{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,26]],"date-time":"2026-03-26T13:37:25Z","timestamp":1774532245604,"version":"3.50.1"},"reference-count":29,"publisher":"Oxford University Press (OUP)","issue":"1","license":[{"start":{"date-parts":[[2019,11,9]],"date-time":"2019-11-09T00:00:00Z","timestamp":1573257600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"Joint Design of Advanced Computing Solutions for Cancer","award":["JDACS4C"],"award-info":[{"award-number":["JDACS4C"]}]},{"DOI":"10.13039\/100000015","name":"U.S. Department of Energy","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000015","id-type":"DOI","asserted-by":"publisher"}]},{"name":"National Cancer Institute of the National Institutes of Health"},{"DOI":"10.13039\/100006224","name":"Argonne National Laboratory","doi-asserted-by":"publisher","award":["DE-AC02-06-CH11357"],"award-info":[{"award-number":["DE-AC02-06-CH11357"]}],"id":[{"id":"10.13039\/100006224","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006227","name":"Lawrence Livermore National Laboratory","doi-asserted-by":"publisher","award":["DEAC52- 07NA27344"],"award-info":[{"award-number":["DEAC52- 07NA27344"]}],"id":[{"id":"10.13039\/100006227","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100008902","name":"Los Alamos National Laboratory","doi-asserted-by":"publisher","award":["DE-AC5206NA25396"],"award-info":[{"award-number":["DE-AC5206NA25396"]}],"id":[{"id":"10.13039\/100008902","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006228","name":"Oak Ridge National Laboratory","doi-asserted-by":"publisher","award":["DE-AC05-00OR22725"],"award-info":[{"award-number":["DE-AC05-00OR22725"]}],"id":[{"id":"10.13039\/100006228","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"NIH","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2020,1,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Objective<\/jats:title>\n                  <jats:p>We implement 2 different multitask learning (MTL) techniques, hard parameter sharing and cross-stitch, to train a word-level convolutional neural network (CNN) specifically designed for automatic extraction of cancer data from unstructured text in pathology reports. We show the importance of learning related information extraction (IE) tasks leveraging shared representations across the tasks to achieve state-of-the-art performance in classification accuracy and computational efficiency.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Materials and Methods<\/jats:title>\n                  <jats:p>Multitask CNN (MTCNN) attempts to tackle document information extraction by learning to extract multiple key cancer characteristics simultaneously. We trained our MTCNN to perform 5 information extraction tasks: (1) primary cancer site (65 classes), (2) laterality (4 classes), (3) behavior (3 classes), (4) histological type (63 classes), and (5) histological grade (5 classes). We evaluated the performance on a corpus of 95 231 pathology documents (71 223 unique tumors) obtained from the Louisiana Tumor\u00a0Registry. We compared the performance of the MTCNN models against single-task CNN models and 2 traditional machine learning approaches, namely support vector machine (SVM) and random forest classifier (RFC).<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>MTCNNs offered superior performance across all 5 tasks in terms of classification accuracy as compared with the other machine learning models. Based on retrospective evaluation, the hard parameter sharing and cross-stitch MTCNN models correctly classified 59.04% and 57.93% of the pathology reports respectively across all 5 tasks. The baseline models achieved 53.68% (CNN), 46.37% (RFC), and 36.75% (SVM). Based on prospective evaluation, the percentages of correctly classified cases across the 5 tasks were 60.11% (hard parameter sharing), 58.13% (cross-stitch), 51.30% (single-task CNN), 42.07% (RFC), and 35.16% (SVM). Moreover, hard parameter sharing MTCNNs outperformed the other models in computational efficiency by using about the same number of trainable parameters as a single-task CNN.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusions<\/jats:title>\n                  <jats:p>The hard parameter sharing MTCNN offers superior classification accuracy for automated coding support of pathology documents across a wide range of cancers and multiple information extraction tasks while maintaining similar training and inference time as those of a single task\u2013specific model.