Deep Learning–Based Classification of Impacted Teeth from Panoramic Radiographic Images

Widha Permata Wihasti, Jatmiko Endro Suseno, Catur Edi Widodo

Abstract


Impacted teeth are dental conditions in which the eruption process is obstructed by surrounding teeth, bone, or soft tissue. Manual examination of panoramic radiographs to identify impacted teeth is time-consuming and prone to human error. This study aims to develop an automated classification system capable of distinguishing impacted and non-impacted teeth using a Convolutional Neural Network (CNN). The novelty of this work lies in the implementation of an optimized CNN architecture combined with comprehensive preprocessing and augmentation workflows to enhance classification performance on panoramic dental images. The dataset consisted of 2,002 panoramic radiographs sourced from Dr. Margono Soekarjo Regional General Hospital (Purwokerto) and a public dental image repository. Preprocessing included image resizing, grayscale conversion, and normalization. To increase data diversity, several augmentation techniques—such as Gaussian noise, Gaussian blur, histogram equalization, CLAHE, and sharpening—were applied. The trained CNN achieved high performance, with an accuracy of 94.65%, precision of 94.68%, recall of 94.63%, and an F1-score of 94.65%. These results demonstrate that augmentation plays a critical role in improving the model’s generalization capability. Overall, the integration of an optimized CNN with structured preprocessing and augmentation strategies shows strong potential as a clinical decision-support tool for detecting impacted teeth, contributing to improved diagnostic accuracy and efficiency in dental imaging.

Keywords: CNN, panoramic radiograph, impacted teeth, image augmentation, deep learning, image classification


Keywords


CNN; panoramic radiograph; impacted teeth; image augmentation; deep learning; image classification

Full Text:

PDF

References


F. Bourzgui, M. Sebbar, Z. Abidine, and Z. Bentahar, “Management of Dental Impaction,” in Orthodontics - Basic Aspects and Clinical Considerations, InTech, 2012. doi: 10.5772/31199.

S. S. Alam, A. Ahad, S. Ahmed, J. Dudley, and T. H. Farook, “Using deep learning to segment impacted molar teeth from panoramic radiographs,” Digital Dentistry Journal, vol. 1, no. 1, p. 100007, Jun. 2025, doi: 10.1016/j.ddj.2025.100007.

P. J. Andreasen JO, Textbook and Color Atlas of Teeth Impaction: Diagnosis, Treatment, and Prevention, 1st ed. St.Loius: Mosby, 1997.

M. Alsolamy, F. Nadeem, A. A. Azhari, W. Alsolami, and W. M. Ahmed, “Automated detection and labeling of posterior teeth in dental bitewing X-rays using deep learning,” Comput Biol Med, vol. 183, Dec. 2024, doi: 10.1016/j.compbiomed.2024.109262.

E. P. Fang, D. J. Liew, Y. C. Chang, and C. Y. Fang, “Enhancing wisdom teeth detection in panoramic radiographs using multi-channel convolutional neural network with clinical knowledge,” Comput Biol Med, vol. 192, Jun. 2025, doi: 10.1016/j.compbiomed.2025.110368.

T. Zirek, M. Ü. Öziç, and M. Tassoker, “AI-Driven localization of all impacted teeth and prediction of winter angulation for third molars on panoramic radiographs: Clinical user interface design,” Comput Biol Med, vol. 178, Aug. 2024, doi: 10.1016/j.compbiomed.2024.108755.

L. Perez and J. Wang, “The Effectiveness of Data Augmentation in Image Classification using Deep Learning,” Dec. 2017, [Online]. Available: http://arxiv.org/abs/1712.04621

M. Bonfanti-Gris, A. Herrera, M. P. Salido Rodríguez-Manzaneque, F. Martínez-Rus, and G. Pradíes, “Deep learning for teeth detection and segmentation in panoramic radiographs: a systematic review and meta-analysis,” Dec. 01, 2025, BioMed Central Ltd. doi: 10.1186/s12903-025-06349-9.

A. K. V. P. L. B. F. S. M. B. Lisa Schneider, “Assessment of CNNs, transformers, and hybrid architectures in dental image segmentation,” J Dent, vol. 156, 2025.

Y. Y. Huang et al., “Application of Convolutional Neural Networks in an Automatic Judgment System for Teeth Impaction Based on Dental Panoramic Radiography,” Diagnostics, vol. 15, no. 11, Jun. 2025, doi: 10.3390/diagnostics15111363.

R. Widyaningrum et al., “Hybrid two-stage CNN for detection and staging of periodontitis on panoramic radiographs,” J Oral Biol Craniofac Res, vol. 15, no. 6, pp. 1392–1399, Nov. 2025, doi: 10.1016/j.jobcr.2025.08.019.

S. Vinayahalingam et al., “Classification of caries in third molars on panoramic radiographs using deep learning,” Sci Rep, vol. 11, no. 1, Dec. 2021, doi: 10.1038/s41598-021-92121-2.

R. M. Izzetti, “Basic Knowledge and New Advances in Panoramic Radiography Imaging Techniques: A Narrative Review on What Dentists and Radiologists Should Know,” Applied Sciences, 2021.

S. Yilmaz, M. Tasyurek, M. Amuk, M. Celik, and E. M. Canger, “Developing deep learning methods for classification of teeth in dental panoramic radiography,” Oral Surg Oral Med Oral Pathol Oral Radiol, vol. 138, no. 1, pp. 118–127, Jul. 2024, doi: 10.1016/j.oooo.2023.02.021.

G. R. Gonzalez and R. E. Woods, Digital Image Processing, 4th ed. New York: Pearson, 2018.

K. Zaborowicz, M. Zaborowicz, K. Cieślińska, A. Daktera-Micker, M. Firlej, and B. Biedziak, “Artificial Intelligence Methods in the Detection of Oral Diseases on Pantomographic Images—A Systematic Narrative Review,” May 01, 2025, Multidisciplinary Digital Publishing Institute (MDPI). doi: 10.3390/jcm14093262.




DOI: http://dx.doi.org/10.52155/ijpsat.v55.1.7743

Refbacks

  • There are currently no refbacks.


Copyright (c) 2025 Widha Permata Wihasti

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.