Application of Artificial Intelligence in predicting complication after maxillofacial surgery

Background Postoperative complications remain a major concern in maxillofacial and orthognathic surgery and can negatively affect functional and aesthetic outcomes. Artificial intelligence (AI) has recently emerged as a promising approach for analyzing complex clinical and imaging data to support surgical risk assessment. This systematic review aimed to evaluate current evidence on AI‑based models for predicting postoperative complications in maxillofacial surgery. Methods: A systematic search was conducted in PubMed/MEDLINE, Scopus, Web of Science, and IEEE Xplore for studies published between ۲۰۱۵ and ۲۰۲۴. Eligible studies were original investigations that developed or evaluated AI‑based predictive models for postoperative complications following maxillofacial surgical procedures. Study selection followed predefined criteria based on the PICO framework. Data on study characteristics, AI algorithms, input variables, predicted outcomes, and performance metrics were extracted by two independent reviewers. Due to methodological heterogeneity, findings were synthesized qualitatively. Results: Out of ۷۴۲ identified records, ۱۶ studies met the inclusion criteria. Most studies were retrospective and focused on orthognathic or mandibular procedures. AI techniques included artificial neural networks, support vector machines, random forests, convolutional neural networks, and transformer‑based models. Predicted outcomes included neurosensory disturbances, postoperative swelling, infection, and skeletal instability. Overall, AI models demonstrated moderate to high predictive performance, particularly when combining clinical variables with CBCT imaging data. Conclusion: AI‑based predictive models show promising potential for improving complication risk assessment in maxillofacial surgery. However, further prospective multicenter studies with standardized datasets and external validation are required to support clinical implementation.

کلیدواژه ها:

Artificial Intelligence ، Machine Learning ، Deep Learning ، Maxillofacial Surgery ، Orthognathic Surgery ، Postoperative Complications ، Predictive Modeling ، Risk Prediction