[Eeglablist] fmrib plugin gradient correction question

[Makoto Miyakoshi]

Dear Basile,

> - having volume markers, is there any interest in switching to slice
markers (that I could add myself, the scanner being regular)? Does the
gradient template is then computed on slices and windowed on let say 30
slices instead of 30 volumes? Or does this slice splitting is already
performed in the software?

It depends on whether you have identical artifact timecourse for every
slice or not. I believe it is not the case. At least in my case, the last
slice had visibly different artifact pattern from other slices... probably
the first slice is different too. I believe using volume template (i.e.
30-slice long template).

> - I removed the 4 ECG channels recorded from the residual artifact
correction as suggested, however, there remains high-frequency content in
the signal (see screenshot
<http://s3.postimg.org/uyzdb2zgj/gradient_correction.png> with 1 ECG
channel and EEG channels) only for these channels. Is that because the
correction didn't work that well for these channels ? Is that something
that I can fix? Does low-pass filter is compulsory in this step? Why EEG
channels are not affected?

First of all, the primary purpose of having ECG (or EKG if you like a
German flavor) is to build artifact template, and what you need is just
reliable peak detection. Your data, after cleaning, seems ok for this
criterion. Low-pass filtering the ECG/EKG after cleaning is fine (I think
fMRIb apply 20Hz low-pass or something when detecting ECG/EKG peaks...sorry
if I'm wrong.) The reason why EEG channels are fine is probably that
ECG/EKG channels have longer wires and they are less physically stable than
those of EEG channels.

Makoto

On Sat, Jul 4, 2015 at 12:35 PM, basile pinsard <basile.pinsard at gmail.com>
wrote:

> Hi EEGlab users,
>
> I am quite new to EEGlab and trying to use the fMRIb plugin to correct
> from MRI artifacts.
> My questions are:
>
> - having volume markers, is there any interest in switching to slice
> markers (that I could add myself, the scanner being regular)? Does the
> gradient template is then computed on slices and windowed on let say 30
> slices instead of 30 volumes? Or does this slice splitting is already
> performed in the software?
>
> - I removed the 4 ECG channels recorded from the residual artifact
> correction as suggested, however, there remains high-frequency content in
> the signal (see screenshot
> <http://s3.postimg.org/uyzdb2zgj/gradient_correction.png> with 1 ECG
> channel and EEG channels) only for these channels. Is that because the
> correction didn't work that well for these channels ? Is that something
> that I can fix? Does low-pass filter is compulsory in this step? Why EEG
> channels are not affected?
>
> Thank you very much for your help and ideas on this.
>
> basile
>
>
> _______________________________________________
> 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/20150707/c8926dba/attachment.html>