Traditional and Artificial Intelligent Methods in Predicting Maxillofacial Soft Tissue Morphology After Orthognathic...

Area of Science:

• Oral and Maxillofacial Surgery
• Medical Imaging
• Artificial Intelligence in Medicine

Background:

• Preoperative prediction of soft tissue morphology after orthognathic surgery remains a significant challenge.
• Traditional 2D prediction methods have limitations in accuracy and scope.

Purpose of the Study:

• To review current prediction methods for soft tissue morphology in orthognathic surgery.
• To evaluate the potential of artificial intelligence (AI), machine learning (ML), and deep learning (DL) in improving prediction accuracy and speed.
• To identify future research directions for enhanced surgical planning.

Main Methods:

• A narrative review of articles was conducted using PubMed.
• Keywords included orthognathic surgery, prediction methods, simulation, AI, ML, DL, soft tissue, and surgical planning.
• Sixty relevant articles were evaluated from an initial search of sixty-seven.

Main Results:

• The transition from 2D to 3D prediction methods has paved the way for AI, especially DL algorithms like DNN, PointNet, and Transformer.
• DL methods demonstrate high accuracy, speed, and ease of use, indicating broad development prospects.
• Key challenges for DL include data insufficiency, overfitting, and lack of interpretability.

Conclusions:

• Deep learning (DL) offers significant potential for improving the accuracy and speed of soft tissue prediction in orthognathic surgery.
• Future improvements can be achieved through large, high-quality databases, specialized datasets, optimized algorithms, multimodal data integration, and visualization techniques.
• Addressing data limitations and interpretability is crucial for the clinical application of DL in surgical planning.