Answers are grounded in this paper and cite their sources.
Objective: To evaluate the diagnostic accuracy of an artificial intelligence (AI)-based dental caries detection system compared with conventional clinical and radiographic diagnostic methods in patients presenting for routine dental examination. Methods: This comparative diagnostic accuracy study included 210 patients undergoing dental examination. Each participant received a standardized clinical examination followed by digital intraoral bitewing radiography when indicated. Dental caries were assessed using conventional visual-tactile examination and radiographic interpretation by experienced dentists. The same radiographic images were independently analyzed using an AI-based caries detection system based on deep-learning image analysis. A consensus diagnosis established by two experienced dental clinicians, with adjudication by a senior oral and maxillofacial radiologist in cases of disagreement, was considered the reference standard. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), diagnostic accuracy, F1-score and area under the receiver operating characteristic curve (AUC) were calculated. Results: Among the 210 patients, 126 (60.0%) were male and 84 (40.0%) were female, with a mean age of 34.8 ± 12.6 years. Carious lesions were identified in 118 (56.2%) patients according to the reference standard. The AI system demonstrated a sensitivity of 91.5%, specificity of 90.2%, PPV of 91.5%, NPV of 90.2% and overall diagnostic accuracy of 90.9%. Conventional clinical examination demonstrated sensitivity of 82.2%, specificity of 86.9% and diagnostic accuracy of 84.3%. Radiographic interpretation alone demonstrated sensitivity of 87.3%, specificity of 89.1% and diagnostic accuracy of 88.1%. The AUC was 0.934 for AI, compared with 0.846 for conventional clinical examination and 0.882 for radiographic interpretation. AI performance was highest for dentinal lesions and somewhat lower for early enamel lesions. Conclusion: The AI-based system demonstrated high diagnostic accuracy for dental caries and performed better than conventional visual-tactile examination and comparably or better than conventional radiographic interpretation. AI may serve as a useful adjunct to conventional dental diagnosis, particularly for improving consistency and identifying lesions that may be difficult to detect clinically. AI should complement rather than replace professional clinical judgment until validated prospectively in larger, multicenter populations.
No takes yet. Share an insight, caveat, or question.
Idrees et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: