Here I’d like to record the process I’m using to record PSGs and collect my own data for dream realization projects. It is worth repeating that I am an average Joe with zero technical training in polysomnography, bio-scanning, electronics, computer programming, any of this sort of stuff. I’ve never played with an “Arduino”. I’m a humanities guy with an interest in science, that’s it. So there are MANY more people out there that know far better than I how to proceed, and if you can find and understand their blogs you’ll likely be in better hands. However, I’m interested in seeing how much a non-specialist like me can figure out, and recording my process of “discovery”. So in this series of posts I’ll present what I have gleaned from following the instructions on the OBCI site, doing a little research, and getting advice from various forums. I welcome the constructive feedback of my betters, and hope this might be helpful to anybody who is even more of a n00b than I am.
Onto the GEAR. There are increasing numbers of options out there for exploring basic electroencephalography. Neurosky make one that is pretty cheap, and Emotiv have recently developed all kinds of (more expensive) gear, and I expect there are lots of other things out there. I’ve not tried these, so I cannot review them at all. They DO seem to be consumer friendly, slick, and claim to be quite easy to use, depending on what you want to do. These features are likely quite appealing to many, but they don’t have much charm for me! And, at any rate, they concentrate exclusively on EEG readings. This is great if you want to learn about biofeedback or turn brainwaves into art, but if you want to get into DREAMS, you need more than EEG. I don’t recall how I found them, but I picked the “Cyton” board by OpenBCI because it seems like a real bit of gear that legitimate scientists and savvy biohackers can use, rather than a cool yuppie dew-daw. It doesn’t claim to be a polished final product (it comes with quite a lengthy disclaimer card, which I like), but it has an opensource ethos I respect and a growing community of users posting things on the forum. Though more expensive than, say, the Neurosky, my strong sense is that this thing provides excellent bang for the buck.
The OBCI “Cyton” board (w/o Wifi Shield). This provides 8 channels of biodata, which is better than the “ganglion”, and is more than enough for my current interests. It can, however, be expanded to 16 channels by adding a “Daisy”, and a WIFI shield which allows the transmission of data over WIFI rather than just bluetooth. I also built a little case for it. These guys made a great looking case and were selling it, but their site says they’re currently working on a second version (and it has been saying that for a while). If you want to make your own case, just bear in mind that if you plan on using certain tools you may need to plug stuff into the TOP of the board as well, not just the main pins at the front.
With this I also got some “wet” electrodes which require the use of the ominous sounding “Conductive Neurodiagnostic Electrode Paste”. This stuff is a bit of a pain, but not too bad once you get the hang of it.
More recently I obtained the “Headband Kit”. This is NOT necessary for PSG, nor anything else, but it
does make life quite a bit easier. For my purposes, it means I can get three head electrodes, and my two earlobe connections set up in seconds, which is great for dealing with nervous nellies and people who don’t want too much Electrode Paste in their hair. I’ve also been experimenting with adding a strap over the top to secure central electrodes, but haven’t been totally happy with the results so far. And I still use the wet electrodes for eye and muscle readings.
I also bought the Myoware Muscle Sensor and the Pulse Sensor. These I regret buying. The muscle sensor is just not necessary for PSGs, and requires more electrical expertise than I was expecting. For example, I had to buy extra cable, and solder it to the board. My impression is that everybody at OpenBCI has loads of experience with this sort of thing and sort of assumes anybody using their stuff to have lots too, which can be a bit harder on those who don’t. That said, the sensor does have a really thorough tutorial, and if you’re doing lots of EMG stuff I’m sure it is great! I was quite happy to learn soldering (and I think I did a pretty nifty job), but so far I’ve found the sensor unnecessary for PSGs. I’ll have to revisit this at some point.
The pulse sensor is more disappointing. There are zero instructions (at least at the time of writing), with only a sales video (using a pretty silly “Christopher Walken” voice over) to guide you – and it shows connection to an Arduino, not an OpenBCI board. Again, if you’re a heavy Arduino geek this thing is a piece of cake, but if not – good luck. I eventually figured it out, I think from somewhere deep in a forum – it plugs into the TOP of the Cyton (Power from VVD and GND, with readings on D11), and you can look at your heart beat on the special analog “pulse” widget in the OpenBCI GUI. Kind of cool, and maybe I’m doing it wrong, but the signal is pretty crappy compared with what you get if you use regular electrodes connected to the wrists. At some point I might revisit it, but for the time being I feel a bit ripped off. But call it a tax on ignorance – the thing does does mostly work like it said it would, eventually.
The final bit of gear is also very important: TAPE! Particularly for the wretched chin electrodes for the EMG (see upcoming post), securing the electrodes for an entire night of sleep is not super easy. I tried two kinds of (useless) medical tape, then found one pretty good brand (“UrgoFix”!). But the best so far is that absurd sports tape various athletes have been taping to themselves lately. That stuff hangs on really tight, and is very flexible. The only trick is getting it off – particularly around the eyes, this can be an issue, so I’ve been using multiple tapes!
That covers my gear. The next step is deciding what to measure, which determines where you want to place your electrodes.