[Eeglablist] **component time course

[Tarik S Bel-Bahar]

Hello Jaleh, some extra notes for you below. Remember it is generally
better to provide more details/pictures/errorInformation for better or more
specific answers. Best wishes.


***************************BEGIN EXTRA NOTES FOR JALEH

Yes, a visual or mathematical analysis of the time course of component
activations is possible to use to detect "non-neural" components.

However, there are multiple features of IC components that should be used.
See the eeglab tutorials and other teaching materials on how to select good
ICs.

See also the ICA classification website where you can train yourself how to
detect good/bad ICs.  It is developed by eeglab developers, and it also has
a feedback option in practice mode that can help you learn. Here is the
link, please be sure to check it out before moving forward:
*http://reaching.ucsd.edu:8000
<http://reaching.ucsd.edu:8000> *

Overall, it is important to be able to personally understand how/why some
ICs are non-neural. It is important not to be fully dependent on algorithms
or plugins (although some existing ICA classification tools are very useful
and work well for detecting classic bad ICs)

In regards to adjust, if you're having some issues, contact the Adjust
team, although the issues may be based on your data rather than adjust.
Usually adjust works well.

I would recommend checking out and also testing the several other plugins
for ICA classfication, including SASICA and MARC-ICA and others.

***************************END EXTRA NOTES FOR JALEH












On Sun, Jun 18, 2017 at 9:31 AM, Jaleh Mohammad alipoor <sugmad973 at yahoo.com
> wrote:

> Oh, l'm so sorry!
> l wanted to say l want to use  "component time course" through
> "plot>component activations (scroll)" to remove bad components not bad
> channels.
> now could you please tell me is it possible? because l have some strange
> problems in running adjust plugin to find artifactual components!
>
> Best regards,
> Zhaleh
>
>
>
> On Sunday, June 18, 2017 6:06 AM, Tarik S Bel-Bahar <
> tarikbelbahar at gmail.com> wrote:
>
>
> ​​
> Hello Jaleh, Notes below
> ​,
> ​
> best wishes.
>
>
> **BEGIN NOTES*******************************************************
> ********************
>
> *Since the component plots show ICs looking at their time courses cannot
> inform you about bad channels, unless, for example, the ICs are very
> focused on a single channel, or clearly have some bad channels in their
> scalp map.
>
> *One can use component-based  artifactual period rejection tools via the
> eeglab gui (which has multiple kinds of rejection similar to the suite of
> channel-based artifactual period rejection tools in the gui). Further, it
> is possible that some of these methods (component-based artifactual time
> period rejection in the eeglab gui) use the temporal IC course as part of
> the information.
> See the listing of Tools>Reject options at the page describing the eeglab
> menu options.
> https://sccn.ucsd.edu/wiki/EEGLAB_Menu_Functions
>
> *One can also check the reconstructed EEG data before and and after
> removing some components to see which components (when retained),
> contribute to a noisy channel in the rebuilt EEG data.
>
> *One could also build a new function to "guess" worst channels based on IC
> temporal information
>
>  If you haven't had a chance, please make sure to review the following
> chapter on the eeglab wiki, which is really useful for new users (along
> with the other chapters of course! :)
>
> Chapter 01: Rejecting Artifacts - SCCN
>
> https://sccn.ucsd.edu/wiki/Chapter_01:_Rejecting_Artifacts
>
> ​
> **
> ​END
>  NOTES******************************************************
> *********************
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> On Sat, Jun 17, 2017 at 6:40 AM, Jaleh Mohammad alipoor <
> sugmad973 at yahoo.com> wrote:
>
> hello,
> l want to know after running ICA, can l use "component time course"
> through "plot>component activations (scroll)" to identify and remove bad
> channels?
>
> Best regards,
> Zhaleh
>
>
> ______________________________ _________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/ eeglabmail.html
> <http://sccn.ucsd.edu/eeglab/eeglabmail.html>
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.
> ucsd.edu <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-request at sccn.ucsd.edu>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20170618/f9e3571c/attachment.html>