobe, but some of
> them do lie in other areas (although they are the ones with a bigger
> residual variance i.e. the bigger the size of the dots the higher the
> residual variance is). How can I interpret the dipoles lying outside the
> ideal area of interest?
>
> Thanks a lot for your help!!
>
> Best regards,
>
> Valentina Barone
>
> -----Original Message-----
> From: eeglablist <eeglablist-bounces at sccn.ucsd.edu> On Behalf Of Makoto
> Miyakoshi via eeglablist
> Sent: Saturday, 20 August 2022 02:35
> To: eeglablist at sccn.ucsd.edu
> Subject: Re: [Eeglablist] dipole fit position outside head map
>
> Dear Valentina,
>
> Sorry for the late reply. I hope my response still helps!
>
> > To plot each dipole I use the posxyz variable obtained from
> > dipfit.model
> structure.
>
> This means you are using MNI coordinates. It should be fine.
>
> If MNI-Tal confusion is not an issue, the next thing I can think of is the
> coregistration failure. In order to evaluate it, however, you should show
> me how much deviation your dipoles show from the template head. Can you
> upload a screenshot to some server and paste a link so that I can access
> it? Can you also check that your frontal dipoles are shown in the frontal
> regions instead of the locations near the left ear? Such 90 degrees
> rotation to the left can often happen.
>
> Makoto
>
>
> On Wed, Jul 27, 2022 at 9:18 AM Barone, Valentina (UT-TNW) via eeglablist
> < eeglablist at sccn.ucsd.edu> wrote:
>
> > Dear eeglab team,
> >
> > I am defining single equivalent dipole localization using the
> > dipfit_erpeeg() function with eeglab during a generalized seizure. I
> > do that in a loop for each sample (sampling freq is 256 Hz) of the
> > seizure, and then I plot all the dipoles on the MNI BME head model I
> > use to define the dipoles. Once I plot the position of my dipoles in
> > the same 3-d volume map used to define the dipoles, a couple of
> > dipoles are positioned outside the head map. To plot each dipole I use
> > the posxyz variable obtained from dipfit.model structure.
> > I do not understand how can some dipoles be positioned outside the
> > head model used to fit the dipoles. Is anybody familiar with these
> > results and can explain why this happens?
> >
> > Thanks a lot in advance for your help.
> >
> > Best regards,
> >
> > Valentina
> > _______________________________________________
> > 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
> >
> _______________________________________________
> 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
>