CavScan: Intelligent Dental Diagnostic System

• Published in: 2025 5th International Conference on Pervasive Computing and Social Networking (ICPCSN) pp. 1846 - 1851
• Main Authors: Kapse, Vaibhavi, Chaudhari, Gaurav, Tandale, Shivam, Gupta, Megha
• Format: Conference Proceeding
• Language: English
• Published: IEEE 14.05.2025
• Subjects: Accuracy; AI in Healthcare; Artificial intelligence; Clinical Decision Support; Convolutional neural networks; Deep Learning; Dental Anomaly Detection; Dentistry; Medical Image Classification; Mobile applications; Pervasive computing; Planning; Real-time systems; Social networking (online); Teeth; VGG16
• DOI: 10.1109/ICPCSN65854.2025.11035730

Early and accurate detection of dental anomalies is crucial for effective treatment planning and overall oral health management. This paper presents CavScan, an intelligent AI-based dental diagnostic system designed to automate the classification of common dental anomalies and generate structured diagnostic reports. The primary objective of this work is to bridge the gap between manual dental assessments and scalable, real-time AI diagnostics. A VGG16 convolutional neural network (CNN) was fine-tuned using a publicly available Kaggle dataset containing images of six dental conditions: Calculus, Caries, Gingivitis, Hypodontia, Mouth Ulcer, and Tooth Discoloration. The model achieved an accuracy of 79% and was integrated with CrewAI and the Gemini API to generate comprehensive and explainable diagnostic reports. CavScan enhances clinical decision-making by reducing diagnostic time, improving consistency, and enabling remote assessment capabilities. The system provides a scalable framework for deploying AI in dentistry, with future improvements aimed at increasing diagnostic accuracy and supporting real-time mobile applications.