DIY medical science is fun stuff. One can ferret out many of the electrical signals that make the body run with surprisingly accessible components and simple builds. While the medical community predictably dwells on the healthcare uses of such information, the hacker is free to do whatever he or she wants.
A good first start is to look at the relatively strong electrical signals coming off of the heart and other muscles. [Bernd Porr] has put together a simple bioamplifier circuit, and his students have made a series of videos explaining its use that’s well worth your time if you are interested in these things.
The electrically inclined among you are likely to want to start with the “from design to measurement” playlist, which details the construction of the amplifier itself. But the real goodies are hidden in the “EEG essentials: how to measure it and its artefacts” list; getting the signals is only the first step — interpreting them is where it gets interesting. For instance, a lot of what are sold as “mind control” devices these days is much more likely to be simply muscle-controlled, and this video shows you why: small signals buried under bigger ones. (Embedded below).
We’re no strangers to the tricks you can play with biosignals. The MobilECG project folks gave a great talk at Hackaday’s Belgrade 2015 conference, and made this sweet ECG business card as a demo. OpenHardwareExG is a more-sophisticated version of the bioamplifier presented here. And straying from the heart, we’ve seen a slew of “mind-controlled” applications.
But the point of the original post here is that making a bioamp need not be bank-breaking or brain-taxing. It’s the kind of thing that you can do simply on a weekend if you’ve already got the parts. What would you control with your body’s own electrical signals?
Thanks [nic] for the great tip!
OpenBCI has done a lot of great work in the open-source hacker-oriented biosensor realm: openbci.com
This may also be of interest to some, is that video capture is able to detect and monitor a person’s heart rate.
Blood passing through a person’s body changes the skin colour minutely and the camera is sensitive enough to detect these changes. The change in colour is directly related to blood flow rate. And the blood flow rate can then be used to determine how many beats per minute are inside the heart.
I first heard about this on the XBone using the Kinect, but there are open source implementations using a webcam
https://github.com/thearn/webcam-pulse-detector
https://github.com/thearn/webcam-pulse-detector
MIT did similar work a while back (there’s a HaD article somewhere) http://people.csail.mit.edu/mrub/vidmag/ .
They used it to detect infant breathing, pulse, subway car sway and other normally imperceptible motions.
You can also read blood flow in the brain via near infrared light absorption changes, these changes are proportional to processing activity levels and multiple concurrent sensor reading can be assembled into a 3D brain map via tomography.
But sadly like fMRI the BOLD signal, or the blood-oxygen-level dependent signal – the signal that both modalities identify – has a temporal resolution of 1-4 seconds; too slow to be useful as a control signal.
Could you use it at higher rates as input to a Kalman filter to help separate EEG from EMG signals? The thing is that it is very specific, down to 1 cm resolution, yes? So if your electrical signals are showing action in an area that does not correlate at all with the IR data then perhaps it is an error? I thought intent was detectable in advance of conscious awareness anyway?
I stumbled upon BPM Biosignals page about a year ago. Great educational value with all the videos, along with schematics and example results. Still what I remembered the most is how Mrs. Georgiadou is a professional and charming presenter :)
Just wondering… the green glow on their hair, is assume it is caused by a green screen behind them.
The reflection of the screen would cause the green glow on the edges of the subjects. They could also have their hair painted with a green glow, but that isn’t obvious considering the both have it and it isn’t continuous.
But why? If they use a green screen, why would they substitute the green screen by the image of a crummy looking grey “screen”? It really took my mind of the real content of the video, which is a pity because it is certainly a very interesting topic.
I think it’s a camera artefact, it’s trying to separate the hair colour from the sheet and you’re seeing…hmm i don’t know the correct term, notching? where the CCD has a coloured mask and that mask is off a bit. The same reason you don’t see herringbone cloth patterns on tv very much.
bitalino.org is also another kit you’ll want to look at if you’re into biosignal. A very cool project done by very cool people. It was featured in a funny way by an automotive manufacturer that used it to emulate a lie detector. It was kind of a public contest where, if you were able to keep telling the truth until the end of the drive, you were winning the car. The questions were made by a close relative and were more and more difficult to handle… I just can’t remember the link, but, oh man, how funny it was.
bitalino.com (soooorry)
The thing is lacking input protection. The INA is not going to last long. Just give it two 10k resistors and use a modern chip and single rail supply. It’s all AC-signals!
probably no input isolation coupling to prevent a failure in the power supply to giving someone a real need for medical attention. But you know, all that safety stuff in real healthcare equipment is just there to increase the profit margin and placate the lawyers….
The ADC board which is used (USB-DUX sigma) has electrical isolation (medical grade, 8kV immunity) and also the inputs of the INA are protected with diodes. See the biosignal howto page. The new bioamp (Attys) uses bluetooth and is thus isolated from mains and has surge protection via VDRs which act at voltages over 6V.