[Eeglablist] about using the tip of nose as reference channel

[Makoto Miyakoshi]

Dear Hui-bin,

> How to remove the VEOG & HEOG signals perfectly without losing the
genuine brain signals?

We recommend ICA (of course, since you are posting to EEGLAB mailing
list!) to separate VEOG and HEOG from EEG. See DElorme et al. (2007)
NeuroImage for artifact rejection using ICA. However, it seems naive to ask
for 'perfect' separation. What do you mean, and how do you define, the
'perfect' separation if we don't have ground truth measured?

> 1.       If I want to use ICA to remove artifacts, is it a good idea that
the VEOG channel is excluded before run ICA ?  The HEOG channel is placed
on the tip of nose, and the VEOG channel is placed below the eyelid of the
left eye.

Of course yes.

> 2.        Before run ica, the continuous EEG data was segmented into
stimulus-locked ERP epochs of 3000ms(-1000ms to 2000ms) and filtered at
1~30Hz. Is it necessary that the epochs is baseline corrected?

Do not correct baseline. See Groppe et al. (2010) NeuroImage paper for
'split half...'

> 3.       When should I re-reference the channels to the channel on the
tip of nose(i.e., the HEOG channel)? Before run ICA or after the
independent components related to VEOG &HEOG is removed?

Before ICA and after bad-channel rejection. Re-reference to average, don't
use nose tip unless you want to replicate the published results. Don't
remove EOG channels.

> 4.       How to efficiently identify the ICs that related to VEOG & HEOG.
Is it a good idea to calculate the Pearson correlation coefficients between
ICs and VEOG channel ?

You need either experience or Nima's eyeCatch() that is included in his
Measure Projection Toolbox (Bigdely-Shamlo et al., 2013 NeuroImage). It's
not that difficult, and you should also learn it too anyways.

Makoto

On Tue, Oct 21, 2014 at 5:24 AM, Hui-bin Jia <420247417 at qq.com> wrote:

> Hi, all
>
>
>
>       In my ERP experiment, all EEG electrodes were referenced online to
> FCz. But during the preprocessing of EEG data, they should be
> re-referenced to the channel on the tip of nose.
>
>
>
>       Because the tip of nose is in the vicinity of eyes, it has a lot of
> VEOG signals. If VEOG and HEOG signals are partly removed by ICA, using
> this channel as reference
>
> may seriously affect the result. So in order to gain reasonable ERP
> waveforms, the VEOG & HEOG signals should be perfectly removed.
>
>
>
>
>
>     Could someone give me some tips and cautions during the preprocessing?
> How to remove the VEOG & HEOG signals perfectly
>
> without losing the genuine brain signals?
>
>
>
>      Besides I have the following questions:
>
> 1.       If I want to use ICA to remove artifacts, is it a good idea that
> the VEOG channel is excluded before run ICA ?  The HEOG channel is placed
> on the tip of nose, and the VEOG channel is placed below the eyelid of the
> left eye.
>
> 2.        Before run ica, the continuous EEG data was segmented into
> stimulus-locked ERP epochs of 3000ms(-1000ms to 2000ms) and filtered at
> 1~30Hz. Is it necessary that the epochs is baseline corrected?
>
> 3.       When should I re-reference the channels to the channel on the
> tip of nose(i.e., the HEOG channel)? Before run ICA or after the
> independent components related to VEOG &HEOG is removed?
>
> 4.       How to efficiently identify the ICs that related to VEOG & HEOG.
> Is it a good idea to calculate the Pearson correlation coefficients between
> ICs and VEOG channel ?
>
>
>
> Any reply from you will be highly appreciated!
>
>
>
> Hui-bin Jia
>
> Faculty of Psychology, Southwest University, China
>
> _______________________________________________
> 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
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20141023/fd5b582e/attachment.html>