TLNM: Externally Validated Tooth Detection, Numbering and Segmentation from Smartphone Photographs Using Mask R-CNN
Using phone photos to automatically identify and count teeth for remote dental screening
A smartphone camera can automatically locate, identify, and map individual teeth in patient photos almost as reliably as professional dental imaging. The system, trained on over 1,200 annotated images, achieved 90% accuracy on external test data from different populations and phone models—suggesting it could work reliably across real-world conditions without costly equipment.
Billions of people lack access to affordable dental care, and this tool could enable basic tooth screening via smartphone for people in remote or low-resource areas. A freely available system that works with any phone camera could make early detection of oral disease faster and cheaper, potentially reducing the burden on overburdened dental clinics and letting people monitor their own teeth between professional visits.