Studies from Beihang University Yield New Data on Dentistry (Adapting SAM2 Model from Natural Images for Tooth Segmentation in Dental Panoramic X-Ray Images).

A study conducted by Beihang University in Beijing, China, focused on adapting the SAM2 model from natural images for tooth segmentation in dental panoramic X-ray images. The research aimed to address challenges in accurate tooth segmentation for lesion analysis and treatment planning in dentistry. By fine-tuning the SAM2 model and introducing ScConv modules and gated attention mechanisms, the study demonstrated improved segmentation accuracy and model robustness, particularly with limited sample datasets. The proposed LightUNet model showed comparable performance to traditional models but with significantly fewer parameters and reduced inference time, making it feasible for deployment on edge devices. [Extracted from the article]