Magnetometer-based Toothbrush Detection using the OpenEarable 2.0

Background

Earables are sensor-equipped earbuds that enable various forms of sensing near the ears (Röddiger et al., 2022). The recently introduced OpenEarable 2.0 (Röddiger et al., 2025) is notable for its rich array of sensors, which open up a wide range of applications in context-aware and health-related sensing.

This bachelor’s thesis focuses on the detection of toothbrushing regions using the magnetometer in the OpenEarable 2.0. When an electric toothbrush is in use, the electromagnetic field it generates can be detected by the magnetometers embedded in the device. By placing one earable in each ear, it becomes possible to estimate the position of the toothbrush relative to the mouth based on the changes in the magnetometer readings.

Tasks

1. Conduct a brief literature search of existing approaches that utilize magnetometers for detecting toothbrushing activity, as well as alternative approaches that rely on other sensing modalities.
2. Design an experimental setup in which multiple participants brush their teeth at various regions (e.g., different quadrants of the mouth), in order to collect labeled training data.
3. Process the collected data and train a classifier capable of distinguishing between different brushing regions.
4. Bonus: Move beyond basic detection and towards a more realistic application, such as an interface that provides feedback on how long each tooth region was brushed, aiming to encourage better brushing habits.

Relevant works to start with

• Magnetometer-based approaches
  • Lee et al., 2012 – Toothbrushing Region Detection Using Three-Axis Accelerometer and Magnetic Sensor
• Earable-based toothbrushing detection:
  • Yang et al., 2025 – SmarTeeth: Augmenting Manual Toothbrushing with In-ear Microphones
  • Wang et al., 2024 – ToothFairy: Real-time Tooth-by-tooth Brushing Monitor Using Earphone Reversed Signals
  • Yang et al., 2024 – BrushBuds: Toothbrushing Tracking Using Earphone IMUs