[Eeglablist] Trouble inserting Boundary Events at Selecting data

Tarik S Bel-Bahar tarikbelbahar at gmail.com
Sat Oct 21 11:58:52 PDT 2017


HI Augstin. here are notes for you below, best wishes.



**************NOTES for AGUSTIN****************************

First off, make sure you are attempting all steps and checks fully from the
gui, to begin with. That includes all the steps listed further below.

Also, it is important to make sure you tried all this eeglab tutorial data
already, following eeglab tutorial steps, and that you had success, Using
the egelab tutorial data as a proof-of-concept and practice is important
for people new to eeglab.

Please following up after reviewing the below, so that other users on the
list can learn from your experiences.

You're probably doing the steps incorrectly, or there is something wrong
with you matlab/eeglab setup.

Look in urevents field of the EEG structure, there are usually a "start
boundary and end boundary there.

However in the normal events structure, after properly rejecting a period
of data, there should be just one boundary event,
See the online tutorials on visual rejection if you have not already. See
also the events structure and processing tutorials.

Try also the following in a specific sequence. All steps should be done via
GUI.

1. Open a new .set file (preferably one of the eeglab tutorial files).
2. View the data in continuous mode.
3. Select one portion of data that you want to reject (any period, make it
~10 seconds, for example)
4. Click the reject button in the GUI of the continuous EEG plot.
It is important here to make sure the rejection actually happens.
5. Make sure that the EEG structure in matlab is updated.
6. Save the dataset with a new name.
7. Clear all datasets from eeglab.
8. Load up the file you saved.
9. Plot the continuous data. Confirm or disconfirm that you see a boundary
event where the period you deleted used to be.
If you delete/reject two different periods you should see two boundary
events.

Note that you You need to select and "make sure it's rejected" (not just
assume,,,,so check that the whole data got shorter in the eeglab gui after
your rejection...did it decrease by the amount of time you rejected ?

Also, try be more clear about exactly where you looking when you look for
the "events". Best way for you is via both "plot conitnuous data" and also
via the Edit > Events GUI.

If you select one period, and also select a second period, and then reject
both those periods, you should end up with two boundary events, one for the
first selection, and one for the second selection. I am pretty sure you are
just not rejecting or saving/updating correctly, otherwise you would see
the boundary events created by the rejection of periods. This is something
that works normally and very smoothly, you should not be having this issue.


Note that This may be a bug, and if you find no solution and it is
replicable, then you should report it to eeglab bugzilla.

In future sendouts, send pics and possibly the file itself that you are
working with,
Also send screenshots of your steps.









On Tue, Oct 17, 2017 at 10:52 AM, Agustín Solano <
asolano at bioingenieria.edu.ar> wrote:

> Thanks Tarik,
>
> I followed your recommendation bue I am still not having what I think I
> should have.
>
> I loaded a dataset without any boundary event and then selected in the GUI
> a portion of data, for example, points [10000 100000]. If I do that I think
> eeglab should insert two events, one at the beginning and other to the end,
> because I should be able to track if any data has been rejected. Is that
> right? But when I do that no events are inserted.
>
> On the other hand, if I select two portions of data, I get an event
> corresponding to the rejected data between the two portions.
>
> Maybe it is the way eeglab works but I am not sure I was clear at my first
> question.
> Thanks again!
>
> 2017-10-15 1:10 GMT-03:00 Tarik S Bel-Bahar <tarikbelbahar at gmail.com>:
>
>> Hello Agustin.
>>
>> don't worry, one should be having no problem with this, rejecting of
>> continuous works smoothly in eeglab.
>>
>> First reject the periods (pressing okay/reject in the GUI) and then make
>> sure the file gets updated in memory (review eegh output after you do the
>> rejection). Then reopen and view the continuous data. Do you see the
>> boundary break events now ?
>>
>> Look into ur.epochs part of the EEG structure, which should contain start
>> and end times for boundaries. I believe you can google tutorial information
>> on urevent fields, etc...
>>
>> You should see the boundary breaks. Make sure to test this all with the
>> GUI first, and if you are trying to use code instead of the gui, review the
>> eegh output after various steps.
>>
>> Let us know of your progress.
>>
>>
>>
>> On Fri, Oct 13, 2017 at 1:13 PM, Agustín Solano <
>> asolano at bioingenieria.edu.ar> wrote:
>>
>>> Dear community,
>>>
>>> I am working woth continuous sleep data. I have to Select Data
>>> corresponding to different sleep stages and I need to track changes. Doing
>>> some tests I realized that if I select a portion of continuos data I don't
>>> get the Boundaries Events at the begining and at the end of the selected
>>> portion (corresponding to the rejected data).
>>>
>>> In matlab Command Window I see the output (for example):
>>>
>>> *  eeg_insertbound(): 2 boundary (break) events added.*
>>> *  eeg_checkset note: upper time limit (xmax) adjusted so
>>> (xmax-xmin)*srate+1 = number of frames*
>>> *  eeg_insertbound(): 2 boundary (break) events added.*
>>> *  Creating a new ALLEEG dataset 3*
>>>
>>> But EEG.event is empty..
>>>
>>> Any ideas? Am I missing something?
>>> Thanks in advance.
>>>
>>> --
>>> *Bioing. Agustín Solano*
>>> Instituto de Fisiología y Biofísica (IFIBIO) - Bernardo Houssay
>>> Laboratorio de Fisiología de la Acción, Facultad de Medicina (UBA)
>>> Paraguay 2155, Capital Federal
>>> Buenos Aires, C1121ABG
>>> Argentina
>>> Tel: 54 11 5 950 9500 (2132)
>>> http://www.physiologyofactionlab.info
>>>
>>> _______________________________________________
>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.uc
>>> sd.edu
>>> For digest mode, send an email with the subject "set digest mime" to
>>> eeglablist-request at sccn.ucsd.edu
>>>
>>
>>
>
>
> --
> *Bioing. Agustín Solano*
> Instituto de Fisiología y Biofísica (IFIBIO) - Bernardo Houssay
> Laboratorio de Fisiología de la Acción, Facultad de Medicina (UBA)
> Paraguay 2155, Capital Federal
> Buenos Aires, C1121ABG
> Argentina
> Tel: 54 11 5 950 9500 (2132)
> http://www.physiologyofactionlab.info
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20171021/dbbd52c9/attachment-0001.html>


More information about the eeglablist mailing list