Automated Detection of Caries, Ulcers, Tooth Discoloration, and Gingivitis Through Intraoral Image Analysis

This study presents an automated approach for detecting and predicting four oral conditions: caries, ulcers, tooth discoloration, and gingivitis, using intraoral image analysis. Image processing techniques and artificial intelligence models are applied to identify these conditions from intraoral photographs. The developed system achieved a precision of 0.9366 and a recall of 0.9315, demonstrating a high accuracy in correct predictions. Additionally, the model reached a mAP50 of 0.9409 and a mAP50-95 of 0.6948, showing strong performance across both lenient thresholds and stricter evaluation metrics. These results suggest that this technology could become a valuable tool in the field of dentistry, enabling timely diagnosis and improving the quality of treatment.