{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,22]],"date-time":"2026-03-22T12:14:12Z","timestamp":1774181652894,"version":"3.50.1"},"reference-count":30,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2020,2,11]],"date-time":"2020-02-11T00:00:00Z","timestamp":1581379200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"funder":[{"DOI":"10.13039\/100000092","name":"National Library of Medicine","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000092","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["R01LM012918"],"award-info":[{"award-number":["R01LM012918"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["R01LM012973"],"award-info":[{"award-number":["R01LM012973"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2020,4,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Introduction<\/jats:title>\n                  <jats:p>Classifying whether concepts in an unstructured clinical text are negated is an important unsolved task. New domain adaptation and transfer learning methods can potentially address this issue.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Objective<\/jats:title>\n                  <jats:p>We examine neural unsupervised domain adaptation methods, introducing a novel combination of domain adaptation with transformer-based transfer learning methods to improve negation detection. We also want to better understand the interaction between the widely used bidirectional encoder representations from transformers (BERT) system and domain adaptation methods.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Materials and Methods<\/jats:title>\n                  <jats:p>We use 4 clinical text datasets that are annotated with negation status. We evaluate a neural unsupervised domain adaptation algorithm and BERT, a transformer-based model that is pretrained on massive general text datasets. We develop an extension to BERT that uses domain adversarial training, a neural domain adaptation method that adds an objective to the negation task, that the classifier should not be able to distinguish between instances from 2 different domains.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>The domain adaptation methods we describe show positive results, but, on average, the best performance is obtained by plain BERT (without the extension). We provide evidence that the gains from BERT are likely not additive with the gains from domain adaptation.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Discussion<\/jats:title>\n                  <jats:p>Our results suggest that, at least for the task of clinical negation detection, BERT subsumes domain adaptation, implying that BERT is already learning very general representations of negation phenomena such that fine-tuning even on a specific corpus does not lead to much overfitting.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>Despite being trained on nonclinical text, the large training sets of models like BERT lead to large gains in performance for the clinical negation detection task.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/jamia\/ocaa001","type":"journal-article","created":{"date-parts":[[2020,1,6]],"date-time":"2020-01-06T20:09:51Z","timestamp":1578341391000},"page":"584-591","source":"Crossref","is-referenced-by-count":39,"title":["Does BERT need domain adaptation for clinical negation detection?"],"prefix":"10.1093","volume":"27","author":[{"given":"Chen","family":"Lin","sequence":"first","affiliation":[{"name":"Computational Health Informatics Program, Boston Children\u2019s Hospital and Harvard Medical School, Boston, Massachusetts, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9560-6491","authenticated-orcid":false,"given":"Steven","family":"Bethard","sequence":"additional","affiliation":[{"name":"School of Information, University of Arizona, Tucson, Arizona, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dmitriy","family":"Dligach","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Loyola University Chicago, Chicago, Illinois, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Farig","family":"Sadeque","sequence":"additional","affiliation":[{"name":"Computational Health Informatics Program, Boston Children\u2019s Hospital and Harvard Medical School, Boston, Massachusetts, USA"},{"name":"Department of Pediatrics, Harvard Medical School, Boston, Massachusetts, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guergana","family":"Savova","sequence":"additional","affiliation":[{"name":"Computational Health Informatics Program, Boston Children\u2019s Hospital and Harvard Medical School, Boston, Massachusetts, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Timothy