Bimaxillary orthognathic surgery is one of the primary treatments for maxillofacial deformities, which not only corrects skeletal structures but also induces significant changes in facial soft tissues and appearance. Accurate prediction of these changes plays a crucial role in treatment planning, clinical decision-making, and managing patient expectations. However, conventional prediction methods are mainly based on simple, empirical, and linear models and lack the ability to adequately represent the complex, nonlinear, and patient-specific relationships between skeletal movements and soft tissue response; this limitation is more pronounced in bimaxillary orthognathic surgery. Despite recent advances in artificial intelligence and deep learning in medicine, there remains a significant research gap in the development of generative models capable of simultaneously utilizing multimodal imaging and anatomical data. Most previous studies have either focused on limited two-dimensional datasets or relied solely on non-generative predictive models, which do not enable realistic visual simulation of postoperative facial outcomes. The present study aims to develop an AI-based generative model for predicting soft tissue changes following bimaxillary orthognathic surgery. The novelty of this research lies in leveraging multi-source data and modeling the complex relationships between skeletal structures and soft tissues. This approach is expected to reduce clinical uncertainties, improve surgical planning accuracy, and enhance clinical decision-making in maxillofacial surgery.

Principal Investigato: Dr Farzaneh Bplandparva

Project Start Year: 2026