{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/joint-optic-disc-and-cup-segmentation-based","title":"Joint Optic Disc and Cup Segmentation Based on Multi-label Deep Network and Polar Transformation","arxiv_id":"1801.00926","date":"2018-01-03","proceeding":null,"authors":["Huazhu Fu","Jun Cheng","Yanwu Xu","Damon Wing Kee Wong","Jiang Liu","Xiaochun Cao"],"abstract":"Glaucoma is a chronic eye disease that leads to irreversible vision loss. The\ncup to disc ratio (CDR) plays an important role in the screening and diagnosis\nof glaucoma. Thus, the accurate and automatic segmentation of optic disc (OD)\nand optic cup (OC) from fundus images is a fundamental task. Most existing\nmethods segment them separately, and rely on hand-crafted visual feature from\nfundus images. In this paper, we propose a deep learning architecture, named\nM-Net, which solves the OD and OC segmentation jointly in a one-stage\nmulti-label system. The proposed M-Net mainly consists of multi-scale input\nlayer, U-shape convolutional network, side-output layer, and multi-label loss\nfunction. The multi-scale input layer constructs an image pyramid to achieve\nmultiple level receptive field sizes. The U-shape convolutional network is\nemployed as the main body network structure to learn the rich hierarchical\nrepresentation, while the side-output layer acts as an early classifier that\nproduces a companion local prediction map for different scale layers. Finally,\na multi-label loss function is proposed to generate the final segmentation map.\nFor improving the segmentation performance further, we also introduce the polar\ntransformation, which provides the representation of the original image in the\npolar coordinate system. The experiments show that our M-Net system achieves\nstate-of-the-art OD and OC segmentation result on ORIGA dataset.\nSimultaneously, the proposed method also obtains the satisfactory glaucoma\nscreening performances with calculated CDR value on both ORIGA and SCES\ndatasets.","url_abs":"http://arxiv.org/abs/1801.00926v3","url_pdf":"http://arxiv.org/pdf/1801.00926v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"joint-optic-disc-and-cup-segmentation-based","repo_url":"https://github.com/Aniladepu007/Joint-Optic-Disc-and-Cup-Segmentation-Based-on-Multi-Label-Deep-Network-and-Polar-Transformation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"joint-optic-disc-and-cup-segmentation-based","repo_url":"https://github.com/HzFu/DENet_GlaucomaScreen","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"joint-optic-disc-and-cup-segmentation-based","repo_url":"https://github.com/HzFu/MNet_DeepCDR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1801.00926","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}