CUCAI 2025 Archive
← Back to papers

DentAI Vision: AI-Powered Dental X-Ray Analysis for Enhancing Trust and Patient Education

Maha Kesibi, Leila Salem, Het Buddhdev, Kamran Jornacion, Elliott Vince

CUCAI 2025 Proceedings2025

Published 2025/03/26

Abstract

This study addresses the challenge of improving trust between patients and dental professionals by leveraging AIdriven analysis of panoramic dental X-rays. With dental caries being one of the most prevalent oral diseases, early and accurate detection is crucial for effective treatment and prevention. Deep learning algorithms were utilized to detect caries, specifically employing the YOLOv5-Small object detection model, optimized for real-time inference. The methodology involved dataset preprocessing, data augmentation techniques such as horizontal flipping, and model training on the DENTEX MICCAI 2023 dataset. In addition to automated diagnosis, a chatbot powered by DeepSeek-Llama and LangChain was integrated into the platform, providing users with reliable, evidence-based dental health information sourced from over 20 accredited references. The findings demonstrate that the proposed AI system can achieve high diagnostic accuracy while fostering patient education and transparency. This research highlights the potential of AI-powered dental diagnostics in reducing the need for costly second opinions, improving patient-dentist relationships, and promoting informed decision-making in oral healthcare.