Search Dental Tribune

Pearl AI reveals hidden oral health crisis

Artificial intelligence analysis of millions of dental radiographs has revealed substantially higher levels of untreated caries than reported by conventional public health surveys. (Image: LuxeShutter25/peopleimages.com/Adobe Stock)

Mon. 10 August 2026

save

LOS ANGELES, US: Dental AI company Pearl has released the Pearl Oral Health Index, which it describes as the first large-scale radiographic census of oral health in the US. Based on artificial intelligence (AI) analysis of millions of dental radiographs, the findings suggest that the burden of untreated caries may be substantially greater than previous national estimates. The analysis offers fresh insights into population oral health and highlights inequalities in access to dental care.

The index analysed radiographic data collected between April 2024 and March 2026 from more than 26 million dental visits involving approximately 15 million adults, covering around 737 million tooth observations. Using its AI technology—which has US Food and Drug Administration clearance—Pearl examined radiographic data at a scale not previously possible, creating what the company describes as the most comprehensive assessment of American oral health to date.

One of the report’s most striking findings is that untreated caries appears to be at least 4.5 times more common than the figure reported by the US Centers for Disease Control and Prevention’s National Health and Nutrition Examination Survey (NHANES), which relies primarily on visual and tactile examinations of a nationally representative sample. Also, the average number of decayed teeth per patient identified by Pearl’s AI was 8.7 times that reported in NHANES. By comparison, the average numbers of missing and restored teeth closely matched government estimates, a similarity that Pearl argues lends credibility to its AI-based methodology.

The company argues that radiographs can reveal carious lesions that may not be detected using a visual and tactile examination, potentially explaining the discrepancy. This is consistent with recent peer-reviewed research highlighting the value of radiographic examination for detecting hidden proximal caries and showing that AI can accurately identify caries involving enamel, dentine and pulp on panoramic radiographs, supporting the growing role of AI in improving radiographic diagnosis.

Beyond measuring disease prevalence, the analysis identified significant geographical inequalities. Among caries-affected teeth, extraction was about 40% more common in areas described as “dental deserts” than in better-served regions, reinforcing concerns that access to routine dental care remains uneven.

The report also uses UK data as a comparative lens and highlighted differences between the healthcare systems. It noted that treatment patterns varied between the US and the UK and suggested that funding models may influence whether severely damaged teeth are restored or extracted.

The findings illustrate how AI could transform population-level oral health surveillance by enabling radiographs to be analysed consistently at scale. Pearl suggests that the data could inform public health policy, dental education, insurance quality measures and future research, as well as provide a new benchmark against which improvements in oral health and access to care can be measured. Full details of the index can be accessed on the Pearl website.

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