Labeling and Segmentation

Medical image segmentation partitions image pixels or volume voxels into regions. Labeling assigns each region a meaningful label, and enables you to create ground truth data for training new segmentation algorithms. Use the Medical Image Labeler app to interactively label images using manual, semiautomated, and automated techniques including deep learning networks. You can additionally segment images outside the app using the Medical Segment Anything Model (MedSAM), other deep learning networks, and image processing algorithms.