ARTIFICIAL INTELLIGENCE PLATFORMS IN DENTAL CARIES DETECTION: A SYSTEMATIC REVIEW AND META-ANALYSIS

Area of Science:

• Dental diagnostics
• Medical imaging analysis
• Artificial Intelligence in healthcare

Background:

• Dental caries detection relies on imaging, with varying diagnostic accuracy.
• Artificial Intelligence (AI) presents a potential advancement in analyzing dental images.
• Assessing AI's current capabilities in caries detection is crucial for clinical integration.

Purpose of the Study:

• To systematically review AI platforms and machine learning methods for dental caries detection.
• To evaluate the accuracy of AI in identifying dental caries from clinical images and radiographs.
• To synthesize current evidence on AI performance in dental caries diagnostics.

Main Methods:

• Systematic literature search across 8 electronic databases (Jan 2000-Mar 2024).
• Extraction of AI platforms, methodologies, accuracies, and study characteristics.
• Quality assessment using QUADAS-2 and CLAIM; meta-analysis for pooled accuracy estimates.

Main Results:

• 45 studies included, utilizing radiographs (33) and clinical images (12).
• 21 AI platforms reported, with accuracies ranging from 41.5% to 98.6%.
• Meta-analysis: mean sensitivity 76%, specificity 91%, AUC 92%, high heterogeneity observed.

Conclusions:

• AI performance in dental caries detection varies significantly across platforms.
• AI demonstrates superior sensitivity and equal specificity compared to bitewing radiography for caries detection.
• Further AI refinement is needed for consistent and reliable performance across imaging modalities.