Meta-analysis: AI detects proximal caries with 76% sensitivity and 94% specificity across radiographic modalities

Meta-analysis of 10 studies: AI for proximal caries detection on radiographs had pooled sensitivity 76% (95% CI 70–80%), specificity 94% (90–96%), AUC 0.90. Bitewing outperformed panoramic. High heterogeneity and limited external validation remain barriers to clinical use.

• Systematic review and meta-analysis of 20 studies (10 included in quantitative synthesis) evaluating AI for proximal caries detection on bitewing, panoramic, and periapical radiographs.
• Bitewing radiographs yielded slightly better diagnostic performance than panoramic views (exact figures not reported in source).
• High heterogeneity across studies, only half provided data for meta-analysis, and no external validation of AI models were reported, limiting clinical translation.

RadPigeon summaries are original and for information only. They are not clinical advice.