What do systematic reviews say?

Visual caries detection (ICDAS visual/visual-tactile inspection) vs Reference standard (histology, excavation, radiographs/microCT, or tooth separation)

• Visual or visual‐tactile examination to detect and inform the diagnosis of enamel caries 2021 · Cochrane All outcomes

  Up to 3 months between index test and reference standard (in vivo studies)
  Specificity 0.76 [0.67, 0.83]
  36 studies
  GRADE not reported

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

AI algorithms have high pooled specificity (0.90) for detecting dental caries; ungraded, with very high heterogeneity (I2>95%).

• 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)
  Specificity 0.90 [0.85, 0.95]
  20 studies · 29423 participants
  GRADE not reported

Teledentistry (index test): dental caries diagnosis from photographs taken with DSLR or smartphone cameras; store-and-forward vs Visual or clinical examination

Across 13 studies, the specificity of teledentistry caries diagnosis ranged from 83% to 100%; no pooled estimate and certainty not graded.

• Teledentistry accuracy for caries diagnosis: a systematic review of in-vivo studies using extra-oral photography… 2024 All outcomes

  Cross-sectional diagnostic accuracy (index test vs reference standard, in-vivo) · In-vivo: dental clinics/hospitals, schools, and a juvenile detention facility (11 countries) · Overall (studies reporting specificity; narrative synthesis)
  Specificity ranged from 83% to 100% across studies (reported in 13 of 19 studies). No pooled estimate; meta-analysis not performed.
  13 studies
  GRADE not reported

Light-based detection and/or oral spectroscopy (e.g. ViziLite, VELscope, Microlux DL, Identafi 3000) vs Surgical or punch biopsy with histological assessment

• Diagnostic tests for oral cancer and potentially malignant disorders in patients presenting with clinically evident… 2021 · Cochrane All outcomes
  Specificity 0.50 [0.32, 0.68]
  23 studies
  Very low

Vital staining (e.g. toluidine blue, tolonium chloride) vs Surgical or punch biopsy with histological assessment

• Diagnostic tests for oral cancer and potentially malignant disorders in patients presenting with clinically evident… 2021 · Cochrane All outcomes
  Specificity 0.68 [0.58, 0.77]
  20 studies
  Very low

Combined index tests (e.g. vital staining plus light-based detection or cytology) vs Surgical or punch biopsy with histological assessment

• Diagnostic tests for oral cancer and potentially malignant disorders in patients presenting with clinically evident… 2021 · Cochrane All outcomes
  Specificity 0.71 [0.53, 0.84]
  9 studies
  Very low

Brush biopsy (oral cytology) vs Surgical or punch biopsy with histological assessment

• Diagnostic tests for oral cancer and potentially malignant disorders in patients presenting with clinically evident… 2021 · Cochrane All outcomes
  Specificity 0.94 [0.88, 0.97]
  20 studies
  Moderate

COE with vital rinsing (toluidine blue/tolonium chloride) vs Conventional oral examination (COE) alone

• Clinical assessment for the detection of oral cavity cancer and potentially malignant disorders in apparently healthy… 2021 · Cochrane All outcomes
  Specificity 0.91 [0.90, 0.91]
  1 study · 4080 participants
  Moderate
• Clinical assessment for the detection of oral cavity cancer and potentially malignant disorders in apparently healthy… 2021 · Cochrane All outcomes
  Specificity 0.92 [0.91, 0.93]
  1 study · 3895 participants
  Moderate

Remote screening of the oral cavity (mobile application/telemedicine) vs Reference standard: examination and clinical evaluation by a physician/clinician with specialist knowledge or training (long-term follow-up accepted for those screened negative)

• Clinical assessment for the detection of oral cavity cancer and potentially malignant disorders in apparently healthy… 2021 · Cochrane All outcomes
  Summary of findings 4 (no pooled analysis): specificity 0.99 (95% CI 0.99 to 1.00), 1.00 (95% CI 0.91 to 1.00), and 0.72 (95% CI 0.63 to 0.80) across the 3 studies. Certainty VERY LOW.
  3 studies · 3600 participants
  Very low

Mouth self-examination (MSE) vs Reference standard: examination and clinical evaluation by a physician/clinician with specialist knowledge or training (long-term follow-up accepted for those screened negative)

• Clinical assessment for the detection of oral cavity cancer and potentially malignant disorders in apparently healthy… 2021 · Cochrane All outcomes
  Summary of findings 2 (no pooled analysis): specificity 1.00 (95% CI 1.00 to 1.00), 0.95 (95% CI 0.88 to 0.99), 0.44 (95% CI 0.20 to 0.70), and 0.54 (95% CI 0.37 to 0.69) across the 4 studies. Certainty VERY LOW.
  4 studies · 35059 participants
  Very low

Visual examination (VE) vs Reference standard: examination and clinical evaluation by a physician/clinician with specialist knowledge or training (long-term follow-up accepted for those screened negative)

• Clinical assessment for the detection of oral cavity cancer and potentially malignant disorders in apparently healthy… 2021 · Cochrane All outcomes
  Summary of findings 1 (no pooled analysis): for the 7 low-prevalence studies specificity ranged from 0.94 (95% CI 0.88 to 0.97) to 0.99 (95% CI 0.98 to 1.00); for the 3 higher-prevalence studies specificity ranged from…
  10 studies · 25568 participants
  Low