Revolutionize dental diagnostics: AI-powered detection and localization of oral lesions using edge devices

• Published in: Network modeling and analysis in health informatics and bioinformatics (Wien) Vol. 14; no. 1; p. 108
• Main Authors: Bonny, Talal, Hammal, Abd Alrhman, Nassan, Wafaa Al, Elhoseny, Mohamed
• Format: Journal Article
• Language: English
• Published: Vienna Springer Vienna 22.09.2025; Springer Nature B.V
• Subjects: Abnormalities; Accuracy; Algorithms; Applications of Graph Theory and Complex Networks; Artificial intelligence; Artificial neural networks; Bioinformatics; Computational Biology/Bioinformatics; Computer Science; Datasets; Deep learning; Edge computing; Health Informatics; Lesions; Localization; Medical errors; Neural networks; Original Article; Real time; X-rays; Raspberry pi implementation; Bitewing images; Dental image analysis; Convolutional neural network; Automated diagnosis
• ISSN: 2192-6670; 2192-6662; 2192-6670
• DOI: 10.1007/s13721-025-00600-7

Detecting and diagnosing oral lesions in bitewing dental images are crucial in early intervention and improved oral healthcare outcomes since traditional methods depend on dental professionals to identify the abnormalities. The aim of this paper is to propose an automated system using deeplearning to detect and diagnose lesions of bitewing images in order to overcome the limitation of traditional detection method. The solution proposed in this paper combines a Convolutional Neural Network (CNN) with the YOLOv5 algorithm. This integration enables lesions detection in images. Then, the Raspberry Pi is used as a low-cost edge device to implement the AI-based Diagnosing system. The proposed AI-based solution could be used as an assistant for diagnosing and detecting oral lesions. The integration between AI techniques and automated systems can guarantee the accurate and re- peatable execution of dental procedures.