{"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/learning-to-summarize-radiology-findings","title":"Learning to Summarize Radiology Findings","arxiv_id":"1809.04698","date":"2018-09-12","proceeding":"WS 2018 10","authors":["Yuhao Zhang","Daisy Yi Ding","Tianpei Qian","Christopher D. Manning","Curtis P. Langlotz"],"abstract":"The Impression section of a radiology report summarizes crucial radiology\nfindings in natural language and plays a central role in communicating these\nfindings to physicians. However, the process of generating impressions by\nsummarizing findings is time-consuming for radiologists and prone to errors. We\npropose to automate the generation of radiology impressions with neural\nsequence-to-sequence learning. We further propose a customized neural model for\nthis task which learns to encode the study background information and use this\ninformation to guide the decoding process. On a large dataset of radiology\nreports collected from actual hospital studies, our model outperforms existing\nnon-neural and neural baselines under the ROUGE metrics. In a blind experiment,\na board-certified radiologist indicated that 67% of sampled system summaries\nare at least as good as the corresponding human-written summaries, suggesting\nsignificant clinical validity. To our knowledge our work represents the first\nattempt in this direction.","url_abs":"http://arxiv.org/abs/1809.04698v2","url_pdf":"http://arxiv.org/pdf/1809.04698v2.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":"learning-to-summarize-radiology-findings","repo_url":"https://github.com/yuhaozhang/summarize-radiology-findings","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"learning-to-summarize-radiology-findings","repo_url":"https://github.com/xtie97/pet-pgn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.04698","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}