A new Australian study has found that a cutting-edge 3D artificial intelligence system can detect early childhood tooth decay with accuracy comparable to traditional visual examinations. The findings suggest the technology could offer faster, more accessible, and more child-friendly dental care.
The research was conducted by the Murdoch Children’s Research Institute (MCRI) in partnership with the University of Melbourne. It is the first study to show how a wireless, hand-held intraoral scanner, paired with advanced AI software, can produce rapid and detailed assessments of children’s dental health. The device looks similar to an oversized electric toothbrush. It uses fluorescence light to scan the teeth and gums while a small camera records how the light reflects. The AI then assembles thousands of captured images into a precise 3D model of the child’s mouth. Clinicians can store, review, and share these images both in person and remotely.
To evaluate the tool, the research team studied 216 five-year-old children from the Melbourne Infant Study: BCG for Allergy and Infection Reduction (MIS BAIR). Each child received both a traditional dental examination and an assessment using the 3D AI scanning technology. The results, published in JMIR Public Health and Surveillance, showed that 38 percent of the children had signs of dental decay and 18 percent had enamel defects. These rates were the same in both visual exams and the digital scans, showing that the AI system performed just as accurately as trained clinicians.
Tooth decay is one of the world’s most common childhood diseases. More than a billion children globally are affected. In Australia alone, more than 40 percent of children aged 5 to 10 have decay in their baby teeth. Untreated dental problems can make it difficult for children to eat, sleep, or concentrate in school. They also remain a leading cause of avoidable hospital admissions.
Associate Professor Mihiri Silva, who led the study, said the findings help address a major knowledge gap in children’s oral health. She explained that baby teeth play an important role in predicting future dental outcomes. She noted that while visual examinations remain the gold standard, new tools are needed to spot decay earlier and more consistently. She added that digital images may allow clinicians to track plaque buildup and monitor subtle changes over time.
MCRI researcher Dr. Bree Jones said the scanner captured thousands of images that the AI stitched together like a jigsaw puzzle. She said the technology could be especially valuable for children who can only tolerate short dental visits. It may also help families who struggle to understand treatment plans or who live in remote or underserved communities.
Associate Professor Silva said more extensive studies are required before the technology can be fully integrated into everyday dental practice. Parents are also noticing the benefits. One father involved in the study said he welcomed anything that made dental visits easier for his five-year-old child and believed early detection helped build confidence.
The study comes as new data raises concerns about children’s diets. A recent survey of families in the Generation Victoria (GenV) cohort found that 80 percent of toddlers frequently consume highly processed foods. Nearly three-quarters of those foods fail to meet World Health Organization nutrition standards, and many carry misleading marketing claims.
Silva’s team is also leading the long-term Infant2Child study, which aims to strengthen prevention and support families during the first 2,000 days of a child’s life. She said parents consistently ask for practical and evidence-based advice. She believes early habits are essential to preventing decay and reducing long-term health costs.
Silva emphasised that improving children’s oral health requires strong prevention strategies and better coordination across health and social systems. Without early intervention, she warned, dental disease in childhood often continues into adulthood.