Comparative evaluation of new caries diagnostic tools-Intraoral scanner and artificial intelligence -based caries detection software with visual examination on the efficacy of caries detection in children.

Abstract

Background: Early diagnosis of dental caries is fundamental for preserving tooth structure and implementing preventive treatment in pediatric patients. Conventional visual examination remains the most frequently used diagnostic method; however, recent advances such as artificial intelligence (AI) and intraoral scanners (IOS) have introduced new opportunities for improving diagnostic accuracy. Comparative clinical evidence regarding these modalities in children remains limited. The present study evaluated and compared the diagnostic efficacy of AI-based caries detection software, an intraoral scanner, and visual examination in detecting dental caries among pediatric patients.
Materials and Methods: A cross-sectional clinical study was conducted among 54 pediatric patients attending the Department of Pediatric and Preventive Dentistry, Inderprastha Dental College and Hospital. Caries assessment was performed using three diagnostic modalities: visual examination based on the International Caries Detection and Assessment System (ICDAS), intraoral scanner assessment, and AI-based caries detection software (CARIO). Diagnostic findings were recorded and statistically analyzed using SPSS Version 27. Intergroup comparisons were performed using the Chi-square test, with statistical significance set at p ≤ 0.05. Results: Visual examination detected caries in 51 (94.4%) participants, the intraoral scanner detected caries in 30 (55.6%), and AI-based software detected caries in 53 (98.1%) participants. A significant association was observed between visual examination and the intraoral scanner (p = 0.046) and between visual examination and AI (p < 0.001), whereas the comparison between the intraoral scanner and AI was not statistically significant (p = 0.259). AI demonstrated the highest sensitivity for caries detection, while the intraoral scanner showed comparatively lower diagnostic performance.
Conclusion: Artificial intelligence–based caries detection software demonstrated superior diagnostic performance compared with the intraoral scanner and showed excellent agreement with visual examination. AI appears to be a promising adjunctive tool for early caries detection in children, while visual examination continues to remain an essential clinical diagnostic method.