Enhancing Orthodontic Treatment Planning through Support Vector Regression for Predicting Unerupted Permanent Canine and Premolar Widths

Document Type : Original Article

Authors

1 Lecturer of Pediatric Dentistry & Dental Public Health, Kafr El Sheikh University

2 Lecturer of Orthodontics, Faculty of Dentistry, Kafr El Sheikh University, Egypt

Abstract

Aim of this study:
This study aims to enhance orthodontic treatment planning by accurately predicting the mesiodistal widths of unerupted permanent canines and premolars using Support Vector Regression (SVR).
Methodology
The study was conducted in Kafr el-Sheikh, Egypt, involving a sample population of students aged 16 to 22 years. Plaster models of their maxillary and mandibular arches were prepared, and precise measurements of the mesiodistal crown diameters of permanent teeth were recorded. The dataset included dental records, arch dimensions, and demographic information of the patients.
Regression correlation analyses and t-tests were employed to examine the relationship between tooth widths and other dental characteristics. The SVR model was trained on this comprehensive dataset, and its performance in predicting tooth width was evaluated against traditional linear regression methods.
Results
The findings indicate that the SVR model achieved prediction accuracies exceeding 88% across various evaluation metrics. The SVR model demonstrated superior performance in accuracy and precision compared to traditional linear regression techniques.
Conclusion
This study successfully illustrates the potential of SVR to significantly improve orthodontic treatment planning by accurately predicting the widths of unerupted permanent canines and premolars. The experiments prove the efficiency of SVR compared to other traditional regression methods.

Keywords

Main Subjects

Egyptian Dental Journal
Volume 70, Issue 4 - Serial Number 1
October (Orthodontics, Pediatric & Preventive Dentistry)
October 2024
Pages 3053-3061

Files

Share

How to cite

Statistics