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/jamia\/ocz153","type":"journal-article","created":{"date-parts":[[2019,8,1]],"date-time":"2019-08-01T19:11:09Z","timestamp":1564686669000},"page":"89-98","source":"Crossref","is-referenced-by-count":80,"title":["Automatic extraction of cancer registry reportable information from free-text pathology reports using multitask convolutional neural networks"],"prefix":"10.1093","volume":"27","author":[{"given":"Mohammed","family":"Alawad","sequence":"first","affiliation":[{"name":"Computational Sciences and Engineering Division, Health Data Sciences Institute, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA"}]},{"given":"Shang","family":"Gao","sequence":"additional","affiliation":[{"name":"Computational Sciences and Engineering Division, Health Data Sciences Institute, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA"}]},{"given":"John X","family":"Qiu","sequence":"additional","affiliation":[{"name":"Computational Sciences and Engineering Division, Health Data Sciences Institute, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA"}]},{"given":"Hong Jun","family":"Yoon","sequence":"additional","affiliation":[{"name":"Computational Sciences and Engineering Division, Health Data Sciences Institute, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA"}]},{"given":"J","family":"Blair Christian","sequence":"additional","affiliation":[{"name":"Computational Sciences and Engineering Division, Health Data Sciences Institute, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA"}]},{"given":"Lynne","family":"Penberthy","sequence":"additional","affiliation":[{"name":"Surveillance Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, Maryland, USA"}]},{"given":"Brent","family":"Mumphrey","sequence":"additional","affiliation":[{"name":"Louisiana Tumor Registry, Louisiana State University Health Sciences Center School of Public Health, New Orleans, Louisiana, USA"}]},{"given":"Xiao-Cheng","family":"Wu","sequence":"additional","affiliation":[{"name":"Louisiana Tumor Registry, Louisiana State University Health Sciences Center School of Public Health, New Orleans, Louisiana, USA"}]},{"given":"Linda","family":"Coyle","sequence":"additional","affiliation":[{"name":"Information Management Services Inc, Calverton, Maryland, USA"}]},{"given":"Georgia","family":"Tourassi","sequence":"additional","affiliation":[{"name":"Computational Sciences and Engineering Division, Health Data Sciences Institute, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA"}]}],"member":"286","published-online":{"date-parts":[[2019,11,9]]},"reference":[{"issue":"2","key":"2020110613085708500_ocz153-B1","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1007\/s10549-016-4035-1","article-title":"Using machine learning to parse breast pathology reports","volume":"161","author":"Yala","year":"2017","journal-title":"Breast Cancer Res Treat"},{"key":"2020110613085708500_ocz153-B2","author":"Wu","year":"2012"},{"key":"2020110613085708500_ocz153-B3","doi-asserted-by":"crossref","first-page":"23","DOI":"10.4103\/2153-3539.97788","article-title":"The feasibility of using natural language processing to extract clinical information from breast pathology reports","volume":"3","author":"Buckley","year":"2012","journal-title":"J Pathol Inform"},{"key":"2020110613085708500_ocz153-B4","doi-asserted-by":"crossref","first-page":"277","DOI":"10.1016\/B978-0-12-802115-6.00014-8","volume-title":"Oncology Informatics","author":"Penberthy","year":"2016"},{"issue":"9","key":"2020110613085708500_ocz153-B5","doi-asserted-by":"crossref","first-page":"603","DOI":"10.1016\/j.ijmedinf.2014.06.009","article-title":"Text mining of cancer-related information: Review of current status and future directions","volume":"83","author":"Spasic","year":"2014","journal-title":"Int J Med Inform"},{"issue":"C","key":"2020110613085708500_ocz153-B6","doi-asserted-by":"crossref","first-page":"14","DOI":"10.1016\/j.jbi.2017.07.012","article-title":"Natural language processing systems for capturing and standardizing unstructured clinical information","volume":"73","author":"Kreimeyer","year":"2017","journal-title":"J Biomed Inform"},{"key":"2020110613085708500_ocz153-B7","doi-asserted-by":"crossref","first-page":"34","DOI":"10.1016\/j.jbi.2017.11.011","article-title":"Clinical