Automated detection of dental health conditions from intraoral images using a convolutional neural network approach
Procedia Computer Science · 2026 · 284 · 787-796
DOI: 10.1016/j.procs.2026.07.063 ↗Kurz gefasst
Fachartikel in der Zeitschrift Procedia Computer Science, 2026. Verfasst von Wulan Permata Sandhi und 4 weiteren Autor:innen. Die Arbeit nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Detecting dental diseases is very important as it significantly impacts human health and the function of teeth in chewing food. However, low public awareness of dental health and the uneven distribution of dentists have led to an increase in the number of people suffering from dental diseases. This study aims to design a dental disease detection system using intraoral images and Convolutional Neural Network (CNN). The data used in this study were obtained from two sources: Google Images and orthodontist dentists. The dataset used in this study comprises 5404 images divided into five classes: normal teeth, caries, impaction, root residue, and non-dental conditions. The preprocessing stage includes data cleansing, image resizing, and data augmentation. The CNN model designed through several network layers was trained and tested using divided subsets of the data. The results showed an overall accuracy of 96%, with evaluation metrics such as accuracy, precision, recall, and F1 score indicating that the model performs very well in recognizing and classifying dental images. This system is expected to help dentists diagnose patients’ dental conditions, thereby improving the efficiency of dental health services and reducing the workload of dentists. This research opens opportunities for further development using larger and more diverse datasets, as well as exploring other deep learning model architectures for more optimal results.
Zitieren
APA
Sandhi, W. P., Faroby, M. H. Z. A., Lidiawaty, B. R., Putra, H. P. A., & Brahmanta, A. (2026). Automated detection of dental health conditions from intraoral images using a convolutional neural network approach. Procedia Computer Science, 284, 787-796. https://doi.org/10.1016/j.procs.2026.07.063
Vancouver
Sandhi WP, Faroby MHZA, Lidiawaty BR, Putra HPA, Brahmanta A. Automated detection of dental health conditions from intraoral images using a convolutional neural network approach. Procedia Computer Science. 2026;284:787-796. doi:10.1016/j.procs.2026.07.063
BibTeX
@article{Sandhi2026,
author = {Sandhi, Wulan Permata and Faroby, Mohammad Hamim Zajuli Al and Lidiawaty, Berlian Rahmy and Putra, Haqiqi Pamungkas Arby and Brahmanta, Arya},
title = {Automated detection of dental health conditions from intraoral images using a convolutional neural network approach},
journal = {Procedia Computer Science},
year = {2026},
volume = {284},
pages = {787--796},
doi = {10.1016/j.procs.2026.07.063}
}
EUVIMED