Artificial intelligence for oral health in low- and middle-income countries: mobile health, teledentistry, and frugal innovation

Artificial intelligence is being applied to oral health in low- and middle-income countries through three overlapping channels: mobile health, teledentistry, and frugal innovation. This review contrasts the conditions, uses, and evidence of these applications against those of high-income settings. In high-income countries, artificial intelligence in dentistry is dominated by image analysis of radiographs and intraoral scans, supported by rich datasets and established regulatory pathways; in low- and middle-income countries, the applications are oriented toward access, using smartphones as the capture device, teledentistry as the delivery channel, and low-cost models as the means, and they target the caries and oral cancer burden that the workforce shortage leaves unattended. The evidence is strongest for teledentistry as a route to access, moderate for smartphone-based screening, and weakest for the claim that artificial intelligence itself, as opposed to the mobile channel around it, changes oral health outcomes. Most reported performance comes from high-income datasets, and transfer to low-resource populations is largely unproven. The review concludes that the value of artificial intelligence in low- and middle-income countries currently lies less in the algorithm than in the integration of the algorithm into affordable, connected workflows, and that realizing that value depends on local data, local validation, and the infrastructure that the literature, so far, mostly takes for granted.