A Model for Dental Caries Detection Using Machine Learning Based on Mobile Phone Images
نویسندگان
1 PhD Student, Department of Information Technology Management, SR.C., Islamic Azad University, Tehran, Iran
2 Professor, Department of Industrial Management, SR.C., Islamic Azad University, Tehran, Iran
3 Associate Professor, Department of Industrial Management, Ka.C., Islamic Azad University, Karaj, Iran
4 Associate Professor, Department of Oral and Maxillofacial Medicine, School of Dentistry, Tehran University of Medical Sciences, International Campus, Tehran, Iran
doi
10.71856/IMPCS.2025.1207340چکیده
Dental caries is one of the most common oral health problems worldwide, and its early detection plays a crucial role in prevention and reducing treatment costs. This study aims to develop a machine learning-based model for automated detection of dental caries using consumer-grade mobile phone images. The research methodology is based on deep neural networks for classifying images into two groups: healthy and decayed teeth. After collection and preprocessing, the data were used to train and evaluate five convolutional neural network (CNN) models with different architectures. Results showed that the best model achieved 88.1% accuracy, 91.2% sensitivity, 82.9% specificity, and an F1 score of 0.900. This research demonstrates that using ordinary mobile phone images combined with deep learning algorithms can be an efficient tool for initial screening of dental caries. This technology can serve as an accessible tool, especially in underserved areas with limited access to dental services, and help reduce unnecessary visits to healthcare centers.
