Deep Learning-Based Automatic Glenohumeral Joint Segmentation for Determining Whether the Hill-Sachs Lesion Is On-Track or Off-Track

The authors aimed to develop a deep learning-based framework for automated segmentation of CT images to enhance the diagnostic efficiency and consistency in assessing glenoid track width and Hill-Sachs lesions in patients with anterior shoulder dislocation. Their model demonstrated excellent segmentation accuracy and significantly reduced processing time, while also validating the reliability of the Two-Thirds Glenoid Height Technique for measuring glenoid parameters. This approach offers a promising tool for improving surgical planning and treatment strategies in shoulder instability cases.