EUVIMEDDescoperă EUVIMED
Open access · CC BY 4.0

Machine Learning Prediction of Orthodontic Extraction Patterns Using Multimodal Features Derived From Intraoral Scans and Lateral Cephalometric Radiographs

Quinn Roederer, Vinicius Chrisosthemos Teixeira iD, Adriel S Araújo, Fernando J Lau, Carlos Eduardo Brito Mascarello iD, Piedro Rockembach Nunes iD, Jakub Olczyk iD, Hannah Pensyl, Sunna Ahmad, Jay Hughes, Lana Helms iD, Allison Bergdoll iD and 6 more

Orthodontics & Craniofacial Research · 2026-11-14

DOI: 10.1111/ocr.70201 ↗

At a glance

Journal article in Orthodontics & Craniofacial Research, published online on 2026-11-14. Written by Quinn Roederer and 17 co-authors. Institution: Indiana University School of Dentistry, Indianapolis, Indiana, USA. 33 references. The full text is freely available (open access, CC BY 4.0). Funded by: Instituto Nacional de Ciência e Tecnologia em Medicina Assistida por Computação Científica, Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, CNPq Productivity Scholarships.

RezumatSubjectsDate bibliograficeCiteRelated studies

Abstract (original language)

Objective

To develop and evaluate a machine learning framework for predicting orthodontic extraction patterns using features extracted from pretreatment intraoral scans (IOS) and lateral cephalometric radiographs (LCRs). Materials and Methods This retrospective study included 420 subjects (mean age, 17.1 years; 61% female) with treatment plans involving dental extractions. Expert consensus‐derived reference labels were established using a validated Delphi process involving four orthodontic experts. The dataset was randomly split into two mutually exclusive subsets: 30% for hyperparameter optimization using grid search with five‐fold cross‐validation and 70% ( n = 294) for model development and evaluation using leave‐one‐out cross‐validation (LOOCV). Performance was assessed using accuracy, sensitivity, specificity, positive and negative predictive values, likelihood ratios and F1 score.

Results

The Gradient Boosting classifier trained exclusively on canine‐molar relationship features achieved the highest overall classification accuracy (79.9%), whereas the Random Forest multimodal model incorporating all IOS‐ and LCR‐derived features achieved 78.2% accuracy and did not provide a significant improvement over canine‐molar relationship features alone. SHAP analysis identified canine and molar relationships as the most influential predictors. For the Random Forest multimodal model, class‐specific sensitivity (recall) was 84% for Class I, 77% for Class II, and 48% for Class III.

Conclusions

Sagittal canine‐molar relationship features contained the greatest discriminative information for extraction pattern classification in this dataset, and incorporating the full multimodal feature set did not improve predictive performance. The substantially lower sensitivity for the underrepresented Class III pattern highlights the need for larger, more balanced datasets.

Subjects

Author keywords

artificial intelligencecephalometryintraoral scansmachine learningorthodonticstreatment planning

Authors and institutions

  • Indiana University School of Dentistry, Indianapolis, Indiana, USA.
  • School of Technology, Pontifical Catholic University of Rio Grande do Sul, Porto Alegre, Brazil.
  • Private Practice, Indianapolis, Indiana, USA.
  • Indiana University School of Medicine, Indianapolis, Indiana, USA.

ORCID profiles: Vinicius Chrisosthemos Teixeira, Carlos Eduardo Brito Mascarello, Piedro Rockembach Nunes, Jakub Olczyk, Lana Helms, Allison Bergdoll, George Eckert, Vinicius Dutra, Carolina Frota, R Scott Conley, Márcio Sarroglia Pinho, Hakan Turkkahraman

Funding

  • Instituto Nacional de Ciência e Tecnologia em Medicina Assistida por Computação Científica
  • Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
  • CNPq Productivity Scholarships
  • Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul

Cite

APA

Roederer, Q., Teixeira, V. C., Araújo, A. S., Lau, F. J., Mascarello, C. E. B., Nunes, P. R., Olczyk, J., Pensyl, H., Ahmad, S., Hughes, J., Helms, L., Bergdoll, A., Eckert, G., Dutra, V., Frota, C., Conley, R. S., Pinho, M. S., & Turkkahraman, H. (2026). Machine Learning Prediction of Orthodontic Extraction Patterns Using Multimodal Features Derived From Intraoral Scans and Lateral Cephalometric Radiographs. Orthodontics & Craniofacial Research. https://doi.org/10.1111/ocr.70201

Vancouver

Roederer Q, Teixeira VC, Araújo AS, Lau FJ, Mascarello CEB, Nunes PR, et al. Machine Learning Prediction of Orthodontic Extraction Patterns Using Multimodal Features Derived From Intraoral Scans and Lateral Cephalometric Radiographs. Orthodontics & Craniofacial Research. 2026. doi:10.1111/ocr.70201

BibTeX
@article{Roederer2026,
  author = {Roederer, Quinn and Teixeira, Vinicius Chrisosthemos and Araújo, Adriel S and Lau, Fernando J and Mascarello, Carlos Eduardo Brito and Nunes, Piedro Rockembach and Olczyk, Jakub and Pensyl, Hannah and Ahmad, Sunna and Hughes, Jay and Helms, Lana and Bergdoll, Allison and Eckert, George and Dutra, Vinicius and Frota, Carolina and Conley, R Scott and Pinho, Márcio Sarroglia and Turkkahraman, Hakan},
  title = {Machine Learning Prediction of Orthodontic Extraction Patterns Using Multimodal Features Derived From Intraoral Scans and Lateral Cephalometric Radiographs},
  journal = {Orthodontics & Craniofacial Research},
  year = {2026},
  doi = {10.1111/ocr.70201}
}
Start for free →