Please use this identifier to cite or link to this item: https://hdl.handle.net/11499/57428
Title: Ai-assisted detection of interproximal, occlusal, and secondary caries on bite-wing radiographs: A single-shot deep learning approach
Authors: Karakus, Rabia
Özic, Muhammet Usame
Taşsöker, Melek
Keywords: Artificial intelligence
Deep learning
Detection
Dental caries
GUI
YOLOv8
Artificial-Intelligence
Classification
Management
Algorithms
Accuracy
System
Publisher: Springer
Abstract: Tooth decay is a common oral disease worldwide, but errors in diagnosis can often be made in dental clinics, which can lead to a delay in treatment. This study aims to use artificial intelligence (AI) for the automated detection and localization of secondary, occlusal, and interproximal (D1, D2, D3) caries types on bite-wing radiographs. The eight hundred and sixty bite-wing radiographs were collected from the School of Dentistry database. Pre-processing and data augmentation operations were performed. Interproximal (D1, D2, D3), secondary, and occlusal caries on bite-wing radiographs were annotated by two oral radiologists. The data were split into 80% for training, 10% for validation, and 10% for testing. The AI-based training process was conducted using the YOLOv8 algorithm. A clinical decision support system interface was designed using the Python PyQT5 library, allowing for the use of dental caries detection without the need for complex programming procedures. In the test images, the average precision, average sensitivity, and average F1 score values for secondary, occlusal, and interproximal caries were obtained as 0.977, 0.932, and 0.954, respectively. The AI-based dental caries detection system yielded highly successful results in the test, receiving full approval from dentists for clinical use. YOLOv8 has the potential to increase sensitivity and reliability while reducing the burden on dentists and can prevent diagnostic errors in dental clinics.
URI: https://doi.org/10.1007/s10278-024-01113-x
https://hdl.handle.net/11499/57428
ISSN: 2948-2925
2948-2933
Appears in Collections:PubMed İndeksli Yayınlar Koleksiyonu / PubMed Indexed Publications Collection
Teknoloji Fakültesi Koleksiyonu
WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection

Show full item record



CORE Recommender

WEB OF SCIENCETM
Citations

2
checked on Dec 18, 2024

Page view(s)

30
checked on Aug 24, 2024

Google ScholarTM

Check




Altmetric


Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.