[EEG] low budget EEG project
hurtstotouchfire at gmail.com
Tue Jul 13 11:20:06 PDT 2010
For reference in case any readers are not familiar with sleep EEG:
It seems to me that it depends what you want to do with your sleep
data. You can probably characterize the transition from wake to sleep
no problem with power spectrum data alone. Stage 2 sleep and waking
alpha are sometimes similar though as far as just the power spectrum
goes, without being able to look for k-complexes and spindles. You
might be able to differentiate k complexes based on voltage though.
Other than eye movement during REM, you shouldn't have too much high
voltage activity once you're actually sleeping.
Given the charts in the first link, it should be possible. I'm
skeptical of their gamma, but the <35Hz data is totally reasonable. I
think that the big win is that it's a fully packaged device already,
with bluetooth/wireless built in. A lot of the hassle with eeg comes
in just tweaking little stuff to get good contact and shield from
noise. Do you know if the Neurosky electrodes are active? I have
trouble imagining they could get that data from passive electrodes.
And if they're active, that's a pretty big win, because active
electrodes are a little bit of a hassle to build. It would probably be
worth trying to hack directly into the connection and get the raw data
if they're active, because then we could harvest readymade active
electrodes... That sounds nice.
You should talk to Andreas, by the way. He's building his own caps from scratch.
I was just thinking earlier today that it's a bummer that doing EEG
for a living has killed my drive to hack it in my spare time. I'm sure
it will reassert itself when I'm not working so much. In theory, EEG
meets on Sundays, but no one goes anymore. I'm not sure how
Sensebridge will fare without Eric. Feel free to rally people to work
on this stuff, Dan. And best of luck.
On Tue, Jul 13, 2010 at 10:20, Dan Brown <danbrown at gmail.com> wrote:
> Similar to what Tracy sent out recently, there are some blog posts
> about using the EEG modules from Neurosky used in toys for collecting
> EEG data:
> Based on these examples, I am planning to build a mindflex+bluetooth
> radio, which I expect to provide the 8 frequency-band strength numbers
> to a bluetooth host, eg a laptop or mobile phone. Fwiw, I am most
> interested in monitoring brain activity during sleep. If anyone is
> interested in this project, let me know and we can set up some times
> to work together at noisebridge.
> It would be even better to get the raw EEG waveform data from the
> device, but I'm not sure if these cheap Neurosky modules are capable
> of sending that data. Anyone have any suggestions about that?
> EEG mailing list
> EEG at lists.noisebridge.net
More information about the EEG