Edge-AI toothbrush that maps brushing coverage in 3D and provides real-time oral hygiene scoring.

An oral care company wanting to differentiate their electric toothbrush with real-time brushing quality analysis and personalized coaching.
Reconstructing 3D brushing motion from a 6-axis IMU inside the brush handle and classifying which tooth surface (buccal, lingual, occlusal) is being brushed, all on a battery-powered MCU.
The Nordic nRF5340 dual-core Cortex-M33 SoC handles both sensor processing and BLE communication. A Bosch BMI270 6-axis IMU captures motion at 400Hz. The handle houses a 40kHz sonic motor with a 3000mAh battery for 3 weeks of brushing.
We implemented a sensor fusion pipeline on the nRF5340's application core: IMU data → Madgwick filter for orientation → 3D trajectory reconstruction → surface classifier. The classifier is a small LSTM trained on 500 annotated brushing sessions, quantized to 200KB. Brushing coverage is mapped to a 16-zone mouth model and synced to the companion app via BLE after each session.
95% accuracy in classifying brushed surfaces. Users improved coverage by 40% within 2 weeks of use. Dentist-verified plaque reduction of 35% vs. manual brushing.
Let's discuss how ChipTalk can deliver results for your next project.