AI-Assisted Risk Stratification of Peri-Implantitis Using Longitudinal Bone Texture Analysis

Abstract

Peri-implantitis is a progressive inflammatory condition that compromises dental implant stability and can lead to implant failure if undetected. Early identification of individuals at high risk remains a clinical challenge due to subtle bone changes and variability in patient-specific factors. This study proposes an AI-assisted risk stratification framework leveraging longitudinal bone texture analysis from radiographic imaging to predict peri-implantitis onset. Advanced machine learning and deep learning models were employed to extract temporal features from trabecular bone patterns, enabling the detection of early microstructural alterations preceding clinical symptoms. The proposed approach demonstrated improved predictive accuracy compared to conventional assessment methods, providing interpretable insights into bone remodeling dynamics and facilitating personalized intervention strategies. These findings underscore the potential of integrating AI and longitudinal imaging for proactive peri-implant disease management, paving the way for more precision dentistry solutions.