Access to dental care is far from universal. Across the European Union, 4.7 % of nationals and 6.7 % of non‑EU citizens report unmet dental examination or treatment needs, according to the latest Eurostat data. These figures are consistently higher than those for general medical care, underscoring a systemic gap in oral health provision across Europe.
The challenge is even more pronounced when viewed globally. An analysis of 192 countries shows stark disparities: high-income countries average 7.05 dentists per 10,000 population, while low‑income countries average just 0.55—a more than tenfold gap. Additionally, a 2024 scoping review of 27 low‑income countries, using World Health Organization (WHO) workforce data, confirmed an average of 0.51 dentists per 10,000, compared with 2.83 and 7.62 in middle‑ and high‑income countries, respectively.
These gaps in access underscore the urgent need for scalable, low-cost tools that can reach patients before diseases progress. A recent study in BDJ Open offers a glimpse of a transformative solution: an artificial intelligence (AI) model capable of detecting common oral diseases using simple intraoral photographs taken by patients themselves. The model, trained on 5,000 open-source images, identified plaque, gingivitis, calculus, and cavities with an accuracy comparable to that of experienced dentists—81 % versus 82 %, respectively. Importantly, it performed this analysis on everyday, non-standardized photos, the kind anyone could take with a smartphone.
The implications go beyond efficiency. This is about rethinking the pathway of oral health. By allowing individuals to screen themselves at home, such tools can shift dentistry from a reactive to a preventive model. Early triage could reduce waiting times, catch problems before they worsen, and free up clinical resources for complex cases. In regions where dental professionals are scarce, it could be the difference between receiving care and living with chronic disease.
This technology is in its early stages, but evolving rapidly; the study found that the AI tended to miss subtle signs of gingivitis and occasionally overdiagnosed cavities. However, it may evolve quickly as it’s fed more diverse data and integrated into clinical workflows.
What’s striking is the accessibility of this approach: the model was built on a “lightweight” architecture designed to run on ordinary devices. This means the potential to democratize oral health isn’t a distant dream requiring expensive hardware; it’s something that could fit in your pocket. This work reflects a broader movement in AI-driven healthcare—using technology not to replace clinicians, but to extend their reach and shift power toward prevention and self-care.
Imagine a world where checking for cavities is as simple as snapping a photo, where rural clinics can triage patients without a resident dentist, and where early intervention becomes the norm rather than the exception. This isn’t just an innovation in diagnostics; it’s a step toward transforming healthcare into something more proactive, inclusive, and human-centered.
Coordinator at la Verneda-Sant Martí Learning Community and adjunct professor at the University of Barcelona