[Eeglablist] SASICA 1.3.1 now available

[Ghuman, Avniel]

Dear Imam,

Generally this is done the same way as for task data. Either by visual inspection, or by measuring skew and kurtosis. Non-neural signals tend to be non-normal (high skew or kurtosis). HTH

Best wishes,
Avniel

On Jul 22, 2016, at 3:29 AM, Maximilien Chaumon <maximilien.chaumon at gmail.com<mailto:maximilien.chaumon at gmail.com>> wrote:

Dear Iman,

I have little experience with resting state EEG, but here's my hunch: clean neural components should have the same properties as with epoched data. Namely, a smooth/dipolar topography (dipolar fits explaining a lot of variance is a good indicator), relatively large amplitude (i.e. they should be ranked among the first components), they could also have a peak at a physiological frequency (alpha, beta...) and should generally rank low on the artefact measures returned by SASICA.
SASICA is designed to spot specific artifactual components, so it is a bit of a reverse use you want to make of it.
In the paper here<https://github.com/dnacombo/SASICA/blob/master/Chaumon_et_al_JNM_2015.pdf> he examples of Figure 3 G-I are atypical neural components that are likely to be mistaken for ocular ones. More stereotypical neural components are shown on Figure 2.

Hope this helps
Max

Le jeu. 21 juil. 2016 à 18:17, Iman Mohammad-Rezazadeh <irezazadeh at ucdavis.edu<mailto:irezazadeh at ucdavis.edu>> a écrit :
My question is how your method can separate btw neural and non neural components (such as one in Fig 3G) in resting state EEG ( if there is no event related activity)?

From: Maximilien Chaumon [mailto:maximilien.chaumon at gmail.com<mailto:maximilien.chaumon at gmail.com>]
Sent: Wednesday, July 20, 2016 11:49 PM
To: Iman Mohammad-Rezazadeh <irezazadeh at UCDAVIS.EDU<mailto:irezazadeh at UCDAVIS.EDU>>; eeglablist at sccn.ucsd.edu<mailto:eeglablist at sccn.ucsd.edu>
Cc: Rob Coben <drcoben at gmail.com<mailto:drcoben at gmail.com>>
Subject: Re: [Eeglablist] SASICA 1.3.1 now available

Hi Iman,

Thanks for your note. What is your question?


Le mer. 20 juil. 2016 à 19:32, Iman Mohammad-Rezazadeh <irezazadeh at ucdavis.edu<mailto:irezazadeh at ucdavis.edu>> a écrit :
Hi Max,
It is great.
I have a question about the toolbox and its application on continuous EEG processing , especially during resting. Examples in your published paper are based on ERP experiments however if we don’t have any sign on ERP components in the temporal activity I wonder how your algorithm/toolbox can differentiate btw neural and non-neural for something like Fig 3 G.  For example, In your paper it is obvious that for the above figure, you have got a early positive and late negative ERP components.

Best,
Iman
-------------------------------------------------------------
Iman M.Rezazadeh, Ph.D
Semel Intitute, UCLA , Los Angeles
& Center for Mind and Brain, UC DAVIS, Davis



From: eeglablist-bounces at sccn.ucsd.edu<mailto:eeglablist-bounces at sccn.ucsd.edu> [mailto:eeglablist-bounces at sccn.ucsd.edu<mailto:eeglablist-bounces at sccn.ucsd.edu>] On Behalf Of Maximilien Chaumon
Sent: Tuesday, July 19, 2016 4:15 AM
To: eeglablist at sccn.ucsd.edu<mailto:eeglablist at sccn.ucsd.edu>
Subject: [Eeglablist] SASICA 1.3.1 now available

Dear all,

A new version of SASICA, the EEGLAB extension for ICA component selection is available in the extension repository of EEGLAB (File > Manage Extensions > data processing extensions), or directly from github<https://github.com/dnacombo/SASICA/archive/master.zip>.
The associated paper can be downloaded here<https://github.com/dnacombo/SASICA/blob/master/Chaumon_et_al_JNM_2015.pdf>.

New features:
Bug fixes and improved usability.

Note: on a new install, first entering SASICA('resetprefs') in the command window is recommended.

Enjoy!
Max