To develop and validate an artificial intelligence framework for automated risk evaluations of pediatric OSAHS using lateral cephalograms.
Approach:
Key Findings:
Upper airway segmentation achieved a mean DSC of 0.931 and an IoU of 0.872.
The fusion model for OSAHS classification achieved an AUC of 0.945 (95% CI: 0.863–0.994) and an F1 score of 0.933 (95% CI: 0.818–0.995).
AI assistance improved diagnostic accuracy by 0.165 for junior dentists and 0.237 for senior dentists.
Interpretation:
The model demonstrates high accuracy and interpretability, suggesting it can significantly enhance early detection and management of pediatric OSAHS in dental settings.
Limitations:
Retrospective design may introduce selection bias, potentially affecting the reliability of the findings.
Limited generalizability due to the specific patient population and settings.
Conclusion:
The proposed AI framework offers a promising tool for automated diagnosis of pediatric OSAHS using routine LCs, potentially improving early detection and individualized care.
