[Eeglablist] Bilateral dipoles

Fri Mar 27 10:49:32 PDT 2015

Dear Salim,

Well, the assumption in using ICA on EEG is that the decomposed ICs should
have dipolar scalp topos (for detail, see Delorme et al, 2012 Plos One
'Independent components are dipolar') Your IC17 is not dipolar
unfortunately, so it is not likely that IC17 reflects good brain activity.

When you create STUDY, you may want to apply 15% residual variance
threshold (by default). It's like a rule of thumb (there is indirect
evidence that supports it shown in Artoni's recent publication, but it is
not explicitly stated). You may want to exclude any ICs larger than r.v. >
15 % (or localized in outside brain).

Makoto

On Fri, Mar 27, 2015 at 3:25 AM, Salim Al-wasity <salim_alwasity at yahoo.com>
wrote:

> Dear Mr. Makoto
> Please find the attached pictures which illustrate the use of one dipole,
> two dipoles (with symmetry constrain ON) and two dipoles (with symmetry
> constrain OFF).
> My question is if the unsymmetrical dipoles not recommended, What can I do
> with such scalp map to get the best dipoles?
>
> Sincerely
> Salim
>
>
> https://drive.google.com/file/d/0B-waDFAaEHcuRmR0U19JN1VSMG8/view?usp=sharing
>
> https://drive.google.com/file/d/0B-waDFAaEHcudlpkc3NDaUtueVU/view?usp=sharing
> 1 dipole.JPG
> <https://drive.google.com/file/d/0B-waDFAaEHcudloyTXpGMmtTWFk/view?usp=sharing>
>
>
> [image: image]
> <https://drive.google.com/file/d/0B-waDFAaEHcudloyTXpGMmtTWFk/view?usp=sharing>
>
>
>
>
>
> 1 dipole.JPG
> <https://drive.google.com/file/d/0B-waDFAaEHcudloyTXpGMmtTWFk/view?usp=sharing>
> View on drive.google.com
> <https://drive.google.com/file/d/0B-waDFAaEHcudloyTXpGMmtTWFk/view?usp=sharing>
> Preview by Yahoo
>
> without symmetry.JPG
> <https://drive.google.com/file/d/0B-waDFAaEHcud0VObmkySzQwUEE/view?usp=sharing>
>
>
> [image: image]
> <https://drive.google.com/file/d/0B-waDFAaEHcud0VObmkySzQwUEE/view?usp=sharing>
>
>
>
>
>
> without symmetry.JPG
> <https://drive.google.com/file/d/0B-waDFAaEHcud0VObmkySzQwUEE/view?usp=sharing>
> View on drive.google.com
> <https://drive.google.com/file/d/0B-waDFAaEHcud0VObmkySzQwUEE/view?usp=sharing>
> Preview by Yahoo
>
>
>
>
>
>
>
>
>
>
>   On Thursday, 26 March 2015, 18:01, Makoto Miyakoshi <mmiyakoshi at ucsd.edu>
> wrote:
>
>
> No, that is not recommended. If you put more dipoles, residual variances
> become always smaller, which is meaningless. You have no reason to justify
> the two dipoles in terms of ICA; ICA usually gives you dipolar scalp topos
> for true EEG sources...
>
> Makoto
>
> On Sat, Mar 21, 2015 at 4:39 AM, Salim Al-wasity <salim_alwasity at yahoo.com
> > wrote:
>
> Dears
> I ran the 'fine_fit' with 'symmetry constrain for dipole' is OFF. I did
> that to get lower RV value.
>
>
>
>   On Thursday, 19 March 2015, 17:48, Makoto Miyakoshi <mmiyakoshi at ucsd.edu>
> wrote:
>
>
> Dear Salim,
>
> Well, I'm concerned with the two dipole fit results. They don't seem
> symmetrical. How did you obtain the current positions?
>
> Makoto
>
> On Thu, Mar 5, 2015 at 1:28 AM, Salim Al-wasity <salim_alwasity at yahoo.com>
> wrote:
>
> Dear Mr. Makoto
> Thanks for your reply. I attached a picture of bilateral dipoles obtained
> by (Locate Dipoles using DIPFIT2> Plot component dipoles) and it shows only
> only coordinate. However (Locate Dipoles using DIPFIT2> Fine fit) shows two
> coordinates. Which coordinates are the correct one?
>
> Sincerely
> Salim
>
> dipoles.JPG
> <https://drive.google.com/file/d/0B-waDFAaEHcudzlhTzJEU21xUUk/view?usp=sharing>
