Artificial intelligence-assisted diagnosis of dental caries and periodontitis on periapical radiographs: a clinical validation study

To evaluate the diagnostic performance of an artificial intelligence (AI)-assisted system for detecting dental caries and staging periodontitis using periapical radiographs, and to assess its impact on diagnostic accuracy among clinicians with different levels of clinical experience. This study comprised retrospective AI model development and internal evaluation using 2700 periapical radiographs containing 10,211 annotated teeth, followed by a prospective single-center clinical reader study involving 54 patients. Dental caries detection was performed using a ConvNeXt-Tiny model, whereas periodontitis staging was conducted using a multi-stage framework integrating You Only Look Once (YOLO) instance segmentation, YOLO semantic segmentation, residual neural network (ResNet)-50 regression, and rule-based post-processing. Model performance was assessed using accuracy, sensitivity, specificity, precision, recall, F1-score, and AUC. Diagnostic performance was further compared between experienced and less experienced dentists, with and without AI assistance. AI assistance improved the diagnostic accuracy of less-experienced clinicians from 67.9% to 86.6%, whereas the senior dentist maintained comparable accuracy (94.0% without AI vs. 94.1% with AI). The ConvNeXt-Tiny model achieved an AUC of 0.8828, accuracy of 76.49%, and recall of 88.24%. For periodontitis staging, the AI framework achieved an overall accuracy of 61.3%, weighted F1-score of 61.3%, and a macro-ROC-AUC of 75.4%. The highest recall was observed for Stage III periodontitis (79.6%), whereas lower recall was observed for Stages I and II. The proposed AI-assisted framework enables simultaneous detection of dental caries and automated staging of periodontitis from periapical radiographs. AI substantially improved diagnostic accuracy for less-experienced clinicians while maintaining the high performance of experienced clinicians, supporting its application as a clinical decision-support system in routine dental practice. This proof-of-concept framework illustrates the potential of AI to assist clinicians in assessing dental caries and radiographic periodontal bone loss on periapical radiographs. However, AI outputs require confirmation through clinical examination and professional judgment, particularly for early periodontal bone-loss categories.