Related Experiment Video
Updated: Nov 7, 2025
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
2.0K
Three-dimensional virtual planning in mandibular advancement surgery: Soft tissue prediction based on deep learning
Rutger Ter Horst1, Hanneke van Weert1, Tom Loonen2
1Department of Oral and Maxillofacial Surgery, Radboud University Nijmegen Medical Centre, Geert Grooteplein Zuid 10, 6525, GA, Nijmegen, the Netherlands.
Summary
A new deep learning algorithm accurately predicts soft tissue changes after mandibular advancement surgery. This AI model offers a clinically relevant tool for orthognathic surgery planning, outperforming traditional methods.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Surgery
- Orthognathic Surgery
Background:
- Predicting soft tissue changes after mandibular advancement surgery is crucial for treatment planning.
- Traditional methods like the mass tensor model (MTM) have limitations in accuracy.
- Deep learning (DL) offers a potential advancement in simulating these complex anatomical changes.
Purpose of the Study:
- To develop and evaluate a deep learning (DL) algorithm for predicting virtual soft tissue profiles post-mandibular advancement surgery.
- To compare the accuracy of the DL-based prediction with the mass tensor model (MTM).
Main Methods:
- A DL model was trained using 3D photographs and CBCT data from patients undergoing mandibular advancement surgery.
- The DL algorithm and MTM generated soft tissue simulations based on actual surgical movements.
- Simulations were compared to postoperative 3D photographs using distance mapping and mean absolute error (MAE) analysis.
Main Results:
- The DL-based algorithm achieved a mean absolute error (MAE) of 1.0 ± 0.6 mm in the lower face region.
- This was significantly lower than the MAE of MTM-based simulations (1.5 ± 0.5 mm, p=0.02).
- The DL model demonstrated clinically acceptable accuracy for predicting soft tissue changes.
Conclusions:
- The developed DL-based algorithm effectively predicts 3D soft tissue profiles following mandibular advancement surgery.
- The DL approach offers superior accuracy compared to the MTM for soft tissue prediction.
- This AI-driven tool represents a relevant and accurate option for pre-surgical planning in orthognathic surgery.
