The electrical signals emitted by the human body tell us a lot about what’s going on inside. But getting those signals inside your microcontroller is not straightforward: the voltages are too small for most ADCs, and the ever-present 50 or 60 Hz mains frequency makes it hard to discern subtle changes. Over at Upside Down Labs, [Deepak Kathri] developed a universal biosensor interface called the BioAmp EXG Pill to make all this a lot easier.
Its name refers to the fact that it can be used for several different bio-electrical sensing applications: ECG, EMG, EOG and EEG, which deal with signals coming from the heart, muscles, eyes and brain, respectively. To enable such flexibility, the board has connectors for two or three electrodes, as well as solder pads to mount resistors and capacitors to adjust the gain and bandwidth. An instrumentation amplifier increases the strength of the desired signal while rejecting noise and interference.
The form factor allows easy connection to electrodes on one side and a data acquisition system on the other. Measuring just 25.4 mm long and 10 mm wide, it should be easy to integrate into any type of biosensing gizmo you can come up with. [Deepak] has made several demo setups, showing him using the Pill with an Arduino to measure his heart rate, detect eye blinks, and even control a robot arm using his own arm muscles!
The EXG Pill is an evolution of an earlier EMG-only project. We’ve seen several great ECG and EEG projects before, but is the first time we’ve seen one amplifier that can do them all.
[Eric] tipped us about the OpenHarwareExG project which goal is to build a device that allows the creation of electrophysiological signal processing applications. By the latter they mean electrocardiography (ECG, activity of the heart), electroencephalography (EEG, signals on the scalp), electromyography (EMG, skeletal muscles activity), electronystagmography and electrooculography (ENG & EOG, eye movements) monitoring projects. As you can guess these signals are particularly hard to measure due to their small amplitude and therefore susceptibility to electrical noise.
The ADS1299 8-channel 24-bit analog front end used in this platform is actually electrically isolated from the rest of the circuit so the USB connection wouldn’t perturb measurements. An Arduino-compatible ATSAM3X microcontroller is used and all the board is “DIY compatible” as all parts can be sourced in small quantities and soldered by hand. Even the case is open source, being laser cut from acrylic.
Head to the project’s website to download all the source files and see a quick video of the system in action.
Interested in measuring the body’s potential? Check out an ECG that’s nice enough to let you know you have died, or this Android based wireless setup.
[Michael] from Lucidcode is at it again, this time with an Android app called Halovision.
In case you don’t remember, this is the guy who has been working on the Lucid Scribe Project, with the end goal of communicating from inside your dreams! Here’s the basic gist of it. If we can use a sensor to detect REM (rapid eye movement) or body movement during sleep, we can tell if we’re dreaming — then it’s just a matter of using an audible cue to inform the sleeper of the dream, so they can take control and become lucid.
The first way they did this was by using commercial EEG headsets to detect REM. We covered a hack on modifying one so it would be more comfortable to wear at night, but what is really exciting is [Michael’s] new app, Halovision — No EEG required
It’s an Android app that uses the camera to detect movement during sleep, and it is only the first plugin planned for Lucid Scribe. The algorithm is still in its experimental stages, but it is at least somewhat functional at this time. They note it’ll only work for day-time naps or with a bright night light, but this could be easily solved with an IR webcam and a few IR LEDs.
It will be interesting to see where this all goes, has anyone else been following or participating in Lucid Scribe?