{"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/adversarial-deep-structural-networks-for","title":"Adversarial Deep Structural Networks for Mammographic Mass Segmentation","arxiv_id":"1612.05970","date":"2016-12-18","proceeding":null,"authors":["Wentao Zhu","Xiang Xiang","Trac. D. Tran","Xiaohui Xie"],"abstract":"Mass segmentation is an important task in mammogram analysis, providing\neffective morphological features and regions of interest (ROI) for mass\ndetection and classification. Inspired by the success of using deep\nconvolutional features for natural image analysis and conditional random fields\n(CRF) for structural learning, we propose an end-to-end network for\nmammographic mass segmentation. The network employs a fully convolutional\nnetwork (FCN) to model potential function, followed by a CRF to perform\nstructural learning. Because the mass distribution varies greatly with pixel\nposition, the FCN is combined with position priori for the task. Due to the\nsmall size of mammogram datasets, we use adversarial training to control\nover-fitting. Four models with different convolutional kernels are further\nfused to improve the segmentation results. Experimental results on two public\ndatasets, INbreast and DDSM-BCRP, show that our end-to-end network combined\nwith adversarial training achieves the-state-of-the-art results.","url_abs":"http://arxiv.org/abs/1612.05970v2","url_pdf":"http://arxiv.org/pdf/1612.05970v2.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":"adversarial-deep-structural-networks-for","repo_url":"https://github.com/wentaozhu/adversarial-deep-structural-networks","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":null,"task_name":"Position"},{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[{"method_slug":"crf","method_name":"CRF"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"fcn","method_name":"FCN"},{"method_slug":"max-pooling","method_name":"Max Pooling"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}