Automated segmentation of optic disc and cup depicted on color fundus images using a distance-guided deep learning strategy

•A novel annotation region was introduced from the original manual annotation to enhanced boundary information in image segmentation.•A general distance-guided deep learning strategy (DGLS) was developed based on the introduced annotation regions and a distance transform.•Experiments on four public datasets demonstrated that the developed DGLS largely improved the segmentation performance of a given network, making it superior to several available networks.