extra electrode not visible with Muse2 and Bluemuse?

[danielemarinazzo]

ℹ Computer information

• Platform OS (e.g Windows, Mac, Linux etc): Windows
• Python Version: 3.7
• Brain Interface Used (e.g Muse, OpenBCI, Notion etc): Muse2 with Bluemuse

📝 Provide detailed reproduction steps (if any)

1. Created extra electrode with mini usb 5 pins (https://hackaday.io/project/162169-muse-eeg-headset-making-extra-electrode/details)
2. Connected the Muse2 with Bluemuse
3. Started muse-lsl

✔️ Expected result

See the AUX channel

❌ Actual result

Not see the aux channel.

The Bluemuse page https://github.com/kowalej/BlueMuse says

Muse 2 has AUX channel disabled - if I try to stream from this channel I get errors. It looks like no data comes from the channel when debugging Bluetooth inside a sniffing tool, so I'm making the assumption that Muse 2 doesn't actually support the AUX (secret electrode) input - it just has a (non functioning) GATT characteristic which is the same UUID as the Muse (2016).

Confirmed? Or is there a workaround? Is there an alternative to Bluemuse for windows? Anyone else encountered this issue?

[ErikBjare]

Probably better to open this issue in the muse-lsl repo. There's an experimental bleak backend in muse-lsl (not sure if merged yet) that I've found to work well on Linux (and should work on Windows too). Can't remember if we got the AUX channel, but might be worth trying: alexandrebarachant/muse-lsl#148

…

On Tue, 11 May 2021, 12:01 Daniele Marinazzo, ***@***.***> wrote: ℹ Computer information - Platform OS (e.g Windows, Mac, Linux etc): Windows - Python Version: 3.7 - Brain Interface Used (e.g Muse, OpenBCI, Notion etc): Muse2 with Bluemuse 📝 Provide detailed reproduction steps (if any) 1. Created extra electrode with mini usb 5 pins ( https://hackaday.io/project/162169-muse-eeg-headset-making-extra-electrode/details ) 2. Connected the Muse2 with Bluemuse 3. Started muse-lsl ✔️ Expected result See the AUX channel ❌ Actual result Not see the aux channel. The Bluemuse page https://github.com/kowalej/BlueMuse says *Muse 2 has AUX channel disabled - if I try to stream from this channel I get errors. It looks like no data comes from the channel when debugging Bluetooth inside a sniffing tool, so I'm making the assumption that Muse 2 doesn't actually support the AUX (secret electrode) input - it just has a (non functioning) GATT characteristic which is the same UUID as the Muse (2016).* Confirmed? Or is there a workaround? Is there an alternative to Bluemuse for windows? Anyone else encountered this issue? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#98>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKXDORTSCDOJYM5C6XGZUTTND6AJANCNFSM44U7B37Q> .

[danielemarinazzo]

Thanks! But apparently looks like the muse lsl is not able to see the aux on the Muse 2, while musejs based systems, such as Mind Monitor and EEGedu, do. See kowalej/BlueMuse#43 (comment).
Any chance to integrate this with the eeg notebooks?

[JohnGriffiths]

Sorry our competence with eeg-notebooks is only really dealing with the streamed inputs that we get.

If you're asking whether we can use an EEGedu stream as an input to muse-lsl - not really sure, I don't understand the tool well enough.

One new option you could look into (we haven't yet) is the petal.tech streaming app, which is similar to bluemuse. The developers are quite responsive so you could perhaps ask them to look into this for you if it isn't giving you the aux stream out of the box.

[JohnGriffiths]

Ok I just tried with Petal. The LSL stream is picked up, but it does not (at least for muse S, which I have on me currently) have the extra channel by default.

[ I'm not sure whether or not that is expected on muse S actually, but interaxon peeps tell me it can be used with an additional chan. I'm not sure if this differs on muse2 and muse2016. I can check muse2016 later, but don't have a muse2. ]

Anyways, for reference, once you've initiated an LSL stream in the petal app, the following code works to pull the stream into Python. It's exactly the same as you would with bluemuse.

from pylsl import StreamInlet, resolve_byprop
from muselsl.constants  import LSL_EEG_CHUNK, LSL_SCAN_TIMEOUT

streams = resolve_byprop('type', 'EEG', timeout=LSL_SCAN_TIMEOUT)
inlet = StreamInlet(streams[0], max_chunklen=LSL_EEG_CHUNK)
samples, timestamps = inlet.pull_chunk(timeout=0.0, max_samples=100)

import numpy as np
arr = np.array(samples)
arr.shape

[danielemarinazzo]

Thanks a lot! I looked into the code for petal, and they have coded by default 4 channels only. One solution could be to edit the code to add the aux channel.
On the other hand looks like @urish did this here https://github.com/urish/node-lsl, maybe @jdpigeon has directly addressed this issue (not being able to see the aux channel with lsl, but yes with js, and then convert the js to lsl for use in neurotechx stuff)?