Automated segmentation of dental restorations using deep learning: exploring data augmentation techniques
Oral Radiology · 2025-04 · 41 (2) · 207-215
DOI: 10.1007/s11282-024-00794-y ↗Kurz gefasst
Fachartikel in der Zeitschrift Oral Radiology, online erschienen am 09.12.2024. Verfasst von Berrin Çelik und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Ankara Yıldırım Beyazıt University, Ankara, Turkey. berrincelik@aybu.edu.tr. Hauptthemen laut MEDLINE: Deep Learning, Dental Restoration, Permanent, Radiography, Panoramic, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 28 Literaturangaben.
Abstract (Originalsprache)
OBJECTIVES: Deep learning has revolutionized image analysis for dentistry. Automated segmentation of dental radiographs is of great importance towards digital dentistry. The performance of deep learning models heavily relies on the quality and diversity of the training data. Data augmentation is a widely used technique implemented in machine learning and deep learning to artificially increase the size and diversity of a training dataset by applying various transformations to the original data. METHODS: This work aims to automatically segment implants, prostheses, and fillings in panoramic images using 9 different deep learning segmentation models. Later, it explores the effect of data augmentation methods on segmentation performance of the models. Eight different data augmentation techniques are examined. Performance is evaluated by well-accepted metrics such as intersection over union (IoU) and Dice coefficient. RESULTS: While averaging the segmentation results for the three classes, IoU varies between 0.62 and 0.82 while Dice score is between 0.75 and 0.9 among deep learning models used. Augmentation techniques provided performance improvements of up to 3.37%, 5.75% and 8.75% for implant, prosthesis and filling classes, respectively. CONCLUSIONS: Findings reveal that choosing optimal augmentation strategies depends on both model architecture and dental structure type.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Ankara Yıldırım Beyazıt University, Ankara, Turkey. berrincelik@aybu.edu.tr.
- Department of Electrical Electronics Engineering, Gazi University, Ankara, Turkey.
- Biomedical Calibration and Research Center (BIYOKAM), Gazi University, Ankara, Turkey.
ORCID-Profile: Berrin Çelik, Muhammed Emin Baslak, Mehmet Zahid Genç, Mahmut Emin Çelik
Zitieren
APA
Çelik, B., Baslak, M. E., Genç, M. Z., & Çelik, M. E. (2025). Automated segmentation of dental restorations using deep learning: exploring data augmentation techniques. Oral Radiology, 41(2), 207-215. https://doi.org/10.1007/s11282-024-00794-y
Vancouver
Çelik B, Baslak ME, Genç MZ, Çelik ME. Automated segmentation of dental restorations using deep learning: exploring data augmentation techniques. Oral Radiology. 2025;41(2):207-215. doi:10.1007/s11282-024-00794-y
BibTeX
@article{elik2025,
author = {Çelik, Berrin and Baslak, Muhammed Emin and Genç, Mehmet Zahid and Çelik, Mahmut Emin},
title = {Automated segmentation of dental restorations using deep learning: exploring data augmentation techniques},
journal = {Oral Radiology},
year = {2025},
volume = {41},
number = {2},
pages = {207--215},
doi = {10.1007/s11282-024-00794-y}
}
EUVIMED