Toothbrushing Monitoring using Wrist Watch
In this project, we built a toothbrushing monitoring system that monitors the 16 toothbrushing surfaces and brushing techniques, using a customized toothbrush and smart watch. Our system successfully recognized toothbrushing gestures with an average precision of 85.6%. The details can be found in our Sensys' 16 paper.
A demo video can be found here.
- A manual toothbrush is customized by installing a magnet on it. This is significant because when the user changes the holding of the toothbrush, we can capture such changes using the magnetic sensor on the smart watch.
- The smart watch recognizes the holding of the toothbrush using the magnetic sensor, and recognizes the toothbrushing motions using the accelerometer and the gyroscope. Together, the toothbrushing surface is recognized.
- The recognition results are transmitted to the tablet for real time feedback to the user. So that the user knows how long each surface has been brushed.
- We use the cloud services to enable data collection, user management, and machine learning model computation. Without the cloud services, it will be much more difficult to collect data from different users living at different locations over a long period of time.