Electrode Cap or Ultracortex for BCI-VR research?
Hello,
I will begin a research in the area of BCI-VR and I wonder:
(1) which EEG would be better to use? Most likely it will be used the "Oculus quest" VR headset. The options are the EEG Electrode Cap (wet electrodes) and the Ultracortex Mark IV (dry electrodes).
(2) I wonder if there will be a difference in signal quality? I initially preferred the EEG Electrode Cap because I thought the wet electrodes had a better signal quality and because I think it will not be able to fit the VR headset over the Ultracortex Mark IV due to its large volume and hardness.
(3) Do you think this will not be a problem? An OpenBCI employee answer me with this.
We haven't validated the use of a wet/gel electrode cap with a VR headset, and have reservations about the use of wet electrodes with sensitive electronics of a VR headset. However, we have seen customers use the dry electrode EEG headset (The Ultracortex Mark IV) with an HTC Vive.
I would like your views on this issue.
Thanks a lot,
Marios.
Comments
Marios, hi.
Do you have a link to the customer case using Ultracortex with Vive? (Who was the person you talked with?) Is it the original Vive model? That one has both side straps, top strap, and a front section containing the lens and video output panels.
https://www.engadget.com/2016-04-05-htc-vive-review.html
I'm having a hard time visualizing how this could be used with Ultracortex, either with the UC on top of the Vive, or the Vive on top of the UC.
Similarly, if an elastic cap / electrode cap was worn 'under' the Vive, it would be pressing down on some of the electrodes that are co-located with the straps. Which would put strong pressure on those locations with the electrodes and probably be uncomfortable.
So essentially, both 'full cap' type of EEG headsets / caps, are going to be co-located in the same places with the Vive components. I'm not sure how that could work.
--
Here is another possibility. You could use under the Vive, something like this,
https://brain-trainer.com/product/tc26-trainers-cap-bundle/
This is a flexible electrode cap, where YOU get to choose where you mount the electrodes. Thus you could omit placing electrodes directly under Vive locations (straps and front unit). So there should be no pressure from the Vive down on sharp hard electrodes. But you can mount electrodes in locations that do not interfere with Vive.
Another advantage of this electrode cap, is that it uses 'saline' (salt solution soaked pads) type contact with the skin. This means no messy gel is involved, and no cleanup is needed on the head / hair / scalp after use. Traditional gel based elastic caps leave your hair looking pretty chaotic and slimy. That's ok for one off occasional usage, but awkward for regular daily use. Another advantage of saline is that it gives readings just as accurate as other wet systems such as gel or paste. But again, no mess factor.
There may be other similar caps out there, but this one is getting good reviews. I don't know about the comments the support person made regarding VR headset electronics noise getting into the electrodes. If that is happening, it seems like the effect would be strongest with electrodes right next to the front head unit. Avoiding those locations might solve the issue.
William
Dear William,
Thanks for your response!
TC26 Trainer’s Cap Bundle provides software to receive real-time raw eeg signals. Also, is there an sdk for creating neuro-games with the Unity platform?
So for a BCI-VR survey would you not recommend an OpenBCI product?
Marios
https://brainflow.readthedocs.io/en/stable/GameEngines.html#unity
https://openbci.com/community/neo-noumena-ar-emotion-openbci/ <-- Hololens + Electrode Cap
https://openbci.com/community/measuring-human-trust-using-openbci-and-vr/ <-- HTC Vive + Electrode Cap
OpenBCI hardware has been, and continues to be, one of the premier options for developing BCI and XR applications. This is such a great fit, that OpenBCI has decided to develop an integrated solution: Galea.
https://galea.co
Dear @retiutut,
Thanks for your response!
What do you suggest for a BCI-VR research? EEG Electrode Cap or Ultracortex Mark IV?
I initially preferred the EEG Electrode Cap because I think it will not be able to fit the VR headset over the Ultracortex Mark IV due to its large volume and hardness.
