Comparative Evaluation of Deep Learning Models for the Classification of Impacted Maxillary Canines on Panoramic Radiographs

Tokatlı, Nazlı; Erdem, Buket; Özcan, Mustafa; Turan Maviş, Begüm; Şar, Çağla; Özdemir, Fulya

doi:10.3390/diagnostics16020219

Open AccessArticle

Comparative Evaluation of Deep Learning Models for the Classification of Impacted Maxillary Canines on Panoramic Radiographs

by

Nazlı Tokatlı

^1,*

,

Buket Erdem

²

,

Mustafa Özcan

²

,

Begüm Turan Maviş

³

,

Çağla Şar

²

and

Fulya Özdemir

²

¹

Department of Computer Engineering, Faculty of Engineering and Natural Sciences, Istanbul Health and Technology University, 34275 Istanbul, Turkey

²

Department of Orthodontics, Faculty of Dentistry, Istanbul Health and Technology University, 34275 Istanbul, Turkey

³

Department of Orthodontics, Faculty of Dentistry, Marmara University, 34722 Istanbul, Turkey

^*

Author to whom correspondence should be addressed.

Diagnostics 2026, 16(2), 219; https://doi.org/10.3390/diagnostics16020219

Submission received: 14 November 2025 / Revised: 28 December 2025 / Accepted: 7 January 2026 / Published: 9 January 2026

(This article belongs to the Section Machine Learning and Artificial Intelligence in Diagnostics)

Download

Browse Figures

Versions Notes

Abstract

Background/Objectives: The early and accurate identification of impacted teeth in the maxilla is critical for effective dental treatment planning. Traditional diagnostic methods relying on manual interpretation of radiographic images are often time-consuming and subject to variability. Methods: This study presents a deep learning-based approach for automated classification of impacted maxillary canines using panoramic radiographs. A comparative evaluation of four pre-trained convolutional neural network (CNN) architectures—ResNet50, Xception, InceptionV3, and VGG16—was conducted through transfer learning techniques. In this retrospective single-center study, the dataset comprised 694 annotated panoramic radiographs sourced from the archives of a university dental hospital, with a mildly imbalanced representation of impacted and non-impacted cases. Models were assessed using accuracy, precision, recall, specificity, and F1-score. Results: Among the tested architectures, VGG16 demonstrated superior performance, achieving an accuracy of 99.28% and an F1-score of 99.43%. Additionally, a prototype diagnostic interface was developed to demonstrate the potential for clinical application. Conclusions: The findings underscore the potential of deep learning models, particularly VGG16, in enhancing diagnostic workflows; however, further validation on diverse, multi-center datasets is required to confirm clinical generalizability.

Keywords: deep learning; radiography; panoramic; cuspid; tooth; impacted; artificial intelligence

Share and Cite

MDPI and ACS Style

Tokatlı, N.; Erdem, B.; Özcan, M.; Turan Maviş, B.; Şar, Ç.; Özdemir, F. Comparative Evaluation of Deep Learning Models for the Classification of Impacted Maxillary Canines on Panoramic Radiographs. Diagnostics 2026, 16, 219. https://doi.org/10.3390/diagnostics16020219

AMA Style

Tokatlı N, Erdem B, Özcan M, Turan Maviş B, Şar Ç, Özdemir F. Comparative Evaluation of Deep Learning Models for the Classification of Impacted Maxillary Canines on Panoramic Radiographs. Diagnostics. 2026; 16(2):219. https://doi.org/10.3390/diagnostics16020219

Chicago/Turabian Style

Tokatlı, Nazlı, Buket Erdem, Mustafa Özcan, Begüm Turan Maviş, Çağla Şar, and Fulya Özdemir. 2026. "Comparative Evaluation of Deep Learning Models for the Classification of Impacted Maxillary Canines on Panoramic Radiographs" Diagnostics 16, no. 2: 219. https://doi.org/10.3390/diagnostics16020219

APA Style

Tokatlı, N., Erdem, B., Özcan, M., Turan Maviş, B., Şar, Ç., & Özdemir, F. (2026). Comparative Evaluation of Deep Learning Models for the Classification of Impacted Maxillary Canines on Panoramic Radiographs. Diagnostics, 16(2), 219. https://doi.org/10.3390/diagnostics16020219

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Comparative Evaluation of Deep Learning Models for the Classification of Impacted Maxillary Canines on Panoramic Radiographs

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI