FPN-based tooth segmentation improves deep learning dental caries detection

A deep learning model with FPN-based tooth segmentation achieved 89.28% accuracy for detecting dental caries on intraoral images, outperforming models without segmentation.

• DenseNet169 with FPN segmentation achieved the highest accuracy (89.28%).
• Adding segmentation before classification consistently improved performance across all architectures tested.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.