[Eeglablist] Muscles artifacts

[Tarik S Bel-Bahar]

Very little published research on iclabel, one can play with thresholds,
based on task and data quality in the sample. Settings have been discussed
on eeglablist.

Usually only 10 to 20 ICs are "neural" in dense EEG, and this is normal and
generally accepted.

Researchers differ on how many and which ICs to drop. Some researchers just
remove a few (or a lot) of ICs and then analyze the cleaned sensor data. On
the other hand, researchers that analyze ICs usually find  just neural ICs
of interest that match up with well know ICs, and focus on just those
clearly neural components.

Drop ICs that are polluted with clear muscle noise. Some researchers keep
mixed components (eg alpha IC plus muscle noise) or only remove ICs that
are clearly artifactual, and keep rest.

Probably best to check follow a procedure from high quality ICA-EEG
publications.




On Tue, Sep 8, 2020, 5:35 PM Lina Ismail <linaelsherif at knights.ucf.edu>
wrote:

> So my question if I have out of 64 Ic more than half are muscles is it
> okay to remove all of them
>
> Get Outlook for iOS<
> https://urldefense.com/v3/__https://aka.ms/o0ukef__;!!Mih3wA!XIq54NYAGrp4w4e99sa1bPGpwuYSeu_AJekjCbmV4K1IiPo3a3bjUG6uNNzZRHxFzBVfxQ$
> >
> ________________________________
> From: Cédric Cannard <ccannard at protonmail.com>
> Sent: Tuesday, September 8, 2020 5:17:16 PM
> To: Lina Ismail <linaelsherif at Knights.ucf.edu>
> Cc: eeglablist at sccn.ucsd.edu <eeglablist at sccn.ucsd.edu>
> Subject: Re: [Eeglablist] Muscles artifacts
>
> Hi,
>
> You can remove the remaining IC muscles with IClabel.
> I believe it is recommended to remove ICs with threshold =>  95%
> confidence.
>
>
> Cédric
>
>
> ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
> On Tuesday, September 8, 2020 9:22 AM, Lina Ismail <
> linaelsherif at Knights.ucf.edu> wrote:
>
> > Dear Experts
> > I am applying ASR and ICA using AMICA for cleaning artifacts. I am
> following the recent Makotos's pre-processing pipeline. Any further advice
> to clean muscle artifacts as in IC label I found many IC appears as muscles.
> >
> > Eeglablist page:
> https://urldefense.com/v3/__https://nam02.safelinks.protection.outlook.com/?url=http*3A*2F*2Fsccn.ucsd.edu*2Feeglab*2Feeglabmail.html&data=02*7C01*7Clinaelsherif*40Knights.ucf.edu*7Cfa01ffc422b649e7660608d8543c958e*7C5b16e18278b3412c919668342689eeb7*7C0*7C0*7C637351966473107760&sdata=4jrmA*2FFe3tk9W7sBstDggan1h8WVQQaHFGlSpSMifZY*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUl!!Mih3wA!XIq54NYAGrp4w4e99sa1bPGpwuYSeu_AJekjCbmV4K1IiPo3a3bjUG6uNNzZRHzPbwINqg$
> > 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