LSL Data Connection Between OpenBCI & NeuroPype

in OpenBCI_GUI
Hi all,
I am seeking help connecting the OpenBCI GUI to NeuroPype via LSL on the same PC.
Currently, we are using a Cyton + Daisy setup for 16 channels on the OpenBCI side, and I am looking to replicate something like this
https://openbci.com/forum/index.php?p=/discussion/comment/19318#Comment_19318
to verify the movement of data between programs and NeuroPype's usefulness.
Any help would be greatly appreciated.
We are currently running OpenBCI Gui V5.2.2 and NeuroPype Academic Suite V2023.0.0
Thank you so much
Comments
In our current setup, I am outputting a timeseriesraw stream from OpenBCI via LSL and NeuroPype will verify that it has received the data but will stop responding on the PC as the LSL stream continues
Hi Sam,
I understand from this page: "NeuroPype Broad Hardware Support List"
https://www.neuropype.io/#features
That it can directly read from Cyton. Is that true? Does it work when directly streaming to NeuroPype from the Cyton serial port?
William
Hi William,
Currently, we are using the OpenBCI GUI to generate the LSL stream, while it seems that the link you mentioned directs me to a page to set up a Python script, which I believe has been absorbed into the GUI.
Nevertheless, I plan to test direct streaming to NeuroPype the next time I am in the lab.
Hmm, when I go to the link I gave:
https://www.neuropype.io/#features
I can scroll down near the bottom to the "NeuroPype Broad Hardware Support List" section, which lists OpenBCI support. But the link there for OpenBCI goes to:
https://docs.openbci.com/Software/CompatibleThirdPartySoftware/LSL/
Which implies they may NOT have direct support, but rely on LSL. The repo mentioned,
https://github.com/openbci-archive/OpenBCI_LSL
Has been 'deprecated', meaning it is no longer supported.
That's good. Let us know if it accepts direct connection to the Cyton serial port.
There is an example in the 'docs' tree regarding feeding GUI LSL data to NeuroPype,
https://docs.openbci.com/Examples/EEGProjects/MotorImagery/#acquition-of-eeg-data-and-markers
However a number of users have had issues with that tutorial, and the author refuses to answer any questions.
Hi William,
I was able to get back into the Lab today,
I tried to setup the aforementioned pipeline and unfortunately, I could not get any type of output.
Here is a portion of the NeuroPype Log showing some errors that I recieved.
Thank you so much for your help
I'm sorry, but that Motor Imagery tutorial is not being supported by the author. I only gave that link as an EXAMPLE of how he linked NeuroPype to the output of OpenBCI_GUI LSL stream. You're going to have to try workarounds on your own. Please abandon any attempts to get the tutorial to run, that was not the point.
This is probably useless: but the NeuroPype people COULD in the future use the Brainflow library to read OpenBCI streams directly, without involving the GUI or LSL. But possibly they depend on LSL so much that this is infeasible.
https://brainflow.readthedocs.io/en/stable/Examples.html#python
No worries,
We are just trying to prove some type of interconnection currently so we don't have any expectations of an output making sense.