{"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/deepem-deep-3d-convnets-with-em-for-weakly","title":"DeepEM: Deep 3D ConvNets With EM For Weakly Supervised Pulmonary Nodule Detection","arxiv_id":"1805.05373","date":"2018-05-14","proceeding":null,"authors":["Wentao Zhu","Yeeleng S. Vang","Yufang Huang","Xiaohui Xie"],"abstract":"Recently deep learning has been witnessing widespread adoption in various\nmedical image applications. However, training complex deep neural nets requires\nlarge-scale datasets labeled with ground truth, which are often unavailable in\nmany medical image domains. For instance, to train a deep neural net to detect\npulmonary nodules in lung computed tomography (CT) images, current practice is\nto manually label nodule locations and sizes in many CT images to construct a\nsufficiently large training dataset, which is costly and difficult to scale. On\nthe other hand, electronic medical records (EMR) contain plenty of partial\ninformation on the content of each medical image. In this work, we explore how\nto tap this vast, but currently unexplored data source to improve pulmonary\nnodule detection. We propose DeepEM, a novel deep 3D ConvNet framework\naugmented with expectation-maximization (EM), to mine weakly supervised labels\nin EMRs for pulmonary nodule detection. Experimental results show that DeepEM\ncan lead to 1.5\\% and 3.9\\% average improvement in free-response receiver\noperating characteristic (FROC) scores on LUNA16 and Tianchi datasets,\nrespectively, demonstrating the utility of incomplete information in EMRs for\nimproving deep learning\nalgorithms.\\footnote{https://github.com/uci-cbcl/DeepEM-for-Weakly-Supervised-Detection.git}","url_abs":"http://arxiv.org/abs/1805.05373v3","url_pdf":"http://arxiv.org/pdf/1805.05373v3.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":"deepem-deep-3d-convnets-with-em-for-weakly","repo_url":"https://github.com/uci-cbcl/DeepEM-for-Weakly-Supervised-Detection","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null},{"paper_slug":"deepem-deep-3d-convnets-with-em-for-weakly","repo_url":"https://github.com/wentaozhu/DeepEM-for-Weakly-Supervised-Detection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"computed-tomography-ct","task_name":"Computed Tomography (CT)"},{"task_slug":"lung-nodule-detection","task_name":"Lung Nodule Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}