Temporomandibular disorders (TMD) comprise a group of conditions affecting the temporomandibular joint and the masticatory muscles. A common symptom of TMD is unilateral chewing, where affected individuals predominantly chew on one side of the mouth. Early and reliable detection of such asymmetric chewing patterns could support diagnosis and monitoring of TMD.

In 2021, a detection method for identifying unilateral chewing using the OpenEarable was patented at TECO. This approach relied primarily on pressure sensors integrated into the earable to detect chewing patterns. Subsequent investigations, however, revealed that pressure sensor data alone is not sufficient to reliably detect unilateral chewing caused by TMD.

Beyond pressure sensors, the OpenEarable is equipped with additional sensing modalities, including an in-ear microphone and a six-degree-of-freedom inertial measurement unit (IMU). These sensors may provide complementary information for chewing pattern detection. Prior research, for example at the University of Cambridge, has demonstrated that IMU data from earables can be used to recognize fine-grained movement patterns, such as the exact location where a user brushes their teeth. Building on these findings, this thesis explores whether alternative sensors or combinations of sensors (pressure sensor, IMU, and in-ear microphone) can be used to develop a reliable algorithm for detecting unilateral chewing as an indicator of TMD. Depending on the project’s progress, OpenEarable 2.0 with an additional bone-conduction microphone may optionally be incorporated.

• Independently collecting chewing data using the OpenEarable, recording mastication movements with a variety of different food types to create a comprehensive dataset

• Conducting data collection with multiple participants to ensure sufficient inter-individual variability

• Designing, implementing, and training an algorithm to detect unilateral chewing patterns based on the collected sensor data

• Evaluating sensor combinations (pressure sensor, IMU, in-ear microphone, and optionally bone-conduction microphone) with respect to detection accuracy and robustness

• Developing a final model that is robust to individual differences and compact and efficient enough to run either directly on the OpenEarable or on a connected smartphone

• A paragraph explaining your motivation.
• Your study program (Bachelor/Master), current semester, and field of study.
• A transcript of records (courses and grades).
• Your programming experience.
• Any areas of interest relevant to the topic.
• Your CV (if available)