AI-Enhanced Periodontology: Detecting Bone Loss and Predicting Disease Progression

Periodontal disease remains one of the most prevalent causes of tooth loss worldwide, due at least partly to a lack of means for early detection, objective assessment of alveolar bone loss, and reliable prediction of disease progression. In the last decade, AI has emerged as an increasingly powerful tool, one with the potential to greatly enhance periodontal diagnostics and prognostics beyond traditional clinical and radiographic assessment. This narrative review aims to critically discuss selected current and emerging applications of AI in the specialty of periodontology, with a particular emphasis on automated bone loss detection and predictive modeling regarding periodontal disease progression. Using a narrative, non-systematic approach, the article synthesizes evidence from contemporary peer-reviewed literature spanning dental imaging, machine learning, deep learning, and clinical decision-support systems. The review covers fundamental AI concepts relevant to clinicians, data sources, and model architectures, and the application of AI to two-dimensional radiographs, three-dimensional CBCT imaging, and radiomics-based analysis. It also covers how machine learning models synthesize clinical, radiographic, and systemic variables to predict periodontal disease trajectories and inform personalized care. The results of this review provide evidence that AI systems can attain a high degree of accuracy and reproducibility in detecting periodontal bone loss and hold significant promise in predicting disease trajectories and responses to different treatments. However, daunting challenges are yet to be overcome, such as heterogeneity in data, algorithmic bias, limited external validation, and major ethical and regulatory concerns. Overall, the evidence indicates that AI can help improve diagnostic precision and will accelerate periodontology toward a more predictive and preventive discipline if it is developed responsibly and integrated clinically. Advances will depend on sound validation, ethical stewardship, and sustained collaboration among clinicians, researchers, and developers of technology.

کلیدواژه ها:

Artificial Intelligence ، Periodontology ، Bone Loss ، Deep Learning ، Predictive Models ، Periodontal Disease