[Eeglablist] Baseline issues, trying again, no pictures this time :-)

[Makoto Miyakoshi]

Dear Ilaria,

I know you tried to post it several times because of the size limitation
for the attachment. Thank you for your patience.

It is possible that what you expect for 'baseline correction' is actually a
high-pass filtering.
Try to high-pass filter at 1 Hz with default EEGLAB FIR filter (which is
0.5 Hz at -6dB) and see the plot. If you find the result satisfactory to
you, it means you actually needed high-pass filtering.

Baseline correction on continuous data only performs mean subtraction. For
example, if you specify the entire data length for 'baseline correction',
then the whole data will be now zero-sum. But any local time window, such
as your one-page plot data length, can have large deviation from zero line.

Makoto

On Sat, Jul 4, 2020 at 6:11 PM Ilaria Berteletti <
ilaria.berteletti at gallaudet.edu> wrote:

> Hello again,
>
> I am hoping for some insight on why my baseline removal is not working.
> This time, I am adding a step-by-step explanation of what I'm doing.
> (pictures available if needed)
>
> The file I start with is a file that was imported from Brain Vision and
> then re-referenced to the mastoids.
> When I open to check the "scroll" plot, it appears that a baseline
> correction is needed, based on my understanding of Arnaud's tutorials, as
> all channels are wide apart and off the plot (ref: 'EEGLAB preprocessing
> part 4: Filtering' video).
> Using the "remove DC offset" from the plot options, the traces do improve
> and nicely show.
> I then unflag the plot option (just to make sure although it should have
> not changed anything to the file itself) and select the "remove baseline"
> option from the drop menu.
> If it is relevant, after I click 'OK', I do see the pop-rmbase process show
> up in my matlab command window but *no saving GUI pops up *(this shows up
> in the video tutorial, maybe differences in version?).
>
> ---Command Window text showing up ----
> pop_rmbase () : Removing baseline ...
> pop_rmbase () : finding continuous data discontinuities
> Done.
> -------------------------------------------------------
>
> To be sure everything is saved, I manually go in the drop down menu to save
> the new file adding a _BL to the end.
> When I visualise the "scroll" plot on the newly saved file, nothing has
> changed.
>
> Again, my understanding of Arnaud's tutorial was that this operation would
> be equivalent to the Remove DC offset option found in the plot menu and
> appears to be an important step before filtering.
>
> my question is:
> why is this happening?
> where am I going wrong?
> is the issue in the original file elsewhere?
>
> I have tried running this as a script by saving the history of the baseline
> removal and it does not seem to do much.
>
> I would really appreciate any insight as I don't want to simply go on and
> process data pretending all is good. Maybe the step is not necessary or the
> problem is elsewhere. I'd really like the advice from someone who has more
> experience than myself
>
> Thank you,
> ilaria
>
>
>
> *Assistant Professor, **PhD in Educational Neuroscience (PEN) Program*
> *Director of the PEN Distinguished Lecture Series **and University
> Partnerships (MOUs)*
> *Director of the Numeracy and Educational Neuroscience Laboratory*
> *Gallaudet University*
> *Sorenson Language and Communication Center, 1223*
> *800 Florida Ave, N.E.*
> *Washington, D.C. 20002*
> *USA*
> *Tel: 202 651 5673*
> _______________________________________________
> 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
>