Related Experiment Video
Updated: Jul 12, 2025
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
1.8K
Orthognathic surgical planning using graph CNN with dual embedding module: External validations with multi-hospital
In-Hwan Kim1, Jun-Sik Kim1, Jiheon Jeong1
1Department of Biomedical Engineering, Asan Medical Institute of Convergence Science and Technology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea.
Computer Methods and Programs in Biomedicine
|October 19, 2023
Summary
A new AI model accurately predicts surgical movements in orthognathic surgery (OGS). The dual embedding module-graph convolution neural network (DEM-GCNN) shows improved accuracy over existing methods for planning complex jaw surgeries.
Area of Science:
- Artificial Intelligence in Medicine
- Medical Imaging Analysis
- Surgical Planning
Background:
- Predicting surgical movements in orthognathic surgery (OGS) is challenging.
- Current AI advancements may not fully address the complexity of jaw movement prediction.
- Evaluating AI model accuracy for OGS planning is crucial.
Purpose of the Study:
- To assess the prediction accuracy of surgical movements in OGS using a novel DEM-GCNN model.
- To compare the DEM-GCNN model's performance against a pre-existing CNN-based model.
- To validate the model's generalizability across diverse patient data.
Main Methods:
- Utilized 599 lateral cephalogram pairs for training/internal testing and 201 for external testing from 9 institutions.
- Developed a DEM-GCNN model incorporating image and landmark learning for predicting jaw movements.
- Compared prediction accuracy of DEM-GCNN against a CNN-based model (Model-C) using landmark coordinate differences.
Main Results:
- DEM-GCNN showed no significant difference from ground truth for all predicted landmarks (ANS, PNS, B-point, Md1crown).
- DEM-GCNN demonstrated higher successful detection rates than Model-C in both internal and external tests.
- DEM-GCNN significantly outperformed Model-C in prediction accuracy and reduced error values for all jaw movements.
Conclusions:
- A robust OGS planning model (DEM-GCNN) was successfully developed.
- The model exhibits maximized generalizability across varied lateral cephalogram qualities.
- This AI tool enhances the precision of orthognathic surgery planning.
