A Wearable Earplug-Shaped Piezoresistive Sensor Based on Ear Canal Deformation for the Screening of Temporomandibular Joint Motion Abnormalities

Temporomandibular disorder (TMD), as a multifactorial oral disease, has long relied on diagnostic methods of clinical palpation and imaging. However, these approaches are highly subjective, costly, and invasive. Inspired by the principle of clinical palpation, this study proposed a noninvasive detection method based on reduced graphene oxide (rGO) composite aerogel piezoresistive sensing technology. A smart earbud sensor was designed and assembled to monitor the deformation of the anterior wall of the external auditory canal. The sensor exhibits an extremely low detection threshold of 14.5 Pa and a short response time of 64 ms, remains stable over 1000 compression cycles, and successfully captures subtle facial activities such as swallowing and mouth opening. In subsequent evaluations of temporomandibular joint (TMJ) motion, the sensor captured different signal patterns between volunteers with normal and abnormal mandibular motion patterns, and the sensing results showed consistency with clinical evaluation and cone-beam computed tomography-based osseous structural observations. To reduce the limitations associated with subjective judgment and further analyze the collected signals, a support vector machine model was introduced for preliminary classification analysis, achieving an accuracy of 91.7%. This work provides a new wearable perception strategy for objective and rapid screening of TMJ movement abnormalities, offering a potential complementary approach for future TMD-assisted assessment.