Hi everyone,
Our lab is using a neurofeedback script that we built in Openvibe to enhance theta/beta bands.
It would appear that the application also responds to motion artifacts such as eye blinks.
Is there functionality in Openvibe that would allow online artifact rejection?
For example, to reject any signal that exceeds 100 μV?
Thanks,
Aviv
Online Artifact Rejection
-
- Posts: 775
- Joined: Tue Dec 04, 2012 3:53 pm
- Location: INRIA Rennes, FRANCE
Re: Online Artifact Rejection
Hi Aviv, there's some discussion about implementing a threshold here,
viewtopic.php?f=11&t=9734&p=15028&hilit ... old#p15028
There's also some stuff for artifact removal,
http://openvibe.inria.fr/documentation/ ... ation.html
http://openvibe.inria.fr/documentation/ ... ising.html
Or do you mean these components don't work in online use (yes, the calibration will need data, but after that).
You could also hack with the ICA box to get what you want. Run ICA on data with eyeblinks, choose by eye the ICA channel that has the eyeblinks, and then remove it. I can't provide detailed instructions for this, but basically the main component you need is ICA and then a bit of understanding what to do with the result (basically project the data online to ICA space, set the eyeblink channel to zero, project back, using the backward and forward matrices W and A, respectively).
Happy hacking,
Jussi
viewtopic.php?f=11&t=9734&p=15028&hilit ... old#p15028
There's also some stuff for artifact removal,
http://openvibe.inria.fr/documentation/ ... ation.html
http://openvibe.inria.fr/documentation/ ... ising.html
Or do you mean these components don't work in online use (yes, the calibration will need data, but after that).
You could also hack with the ICA box to get what you want. Run ICA on data with eyeblinks, choose by eye the ICA channel that has the eyeblinks, and then remove it. I can't provide detailed instructions for this, but basically the main component you need is ICA and then a bit of understanding what to do with the result (basically project the data online to ICA space, set the eyeblink channel to zero, project back, using the backward and forward matrices W and A, respectively).
Happy hacking,
Jussi