>
>
> [image: image]
> <https://drive.google.com/file/d/0B-waDFAaEHcudzlhTzJEU21xUUk/view?usp=sharing>
>
>
>
>
>
> dipoles.JPG
> <https://drive.google.com/file/d/0B-waDFAaEHcudzlhTzJEU21xUUk/view?usp=sharing>
> View on drive.google.com
> <https://drive.google.com/file/d/0B-waDFAaEHcudzlhTzJEU21xUUk/view?usp=sharing>
> Preview by Yahoo
>
>
> fine fit.JPG
> <https://drive.google.com/file/d/0B-waDFAaEHcuRGdyNm9IUFg5azA/view?usp=sharing>
>
>
> [image: image]
> <https://drive.google.com/file/d/0B-waDFAaEHcuRGdyNm9IUFg5azA/view?usp=sharing>
>
>
>
>
>
> fine fit.JPG
> <https://drive.google.com/file/d/0B-waDFAaEHcuRGdyNm9IUFg5azA/view?usp=sharing>
> View on drive.google.com
> <https://drive.google.com/file/d/0B-waDFAaEHcuRGdyNm9IUFg5azA/view?usp=sharing>
> Preview by Yahoo
>
>
>
>
>   On Wednesday, 4 March 2015, 15:50, Salim Al-wasity <
> salim_alwasity at yahoo.com> wrote:
>
>
> Dear Mr. Makoto
> Thanks for your reply. I attached a picture of bilateral dipoles obtained
> by (Locate Dipoles using DIPFIT2> Plot component dipoles) and it shows only
> only coordinate. However (Locate Dipoles using DIPFIT2> Fine fit) shows two
> coordinates. Which coordinates are the correct one?
>
> Sincerely
> Salim
>
>
>   On Tuesday, 3 March 2015, 2:03, Makoto Miyakoshi <mmiyakoshi at ucsd.edu>
> wrote:
>
>
> Dear Salim and Arno (cc Caterina)
>
> Sorry for the slow response.
>
> > Therefore I used bilateral dipoles to reduce the RV.
>
> Use bilateral dipole only when you see clear bilateral scalp topography!
>
> > 1- In (plot component dipoles) options, it gives me only one
> coordination ever if I have two dipoles. How can I know where these two
> dipoles are?
>
> Really? That sounds weird to me. You should close the fine fit window by
> pressing 'ok' and not 'x' to update the data.
>
> >2- Using a PCA clustering approach, I might find one cluster located at
> the right hemisphere but it has some dipoles at the left side (bilateral
> dipoles), Is this cluster correct?
>
> Yes, that should be correct. To be honest I don't know how STUDY
> clustering handles dual dipoles, and I (and Caterina also) have wanted to
> know about it. Arno?
>
> Makoto
>
> On Tue, Feb 24, 2015 at 7:08 AM, Salim Al-wasity <salim_alwasity at yahoo.com
> > wrote:
>
> Dears
> After decompose my EEG data to components using ICA and estimate their
> dipoles, some of these dipoles have high residual variance(RV). Therefore I
> used bilateral dipoles to reduce the RV.
> 1- In (plot component dipoles) options, it gives me only one coordination
> ever if I have two dipoles. How can I know where these two dipoles are?
> 2- Using a PCA clustering approach, I might find one cluster located at
> the right hemisphere but it has some dipoles at the left side (bilateral
> dipoles), Is this cluster correct?
>
> Sincerely
> Salim
>
>
>
> _______________________________________________
> 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
>
>
>
>
> --
> Makoto Miyakoshi
> Swartz Center for Computational Neuroscience
> Institute for Neural Computation, University of California San Diego
>
>
>
>
>
>
>
> --
> Makoto Miyakoshi
> Swartz Center for Computational Neuroscience
> Institute for Neural Computation, University of California San Diego
>
>
>
>
>
> --
> Makoto Miyakoshi
> Swartz Center for Computational Neuroscience
> Institute for Neural Computation, University of California San Diego
>
>
>

-- 
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20150327/33c60bd6/attachment.html>