Artificial Intelligence for Autonomous Growth Pattern Forecasting in Mixed Dentition Using Skeletal Maturation Signals

Abstract

Accurate prediction of craniofacial growth during the mixed dentition phase is critical for timely orthodontic diagnosis and treatment planning, yet conventional methods rely heavily on clinician experience and static growth indicators. This study presents an artificial intelligence–driven framework for autonomous growth pattern forecasting in mixed dentition using skeletal maturation signals. Multimodal data, including cephalometric radiographs, hand–wrist or cervical vertebral maturation indicators, and clinical parameters, are integrated into machine learning and deep learning models to capture nonlinear growth dynamics. The proposed approach enables automated identification of skeletal maturity stages and individualized prediction of future growth trajectories. Model performance is evaluated using accuracy, robustness, and clinical agreement metrics, demonstrating improved predictive consistency compared to traditional assessment methods. By reducing subjectivity and enhancing early decision-making, this AI-based system supports precision orthodontics and proactive intervention during critical growth periods. The findings highlight the potential of intelligent, data-driven tools to transform growth assessment and personalized care in orthodontics.