What EEG headset you think is better to combine with a VR headset? (1) EEG Electrode Cap or (2) Ultracortex Mark IV and why?
I'm just going to share these examples and let you choose. It should be clear that the electrode cap will fit with the HTC Vive from the second example.
Another example of OpenBCI hardware + XR: https://openbci.com/forum/index.php?p=/discussion/2860/mixed-reality-ssvep-research-platform-open-source#latest
@retiutut,
The question is whether it also fits with the Ultracortex Mark IV. I would prefer to use dry electrodes but I'm afraid it may not be combined with a VR headset due to its large volume and hardness.
Do you think this will not be a problem? So I can get Ultracortex Mark IV and wear under a VR headset?
Please review the examples I have shared carefully. You will then understand that Mark IV + any XR headset requires lots of work.
To this end, this is why OpenBCI is developing an integrated solution for XR called Galea.
@retiutut, @wjcroft
Unfortunately I can't wait until the release of Galea. So I have to go with the EEG Electrode Cap. I was hoping for an EEG without gel or paste. Is there this option?
From the data I have shared, electrode cap is probably the best option for you. Generally, this will provide better connections than the dry electrodes and the Mark IV, which is very important when conducting research with expensive hardware/software in a lab or at home.
Electrode cap benefits:
@wjcroft Is this information correct? Are there any more benefits to using the electrode cap for this use-case?
Here, you can see one of the examples I shared and how the engineer started with the Mark IV, then designed a solution that would fit better with the Hololens:
Start at 41 minute mark.
Richard and Marios, thanks for the comments. Marios mentioned:
The saline electrode cap I mentioned has some advantages, such as no gel required, and custom positioning of the electrodes. Most other caps I'm aware of, including OpenBCI's, have fixed electrodes. The electrodes underneath the straps / bands / face unit, may be problematic for EEG. Both in terms of the signal quality being changed by the band pressure, and skin comfort may also be less than optimal because of pressure on a small hard object next to the skin.
The TC26 cap I mentioned is only a cap. All the other software you would use is the same as for any OpenBCI board.
As Richard and I have pointed out, it appears the Ultracortex would occupy the same space as some of the VR headset components. Thus making it difficult to use both at once. There are other dry electrode alternatives that can be constructed using the headband kit and additional Velcro One Wrap material for bands going over top of head / chin strap etc.
https://shop.openbci.com/collections/frontpage/products/openbci-eeg-headband-kit?variant=8120393760782
The signal quality with wet electrodes is always stronger than passive dry electrodes.
Regards, William
@wjcroft Thanks for your response.
That is, I can use the TC26 Trainer’s Cap Bundle (https://brain-trainer.com/product/tc26-trainers-cap-bundle/) by connecting it to the Cyton + Daisy Biosensing Boards (16-Channels) (https://shop.openbci.com/collections/frontpage/products/cyton-daisy-biosensing-boards-16-channel?variant=38959256526) and receive raw eeg signals with the OpenBCI software?
How many channels does the TC26 Trainer’s Cap Bundle have?
The TC26 page mentions it comes with 7 electrodes, more can be purchased. It also states "Electrode covers (10)", which I assume fit around the electrodes as shown in the photos. So you may want to inquire if they can include some more; since these are just molded plastic they should be cheap.
It's certainly possible that the cap that OpenBCI sells would work fine.
Note that the electrodes that overlap with VR components, such as Fp1 Fp2 Fz Cz Pz F7 F8 T3 T4 T5 T6 O1 O2 (whew! 13 positions) may be problematic because of pressure from the VR side / top bands and face unit. From my quick look only 6 standard positions F3 F4 C3 C4 P3 P4 are not under bands or face unit. Standard EEG 10-20 locations are 19 positions. It's also possible that only the forehead points Fp1 Fp2 are the most problematic, and pressed upon by the VR face unit frame cushion.
My suggestion is that if you are in a university setting that may have electrode caps - to try out the fit of your VR unit with that cap to see how it feels. Maybe also gel it up to see how it performs.
