So I first came across when the Embrace watch was launched on Indiegogo. Essentially by monitoring the resistance across the skin of a person, it is able to reliably detect seizures. Yes, the majority of seizure related deaths have causes that are unfortunately unknown and “just happen” (SUDEP), but even the slightest probability of someone alone having a seizure at the wrong place at the wrong time (such as a flight of stairs) can be extremely worrying for families. Various attempts and products have existed on the market utilising accelerometers/gyros to determine the event of seizures and message loved ones, but they often give off false detections and they require the epileptic to have clonic seizures (i.e. jerking). From what I read the Embrace does work. Its origin came from research into EDA (electrodermal activity) with a sensor strapped to a subject who just happened to have epilepsy. One day, one of the researchers looked through the data and discovered an unusual phenomenon in the graph and asked the subjects family what had happened that day at the exact day: a seizure!
So what is EDA?
EDA basically monitors the resistance across your skin. Emotions are often attributed to changes in the resistance of skin, as well as sweat, and are often found in lie detectors.
Upon discovering this wonderful application of EDA, naturally I wanted to get a sensor to read it. Yep, I wanted to make my own, but that would require time (I was a few weeks before exam block when I was tangled in a bunch of assignments). Obviously I wanted to make a nice pretty little board with two electrodes across the bottom, but each PCB revision would take ages to receive here in Australia. Sure I could probably just prototype it on protoboard, but still I’d need components which I simply don’t have. Then I found the inexpensive Grove GSR … It came in the mail around a week later and I’ve finally got to use it today.
It “kind of” worked, but:
- I couldn’t see changes due to breathing
- There seemed to be some ridiculous rise time for the sensor and an even more ridiculous fall time, probably attributed to some cap on the board. As a result, I had to turn the sensor on/off every time I polled it (I haven’t tuned it but currently I’ve got a 20 ms off time and 5ms on time; 40Hz sample)
Sensor output of Grove GSR. The graph above is generated using pyqtgraph.
In the fact the only thing I could detect was me clenching my wrist. When I clenched my wrist there would be a rise in the EDA reading (i.e. lowering of resistance) which I would assume to be due to the stretching of skin.
Put it this way, the people behind the Embrace watch have been working on EDA experiments since the late 90’s at MIT and there sensors have been strapped onto hundreds of wrists. On the other hand the Grove GSR was probably designed by some guy at Seeedstudio as a toy for hobbyists. But then again, monitoring is a relatively simple task. Perhaps I’m doing something wrong?