Method and system for analysing orbital mr images based on deep learning

Abstract

The invention discloses a method and a system for analysing orbital MR images based on deep learning, belonging to the field of image analysis. The method includes: acquiring MR image data of the patient's orbital region, the MR image data including conventional T1WI sequences, T2Wl plain scan sequences, conventional enhanced sequences, and extraocular muscle fibrosis enhanced sequences; performing preprocessing on the MR images to obtain standardized images; inputting the standardized images into a deep learning segmentation model, and outputting segmentation masks of extraocular muscles, lacrimal glands, and intraorbital fat; calculating quantitative indicators of target structures based on the segmentation masks, the target structures including extraocular muscles, lacrimal glands, and intraorbital fat; generating a diagnostic report containing the quantitative indicators. The invention avoids dependence on manual experience and ensures objective and unified segmentation results of anatomical structures.