Related Experiment Video
Updated: Apr 20, 2026
10:42
A Postoperative Evaluation Guideline for Computer-Assisted Reconstruction of the Mandible
Published on: January 28, 2020
7.2K
A more accurate soft-tissue prediction model for Class III 2-jaw surgeries
Yun-Sik Lee1, Hee-Yeon Suh1, Shin-Jae Lee2
1Postgraduate student, Department of Orthodontics, School of Dentistry and Dental Research Institute, Seoul National University, Seoul, Korea.
Summary
A new prediction model using partial least squares (PLS) accurately estimates surgical outcomes for Class III malocclusion patients undergoing maxillary or mandibular surgery. This model offers improved predictability for both single-jaw and double-jaw surgical corrections.
Area of Science:
- Orthodontics and Maxillofacial Surgery
- Biostatistics
- Predictive Modeling
Background:
- Bimaxillary surgeries for Class III malocclusions present challenges in accurately predicting outcomes.
- A universal soft-tissue prediction model is needed for Class III surgical-orthodontic patients, accommodating various surgical approaches.
Purpose of the Study:
- To develop an accurate soft-tissue prediction model for Class III malocclusion patients.
- To ensure the model's universal applicability across different surgical corrections (maxillary, mandibular, with/without genioplasty).
Main Methods:
- A study involving 204 patients with severe skeletal Class III malocclusions who underwent mandibular setback surgery.
- Development of a prediction model with 226 independent and 64 dependent variables.
- Comparison of ordinary least squares and partial least squares (PLS) methods using leave-1-out cross-validation, with actual surgical outcomes as the gold standard.
Main Results:
- The multivariate PLS prediction model, utilizing 30 orthogonal components, demonstrated superior prediction quality.
- No significant difference in prediction error patterns was observed between one-jaw and two-jaw surgeries when using the PLS method.
Conclusions:
- A multivariate PLS prediction model with approximately 30 latent variables offers an enhanced algorithm for predicting surgical outcomes in Class III patients.
- This model improves predictability for both single-jaw and double-jaw surgical corrections in Class III malocclusions.
