Search Dental Tribune

New AI platform radically improves detection of early childhood caries

Researchers at the University of Hong Kong have developed a machine learning model capable of accurately detecting early childhood caries before it becomes visible. (Image: galitskaya/Adobe Stock)

HONG KONG: According to the Global Burden of Disease study, early childhood caries affects approximately 530 million children around the world, making it one of the most prevalent chronic childhood diseases globally. In an attempt to mitigate this public health concern, a collaborative research team from various institutions in China has developed a revolutionary artificial intelligence (AI) model capable of detecting subtle microbial shifts that indicate the development of caries in specific teeth long before it becomes clinically evident. The implications for patient health and treatment are thus substantial.

Over an 11-month period, the team tracked changes in plaque microbiota in 89 preschool children and in this manner established a microbial map of the primary dentition—constituting a unique research contribution. As reported in the study, normally an individual exhibits a stable pattern of microbe distribution between the anterior and posterior maxillary teeth, each region harbouring its own unique communities. Of vital importance is that this structured distribution of microbes throughout the mouth is disrupted once caries begins to develop. Microbes associated with the incisal region migrate to the molar region and vice versa.

The researchers developed a highly utile AI platform on these spatial microbial indicators of caries and using it were able to detect these specific bacterial shifts occurring on individual teeth well in advance of any caries becoming visible. The end result is essentially an early microbial risk indicator for each tooth in the mouth.

The model is highly accurate. By integrating the data on a tooth’s unique microbial profile with that of adjacent teeth, the platform achieved 98% accuracy in detecting existing caries and 93% accuracy in predicting caries two months before it became clinically apparent.

Speaking in a university press release on this pioneering development, co-author Dr Shi Huang, assistant professor in the Division of Applied Oral Sciences and Community Dental Care at the Faculty of Dentistry of the University of Hong Kong, commented: “These findings fundamentally change how we understand tooth decay.” Dr Huang explained: “We’ve moved from seeing cavities as inevitable to being able to predict and prevent them at the microbial level, tooth by tooth.”

The clinical significance of the research is undoubtedly enormous, and the model if widely adopted could revolutionise how early childhood caries is treated. The AI platform amounts to nothing less than a powerful advance warning system for the development of caries in children, using these minute microbial shifts as precise indicators of future caries, infection, pain and overall health. If employed, the platform will thus be capable of greatly improving the oral health and well-being of young children, making it especially valuable.

The study, titled “Single-tooth resolved, whole-mouth prediction of early childhood caries via spatiotemporal variations of plaque microbiota”, was published online on 11 June 2025 in Cell Host and Microbe.

Topics:
Tags:
To post a reply please login or register
advertisement