can OpenBCI train Slow Cortical Potentials ?
I'd like to train SCPs (slow cortical potentials) and I am looking for an affordable solution.
Do I understand correctly that neither OpenEEG nor Muse can do that because they are not DC coupled?
Do you know of any projects who were successful in doing that?
On top of the hardware it seems that it also needs support on the software side, is there any open/affordable software which supports it?
Thank you
Comments
Hi IFS,
The ADS1299 analog to digital converter used in Cyton, IS DC coupled. So can be used for SCP or other related protocols. There are many threads here on the Forum and Docs covering 'neurofeedback' applications. SCP is a type of neurofeedback.
https://www.google.com/search?as_q=neurofeedback&as_sitesearch=openbci.com
William
https://www.google.com/search?as_q=scp&as_sitesearch=openbci.com
Hi,
yes SCP is a specific type of neurofeedback, I didn't include that explanation because I assumed that to be common knowledge in this forum. I searched for SCP or slow cortical potentials in this forum and the last entry is three years old and rather unspecific, thus my question. Have a look at the links you posted, they don't answer my question.
Cyton is a board you can buy here, which is still 1k Euro. Definitely better than 6k, but in comparison OpenEEG is 100 Euro. Are there existing openBCI v3 compatible boards[?]
And then we still have the software question open: Assume I have a cyton board, how do I do SCP training?
The post I referred to in the above comment is https://openbci.com/forum/index.php?p=/discussion/2342/neurofeedback-for-adhd – in which you provide some info, but it's rather difficult to get all pieces together.
So from what I understand: One needs to install the Cyton GUI software, then this connects to BioEra (https://docs.openbci.com/Software/CompatibleThirdPartySoftware/BioEra/, $250), and then you speak of BioExplorer (https://brainmaster.com, $575) but linking to Brain-Trainer ($1300). Neither of them writing something about SCP in their description.
IFS, hi again.
re: "OpenBCI compatible boards". Any boards you see from other websites advertising "compatibility" are not supported by OpenBCI software or firmware. Thus are not recommended.
re: "not answering your question". The initial reply above did suggest searching on 'neurofeedback'. There are many options as you have determined. BioEra is indeed the platform used by the Othmer's Cygnet system, which is a DC coupled ILF Infra-Low Frequency neurofeedback training. Not the same as SCP, but SCP was a precursor to the other ILF / ISF protocols.
If you are a programmer in languages such as Python, C++, C#, Java, etc, you could write your SCP trainer yourself in that language, getting the raw data stream from the Cyton using the Brainflow library:
https://brainflow.org/
https://brainflow.readthedocs.io/en/stable/Examples.html
A similar app to BioEra, but free is 'neuromore'. Another is 'BrainBay'. All of these (BioEra, BrainBay, neuromore, OpenViBE, Bioexplorer, LabView), are all examples of VPL, Visual Programming Languages, which are programmed via graph-like data-flow diagrams. Starting with the input block / element, the data stream from the amplifier, then flowing through various computation and filtering elements until the final feedback is provided.
Bioexplorer was never converted to receive data from OpenBCI, but it was a precursor to all of these. Since SCP is a relatively simple training paradigm, you likely could do a decent job with the Python / Brainflow approach. In fact some things may actually be easier that way, such as structuring specific successive 'trials'. It is not easy doing procedural / sequencing things with the VPLs.
Here is an example of a BrainBay design, but as you can likely see is not procedural / step-wise, so quite unlike an actual SCP trainer.
https://sites.google.com/site/biofeedbackpages/brainbay-openbci
Regards, William
Hmm, I went searching for existing SCP training hardware / software. My earlier statement about SCP being "relatively simple", is likely not in accordance with what I am seeing. Here is a manual for the Thought Tech system:
https://bio-medical.com/media/support/sa7986-slow_cortical-potential-scp-suite-reference-manual-rev-0.pdf
Which looks to be hugely complex with specialized add-on hardware components. Perhaps a simplified SCP app could be derived, but all these components were assembled to make the process easier for the clinician. That link was from this page, Support tab:
https://bio-medical.com/slow-cortical-potentials-package-system.html
'Info' sheet:
https://bio-medical.com/media/support/sa7986-slow_cortical-potential-scp-suite-info-sheet-rev-0.pdf
Thank you so mich William, this is tremendously helpful!
From what I've seen the eye movement filtering is critical indeed. They place one electrode near one eye, is that something that's possible with openbci?
And then I still don't understand what they actually measure
They state SCPs are in the range of 0-1hz. I clearly don't have an understanding what's actually measured in an EEG. So far I couldn't find out what they measure with SCP NF.
As I mentioned before, you could write a simpler version of SCP training, using Python and Brainflow library functions (these return the raw EEG data stream.)
SCP is measuring very slow moving 'EEG', below .1 Hz (10 seconds per cycle). More typically you will see SCP waves that take much longer than that, say 100's to 1000's of seconds per cycle. That is the natural rhythm.
Birbaumer's SCP training, has the subject intend to move the wave direction alternately in the up and down directions. This CAN be learned. This very slow rhythm is apparently related to cortical firing potentials, thus how likely the neurons are to fire. If brought under conscious control, there are numerous benefits.
Here is an entire issue devoted to infra low frequency EEG:
http://media.wix.com/ugd/cba323_fdd03c418d2348f59879f475b82439ee.pdf
Excellent Fall 2013 review issue of ISNR Neuroconnections, "What's happening below 0.5 Hz?"
https://openbci.com/forum/index.php?p=/discussion/201/large-millivolt-data-values-fbeeg-full-band-eeg
AFAIK the eye movement control/reduction is important, do you know if anyone has tried that with a cyton board? Is it at all possible, or would that need additional hardware?
https://www.google.com/search?as_q=EOG&as_sitesearch=openbci.com
I am aware of other ILF protocols that do NOT worry about eye movements. These are the Othmer's Cygnet protocols, and Mark Smith's protocols.
Thank you very much!