information extraction applications: A literature review","volume":"77","author":"Wang","year":"2018","journal-title":"J Biomed Inform"},{"issue":"1","key":"2020110613085708500_ocz153-B8","doi-asserted-by":"crossref","first-page":"163","DOI":"10.1016\/j.jbi.2010.07.006","article-title":"Natural language processing methods and systems for biomedical ontology learning","volume":"44","author":"Liu","year":"2011","journal-title":"J Biomed Informatics"},{"key":"2020110613085708500_ocz153-B9","author":"Currie","year":"2006"},{"key":"2020110613085708500_ocz153-B10","first-page":"41","author":"Ou","year":"2014"},{"key":"2020110613085708500_ocz153-B11","author":"Kavuluru","year":"2013"},{"key":"2020110613085708500_ocz153-B12","author":"Nguyen"},{"key":"2020110613085708500_ocz153-B13","first-page":"369","author":"Yoon","year":"2017"},{"key":"2020110613085708500_ocz153-B14","first-page":"2493","article-title":"Natural language processing (almost) from scratch","volume":"12","author":"Collobert","year":"2011","journal-title":"J Mach Learn Res"},{"key":"2020110613085708500_ocz153-B15","doi-asserted-by":"crossref","first-page":"55","DOI":"10.1109\/MCI.2018.2840738","article-title":"Recent trends in deep learning based natural language processing [review article]","volume":"13","author":"Young","journal-title":"IEEE Comput Intell Mag"},{"issue":"1","key":"2020110613085708500_ocz153-B16","doi-asserted-by":"crossref","first-page":"244","DOI":"10.1109\/JBHI.2017.2700722","article-title":"Deep learning for automated extraction of primary sites from cancer pathology reports","volume":"22","author":"Qiu","year":"2018","journal-title":"IEEE J Biomed Health Inform"},{"issue":"3","key":"2020110613085708500_ocz153-B17","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1093\/jamia\/ocx131","article-title":"Hierarchical attention networks for information extraction from cancer pathology reports","volume":"25","author":"Gao","year":"2018","journal-title":"J Am Med Inform Assoc"},{"key":"2020110613085708500_ocz153-B18","author":"Zhang","year":"2018"},{"key":"2020110613085708500_ocz153-B19","author":"Ruder","year":"2017"},{"key":"2020110613085708500_ocz153-B20","doi-asserted-by":"crossref","first-page":"195","DOI":"10.1007\/978-3-319-47898-2_21","volume-title":"Advances in Big Data","author":"Yoon","year":"2017"},{"key":"2020110613085708500_ocz153-B21","first-page":"218","author":"Alawad","year":"2018"},{"issue":"1","key":"2020110613085708500_ocz153-B22","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1023\/A:1007327622663","article-title":"A Bayesian\/information theoretic model of learning to learn via multiple task sampling","volume":"28","author":"Baxter","year":"1997","journal-title":"Machine Learning"},{"key":"2020110613085708500_ocz153-B23","first-page":"676","author":"Yim","year":"2014"},{"key":"2020110613085708500_ocz153-B24","first-page":"3994","author":"Misra","year":"2016"},{"key":"2020110613085708500_ocz153-B25","first-page":"253","author":"Zhang","year":"2017"},{"key":"2020110613085708500_ocz153-B26","author":"Kim"},{"issue":"4","key":"2020110613085708500_ocz153-B27","doi-asserted-by":"crossref","first-page":"427","DOI":"10.1016\/j.ipm.2009.03.002","article-title":"A systematic analysis of performance measures for classification tasks","volume":"45","author":"Sokolova","year":"2009","journal-title":"Inf Process Manage"},{"key":"2020110613085708500_ocz153-B28","first-page":"317","author":"Zhang","year":"2015"},{"key":"2020110613085708500_ocz153-B29","volume-title":"Chapman and Hall\/CRC Monographs on Statistics and Applied Probability","author":"Efron","year":"1994"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/27\/1\/89\/34152435\/ocz153.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/27\/1\/89\/34152435\/ocz153.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2020,11,6]],"date-time":"2020-11-06T19:23:56Z","timestamp":1604690636000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/27\/1\/89\/5618621"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,11,9]]},"references-count":29,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2019,11,9]]},"published-print":{"date-parts":[[2020,1,1]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocz153","relation":{},"ISSN":["1527-974X"],"issn-type":[{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2020,1]]},"published":{"date-parts":[[2019,11,9]]}}}