Most caps connect via 'touchproof' plugs. OpenBCI sells a kit adapter to accept these. Are you sure you need 16 channels? Eight may be plenty.
William
Some EEG caps are not limited to standard 10-20 positions, instead there is the 10-10 system.
https://commons.wikimedia.org/wiki/File:International_10-20_system_for_EEG-MCN.svg
As you can see on this chart, all the points that correspond to the dotted line circle, and central vertical line, are co-located with VR headset elements.
I do believe that with the TC26 cap, you could drill more holes in these 10-10 positions. Thus avoiding VR band / face unit collisions.
@wjcroft
I don't understand how the electrodes will work? Is there any board on their website that I can buy and connect the electrodes? or I need to buy the Cyton Biosensing Board (8 channels)? And how can I get the raw EEG? With the OpenBCI software?
https://shop.openbci.com/collections/frontpage/products/touch-proof-electrode-cable-adapter?variant=31007211715
Most EEG electrodes use this connector system.
@wjcroft
You are not sure if TC26 electrodes can connect with OpenBCI Cyton Biosensing Board? Every EEG electrode that can connect with OpenBCI's board can also work with the OpenBCI's software to get the eeg raw data?
Where did you get that idea?? I gave you the link. The Touchproof adapter is how most EEG electrodes (including the OpenBCI cap and the TC26 cap), connect to OpenBCI boards. The adapter converts the touchproof plug to a board header pin plug. OpenBCI boards use header pins for connection. If you search for electrodes or caps online, touchproof is what they use.
In the Shop, a few OpenBCI items use the header pin plugs. Such as the gold cups, headband kit, ECG snap ons, etc.
Once any electrode is connected to a board, all the OpenBCI or other BCI software is available to use to process raw EEG.
@wjcroft
Thanks a lot William. I appreciate your help very much. Apologies for the pressure in some of my questions. Its because i want to be sure if I understand right. Your idea is very nice.
Many thanks for your help.
You may want to look around on the FRI site. They have dry cap and headband items, as well as separate electrodes and cables to match them.
https://www.fri-fl-shop.com/product/19-channel-eeg-cap-with-electrodes-and-leads/
The cap shown there would probably NOT work well with your VR headset. Because as previously mentioned, 13 positions in the standard 10-20 map, collide with VR headset components (straps, head unit). However the photo gives you an idea of how the cables work. Their other page on kits,
https://www.fri-fl-shop.com/product-category/packages/
Shows some of their headband variants. Note that the headbands are all made from Velcro One-Wrap material, with holes punched it it. If you use / create a headband system using top strap(s), you will want to incorporate a chin strap also to hold it down on the scalp.
https://www.amazon.com/VELCRO-Brand-ONE-WRAP-Double-Sided-Multi-Purpose/dp/B000078CUB
Not trying to confuse you with too many options. But I continue to have a hunch the TC26 cap, with your choice of locations (possibly drilling other holes) may have the most flexibility and non-overlap with VR headset elements. I've heard from some dry headband users, that considerable pressure is required / tightening of the bands, to get reasonable signal quality. With wet electrodes, this is not an issue.
@wjcroft
With TC26 salin electrodes pressure is required / tightening of the bands, to get reasonable signal quality. I mean salin electrodes are considered as normal wet electrodes?
Most EEG elastic caps do not use 'bands'. Instead the cap is made of elastic material that clings to the scalp. Both TC26 and the OpenBCI cap use chin strap (only) to help hold the cap down on the head.
The 'headband' systems using Velcro One-Wrap, are NOT elastic. So they require careful tensioning by hand (with tightening of the bands), to achieve the appropriate pressure required. Dry electrodes require significant downward force to the scalp. With Ultracortex, this downward force is generated by springs behind the electrodes. With headband systems there are no springs.
Wet electrode systems (gel, paste, or saline), do not require strong downward pressure. Just enough force to keep the electrolyte in touch with the scalp.
Ok thanks a lot William. You are so helpful.