What do systematic reviews say?

AI algorithms (predominantly CNN/deep learning) as index test vs Dental expert / visual-tactile clinical assessment (reference standard)

AI algorithms have high pooled sensitivity (0.85) for detecting dental caries; ungraded, with very high heterogeneity (I2>95%) and evidence of publication bias.

• Examining the diagnostic accuracy of artificial intelligence for detecting dental caries across a range of imaging… 2025 All outcomes

  Cross-sectional diagnostic accuracy (index test vs reference standard) · Dental imaging: intraoral photographs, periapical, bitewing, panoramic radiographs, CBCT, near-infrared transillumination · Overall (all imaging modalities pooled)
  Sensitivity 0.85 [0.83, 0.93]
  20 studies · 29423 participants
  GRADE not reported

Low polymerization shrinkage resin composite (silorane, ormocer, bulk-fill or other modified-monomer composite) vs Conventional methacrylate-based resin composite

Postoperative sensitivity at 12 months did not differ significantly between low polymerization shrinkage and methacrylate composites (OR 1.65, 95% CI 0.71 to 3.81; Moderate certainty).

• Meta-analysis of the clinical behavior of posterior direct resin restorations: Low polymerization shrinkage resin in… 2018 All outcomes

  12 months · Posterior class I/II restorations in permanent teeth; multicenter randomized controlled trials conducted across Europe, the Americas, and the Middle East
  OR 1.65 [0.71, 3.81]
  10 studies · 970 participants
  GRADE not reported
• Meta-analysis of the clinical behavior of posterior direct resin restorations: Low polymerization shrinkage resin in… 2018 All outcomes

  24 months · Posterior class I/II restorations in permanent teeth; multicenter randomized controlled trials conducted across Europe, the Americas, and the Middle East
  Six articles (seven experimental groups) evaluated postoperative sensitivity after 24 months; four showed no sensitivity in either group and three favored the control group, but the overall meta-analytic effect was not…
  6 studies
  GRADE not reported

Intervention not reported

• Examining the diagnostic accuracy of artificial intelligence for detecting dental caries across a range of imaging… 2025 All outcomes

  Imaging modality: Intraoral photographic images
  Sensitivity 0.80 [0.69, 0.90]
  7 studies
  GRADE not reported

Intervention not reported

• Examining the diagnostic accuracy of artificial intelligence for detecting dental caries across a range of imaging… 2025 All outcomes

  Imaging modality: CBCT
  Sensitivity 0.73 [0.68, 0.77]
  1 study
  GRADE not reported

Intervention not reported

• Examining the diagnostic accuracy of artificial intelligence for detecting dental caries across a range of imaging… 2025 All outcomes

  Imaging modality: Panoramic radiographs
  Sensitivity 0.83 [0.72, 0.94]
  8 studies
  GRADE not reported

Intervention not reported

• Examining the diagnostic accuracy of artificial intelligence for detecting dental caries across a range of imaging… 2025 All outcomes

  Imaging modality: Periapical radiographs
  Sensitivity 0.86 [0.83, 0.90]
  12 studies
  GRADE not reported

Intervention not reported

• Examining the diagnostic accuracy of artificial intelligence for detecting dental caries across a range of imaging… 2025 All outcomes

  Imaging modality: Bitewing radiographs
  Sensitivity 0.81 [0.72, 0.91]
  5 studies
  GRADE not reported

Intervention not reported

• Examining the diagnostic accuracy of artificial intelligence for detecting dental caries across a range of imaging… 2025 All outcomes

  Overall; trim-and-fill (20 observed + 6 imputed studies)
  Sensitivity 0.818 [0.778, 0.858]
  26 studies
  GRADE not reported