For two and a half years, my bruxism symptoms were manageable. After I stopped wearing the early grind control device in August/September 2013, my symptoms continued to steadily improve. Not completely gone, but most days I felt completely normal, without headaches or dizziness, and no pain or muscle tension while chewing. The symptoms would occasionally come back sometimes occasionally, for example I would feel tension in the morning for a few hours.

I knew my bruxism was not gone, because of these occasional flare-ups, and because I was told I sometimes grind my teeth at night, sometimes right after falling asleep. I also strongly suspected some medication would significantly worsen my night-time bruxism, particularly fluticasone/salmeterol (brand name: Seretide). I only use this medication a few times a year, for a few days, when my airways are irritated, but every single time, it makes my bruxism kick into full gear. Looking back, it’s not a coincidence that the extreme symptoms appeared end of 2010/early 2011, at a time when I was using the asthma inhaler for a good six months stretch.

Most recently, during a weekend trip to Taiwan, I felt the kind of dizziness in the morning that I can only attribute to bruxism. I am almost certain it must have been due to increased teeth grinding the nights before, but I had no way of proving that. An idea started forming in my head to build a new device, comfortable and convenient enough to wear every night, which could alert me to this sort of situation. Taiwan is a huge semiconductor manufacturer and some billboards advertise this in the airport. So my mind was on electronics.

Besides keeping in touch with the fine folks on the Bruxhackers Mailing List, I hadn’t touched any electronics, hadn’t powered up any microcontrollers or held a soldering iron in the last two years. I loaded up Sparkfun.com to see what’s new and came across this little gem:

The Particle Photon is an Arduino-compatible, wifi-enabled dev board, made to work seamlessly with the Particle Cloud. In short, you can call a REST API to invoke a function on the Photon, or query a particular variable value, without worrying about network connectivity behind a router, or setting up an internet-facing server. Firmware flashing happens over the internet as well. This seemed like a perfect platform to build a bruxism-tracking device on. With the previous grind control device, I always felt like I was running blind. I couldn’t see in realtime what the EMG value was, because the only way of outputting data was connecting to the USB/serial port, and that wasn’t compatible with the EMG sensor, which had to be electrically isolated.

With the grind control device, at every development cycle, I had to take the microSD card out, copy the data over, process it, and start over. This was extremely time consuming. The new device definitely had to be Wireless. Additionally, I had to charge 2x 9V batteries every single day, and replace the old ones.

Objectives for the new device:

  • Wireless: for every day usage, it’s too much of a hassle to have to disassemble and take the microSD card out.
  • On-board battery charging: it should charge by plugging in a USB cable, just like every modern gadget.
  • Controllable from a smartphone: because the device is small, and worn against the user’s head, it doesn’t have an LCD display. But it should relay information back to a mobile device, smartphone or tablet.
  • Comfortable: it should be comfortable enough to wear every night. This means the size should be small, and there shouldn’t be a lot of wires sticking out.

An essential component in an anti-bruxism device is the EMG (Electromyography) sensor. The grind control device used the http://www.advancertechnologies.com/ Muscle Sensor V3. It just so happened that they released a new version, the Myoware Sensor right around the time I set out to build a new device.

The Myoware has a number of advantages:

  • Doesn’t require a dual power supply: can be powered from a +3.3V input
  • Has electrode snaps directly on the PCB, which makes wires short and as a result is less susceptible to interference.
  • Very small in size: can be stuck directly on the face.

It seems like a perfect fit for this project. Sparkfun has released a number of shields for the Photon which I can also directly use. Here are the building blocks for the new device:

As a first step, a PCB must be designed that regroups all of these elements onto one board. The Myoware sensor will be external, connected using a short cable.

Check back soon for updates !

Do you have questions about this project ?
Get in touch: contact@sleeptrack.io   @SleepTrackIO   sleeptrackio