Slashdot Mirror

SoundcardEEG (scEEG) prototype BR> The idea

Everyone has a sound input to their computer these days. If we used it to record EEG signals, the EEG hardware itself would be simpler and cost less.

There is only one problem: Most audio systems, including sound cards, filter out frequencies below 20 Hz. The signals we want to capture are mostly below this frequency, so we can not just plug in the EEG amps and go.

There is a way to get past the filter, and it is called frequency modulation. In a frequency modulator, a varying input voltage (say EEG) is translated to an output sine wave of varying frequency.

We can for example build a modulator which takes input voltages between 1 and 2 volts and let it produce a wave as output that changes frequency between 2 kHz and 4 kHz, depending on the input. The output is still well within the audio range, but far above the 20Hz that is filtered out. It all works as long as the input signal varies at a frequency much less than 1 kHz.

When the modulated signal has been captured by the sound card, it is demodulated to its original form, in software.

The whole signal chain looks something like this:

Brain => EEG amplifier => hardware modulator => sound card => software demodulator

1) settle on a particular sound card, like the Creative Soundblaster 64 PCI, and engineering some trivial mods to allow it to pass 20Hz and below. Should be pretty simple.

2) If not, or in any case, just buy a cheap ADC. I bet you can find them for $50 that will talk either RS-232 serial or Parallel Printer Port or USB.

I am currently using a pretty good one that gives you EIGHT channels of 12bit Analog->digital, with USB interface and other I/O for only $110. (http://www.measurementcomputing.com/usb.html). I would bet that you can find even cheaper. You will want more than the TWO-channels that a sound card gives you if you want to be even semi-serious about EEG.

So if you can get a decent 8-channel ADC for $110, why bother building a complicated FM-mod/demod circuit that will probably drift all over the place just to get 2-channels while chewing up all your CPU cycles to do the acquisition ?

Does the $3,500 include the $1,300 cost of the Measurement Computing PCI-DAS4020/12 20

Well, it looks like a DAS4020 board has 12-bit resolution and so does the BBC Micro (although it might be only 10 bits in practice). The difference comes in sample rates: 20MHz versus 100Hz! So people are not going to be recording any UHF broadcasts through the analogue port. Unless they manage some serious overclocking.

Yuck...

I have not yet got a feel for the computational power required to approach real time processing or typical performance. Does anyone else know?

At $1,299.00 for the PCI card that their driver is written for, I do not see this in my future. For that matter, I don't see that in the future of many hobbiests which makes this project rather useless to the general population at present.

See here for information on the product the GNU Radio project wrote the driver for: Measurement Computing

Maybe some day...

A Linux-friendly HDTV recording solution is definitely needed. Unfortunately, it seems that in order to record HDTV you need a $1300 Analog Input Board.

Can anyone with more knowledge about this project please post a less expensive solution if one exists?

This page has the details. Too pricey for a tinkertoy.