To investigate the effectiveness of automated dentition analysis on panoramic X-rays (PX) of partially edentulous patients using Mask R-CNN for determining Kennedy’s classification, highlighting its potential to enhance diagnostic accuracy.
Key Findings:
Automated tooth labeling and segmentation were successfully performed using Mask R-CNN, indicating a significant advancement in dental imaging.
The study established a reliable method for determining Kennedy’s classification based on the automated analysis of PXs, which could transform clinical workflows.
The approach demonstrated potential for reducing errors in tooth classification and improving communication among dental professionals, ultimately enhancing patient care.
Interpretation:
The findings suggest that AI can significantly enhance the accuracy of dental diagnostics and treatment planning, particularly in the context of prosthodontics, by providing reliable and efficient classification methods.
Limitations:
The study was limited to a specific patient population and may not generalize to all demographics, potentially affecting the broader applicability of the findings.
Exclusion of certain cases (e.g., fully edentulous jaws) may limit the comprehensiveness of the findings and their relevance to diverse clinical scenarios.
Conclusion:
The use of Mask R-CNN for automated tooth labeling and Kennedy’s classification shows promise in improving the efficiency and accuracy of dental diagnostics.
Analysis of 50 cases shows moderate diagnostic agreement between FNAC and histopathology and identifies pleomorphic adenoma as the most common neoplasm.
