Deep Learning-Based Segmentation of the Maxillary Sinus on Panoramic Radiographs Using MedSAM and DeepLabv3+

Background Panoramic radiographs are used in dental practice because of low radiation dose, patient comfort, and rapid acquisition. However, segmentation of the maxillary sinus remains difficult because of superimposed anatomical structures. We evaluated the performances of MedSAM and DeepLabv3+, two advanced segmentation models, for delineating the maxillary sinus on panoramic radiographs. Methods A total of 1,046 panoramic radiographs were retrospectively collected from a dental hospital and a private clinic. Maxillary sinus boundaries were manually annotated by two oral and maxillofacial radiologists and one general dentist, using the VGG Image Annotator. The dataset was randomly divided into training, validation, and test sets in a 60:20:20 ratio. Binary masks were generated and segmentation was performed using MedSAM and DeepLabv3 + in Python. Model performance was evaluated using Dice Similarity Coefficient (DSC), Intersection over Union (IoU), precision, recall, F1-score, Normalized Surface Distance (NSD), and 95th percentile Hausdorff Distance (HD95). Results Both models achieved high segmentation accuracy. MedSAM and DeepLabv3 + recorded a DSC of 0.9570 and 0.9534 and an IoU of 0.9183 and 0.9124, respectively. The NSD was 0.931 for MedSAM and 0.928 for DeepLabv3+, with HD95 < 0.02 for both models. MedSAM showed a slightly higher accuracy but demanded substantially greater computational complexity (1,487.9 vs. 593.97 Floating Point Operations per second) and parameters (90.49M vs. 26.68M) than DeepLabv3+. Conclusions MedSAM and DeepLabv3 + provided robust and reliable segmentation of the maxillary sinus on panoramic radiographs. These findings support the clinical feasibility of advanced deep learning models for automated sinus segmentation, particularly when three-dimensional imaging is unavailable.