{"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/deep-neural-networks-improve-radiologists","title":"Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening","arxiv_id":"1903.08297","date":"2019-03-20","proceeding":null,"authors":["Nan Wu","Jason Phang","Jungkyu Park","Yiqiu Shen","Zhe Huang","Masha Zorin","Stanisław Jastrzębski","Thibault Févry","Joe Katsnelson","Eric Kim","Stacey Wolfson","Ujas Parikh","Sushma Gaddam","Leng Leng Young Lin","Kara Ho","Joshua D. Weinstein","Beatriu Reig","Yiming Gao","Hildegard Toth","Kristine Pysarenko","Alana Lewin","Jiyon Lee","Krystal Airola","Eralda Mema","Stephanie Chung","Esther Hwang","Naziya Samreen","S. Gene Kim","Laura Heacock","Linda Moy","Kyunghyun Cho","Krzysztof J. Geras"],"abstract":"We present a deep convolutional neural network for breast cancer screening\nexam classification, trained and evaluated on over 200,000 exams (over\n1,000,000 images). Our network achieves an AUC of 0.895 in predicting whether\nthere is a cancer in the breast, when tested on the screening population. We\nattribute the high accuracy of our model to a two-stage training procedure,\nwhich allows us to use a very high-capacity patch-level network to learn from\npixel-level labels alongside a network learning from macroscopic breast-level\nlabels. To validate our model, we conducted a reader study with 14 readers,\neach reading 720 screening mammogram exams, and find our model to be as\naccurate as experienced radiologists when presented with the same data.\nFinally, we show that a hybrid model, averaging probability of malignancy\npredicted by a radiologist with a prediction of our neural network, is more\naccurate than either of the two separately. To better understand our results,\nwe conduct a thorough analysis of our network's performance on different\nsubpopulations of the screening population, model design, training procedure,\nerrors, and properties of its internal representations.","url_abs":"http://arxiv.org/abs/1903.08297v1","url_pdf":"http://arxiv.org/pdf/1903.08297v1.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":"deep-neural-networks-improve-radiologists","repo_url":"https://github.com/nyukat/breast_cancer_classifier","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"AGPL-3.0"}},{"paper_slug":"deep-neural-networks-improve-radiologists","repo_url":"https://github.com/EdoardoCicero/Neural-Networks-Project","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"breast-cancer-detection","task_name":"Breast Cancer Detection"},{"task_slug":"medical-diagnosis","task_name":"Medical Diagnosis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1903.08297","atlas_url":"https://app.syntology.ai/?focus=1903.08297","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}