Artificial Intelligence-Assisted pH Mapping of Tooth-Surface Acidity Patterns

Dental caries is among the most prevalent infectious chronic diseases worldwide, yet localized acidogenic activity associated with cariogenic processes often remains difficult to visualize at the whole-dentition level. This highlights the urgent need for analytical tools that enable imaging of localized acidity-related changes. However, current methods, such as salivary test strips and hand-held pH meters, remain limited to single-point or averaged measurements, lacking the spatial resolution required to visualize the distribution of acidic microenvironments. Herein, we present a conceptual approach that moves from single-point sensing to spatially resolved imaging through an AI-enhanced optical imaging platform for visualizing tooth-surface acidity patterns. The method integrates an imprint-based colorimetric assay using a food-derived pH indicator, with an AI-assisted image analysis workflow guided by dentist-labeled annotations, and explores the feasibility of supervised, annotation-guided image analysis under pilot conditions. By mapping localized acidogenic activity as an indicator associated with cariogenic bacteria, the method provides interpretable spatial readouts of acidic microenvironments. The method allows visual detection of lactic acid at concentrations as low as 5 mM under the tested conditions, and provides multisite results within 5 min, showing qualitative spatial correspondence with clinical findings within this pilot cohort. This approach demonstrates the initial feasibility of treating the dentition as an analytically accessible surface, and explores an AI-assisted analytical concept for relating spatial pH patterns to clinician-identified caries-associated regions on biological surfaces. By combining accessible materials with intelligent data interpretation, this study presents a proof-of-concept pathway toward low-cost visualization of microenvironments for at-home and low-resource settings, offering particular value for children, the elderly, and individuals in remote areas.