A","family":"Miller","sequence":"additional","affiliation":[{"name":"Computational Health Informatics Program, Boston Children\u2019s Hospital and Harvard Medical School, Boston, Massachusetts, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2020,2,11]]},"reference":[{"issue":"5","key":"2020110613081799800_ocaa001-B1","doi-asserted-by":"crossref","first-page":"552","DOI":"10.1136\/amiajnl-2011-000203","article-title":"2010 i2b2\/VA challenge on concepts, assertions, and relations in clinical text","volume":"18","author":"Uzuner","year":"2011","journal-title":"J Am Med Inform Assoc"},{"key":"2020110613081799800_ocaa001-B2","author":"Pradhan","year":"2013"},{"key":"2020110613081799800_ocaa001-B3","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":"5","key":"2020110613081799800_ocaa001-B4","doi-asserted-by":"crossref","first-page":"301","DOI":"10.1006\/jbin.2001.1029","article-title":"A simple algorithm for identifying negated findings and diseases in discharge summaries","volume":"34","author":"Chapman","year":"2001","journal-title":"J Biomed Inform"},{"issue":"5","key":"2020110613081799800_ocaa001-B5","doi-asserted-by":"crossref","first-page":"839","DOI":"10.1016\/j.jbi.2009.05.002","article-title":"ConText: an algorithm for determining negation, experiencer, and temporal status from clinical reports","volume":"42","author":"Harkema","year":"2009","journal-title":"J Biomed Inform"},{"key":"2020110613081799800_ocaa001-B6","doi-asserted-by":"crossref","first-page":"213","DOI":"10.1016\/j.jbi.2015.02.010","article-title":"DEEPEN: a negation detection system for clinical text incorporating dependency relation into NegEx","volume":"54","author":"Mehrabi","year":"2015","journal-title":"J Biomed Inform"},{"key":"2020110613081799800_ocaa001-B7","article-title":"Dependency Parser-based negation detection in clinical narratives","author":"Sohn","journal-title":"AMIA Jt Summits Transl Sci Proc"},{"key":"2020110613081799800_ocaa001-B8","first-page":"954","author":"Bhatia","year":"2019"},{"issue":"11","key":"2020110613081799800_ocaa001-B9","doi-asserted-by":"crossref","first-page":"e112774","DOI":"10.1371\/journal.pone.0112774","article-title":"Negation\u2019s not solved: generalizability versus optimizability in clinical natural language processing","volume":"9","author":"Wu","year":"2014","journal-title":"PLoS One"},{"key":"2020110613081799800_ocaa001-B10","doi-asserted-by":"crossref","first-page":"165","DOI":"10.18653\/v1\/W17-2320","volume-title":"BioNLP 2017","author":"Miller","year":"2017"},{"key":"2020110613081799800_ocaa001-B11","author":"Blitzer","year":"2007"},{"key":"2020110613081799800_ocaa001-B12","author":"Blitzer","year":"2006"},{"key":"2020110613081799800_ocaa001-B13","author":"Ziser","year":"2017"},{"key":"2020110613081799800_ocaa001-B14","first-page":"414","author":"Miller","year":"2019"},{"key":"2020110613081799800_ocaa001-B15","first-page":"1","article-title":"Domain-adversarial training of neural networks","volume":"17","author":"Ganin","year":"2016","journal-title":"J Mach Learn Res"},{"key":"2020110613081799800_ocaa001-B16","first-page":"1627","author":"Chen","year":"2012"},{"issue":"1\u20132","key":"2020110613081799800_ocaa001-B17","doi-asserted-by":"crossref","first-page":"151","DOI":"10.1007\/s10994-009-5152-4","article-title":"A theory of learning from different domains","volume":"79","author":"Ben-David","year":"2010","journal-title":"Mach Learn"},{"key":"2020110613081799800_ocaa001-B18","first-page":"1638","author":"Akbik","year":"2018"},{"key":"2020110613081799800_ocaa001-B19","first-page":"328","author":"Howard","year":"2018"},{"key":"2020110613081799800_ocaa001-B20","first-page":"4171","author":"Devlin","year":"2019"},{"key":"2020110613081799800_ocaa001-B21","first-page":"2227","author":"Peters","year":"2018"},{"issue":"1","key":"2020110613081799800_ocaa001-B22","doi-asserted-by":"crossref","first-page":"160035","DOI":"10.1038\/sdata.2016.35","article-title":"MIMIC-III, a freely accessible critical care database","volume":"3","author":"Johnson","year":"2016","journal-title":"Sci Data"},{"key":"2020110613081799800_ocaa001-B23","first-page":"72","author":"Alsentzer","year":"2019"},{"key":"2020110613081799800_ocaa001-B24","author":"Huang"},{"key":"2020110613081799800_ocaa001-B25","doi-asserted-by":"crossref","DOI":"10.1093\/jamia\/ocz096","article-title":"Enhancing clinical concept extraction with contextual embeddings","author":"Si","year":"2019","journal-title":"J Am Med Inform Assoc"},{"key":"2020110613081799800_ocaa001-B26","author":"Han","year":"2019"},{"issue":"5","key":"2020110613081799800_ocaa001-B27","doi-asserted-by":"crossref","first-page":"922","DOI":"10.1136\/amiajnl-2012-001317","article-title":"Towards comprehensive syntactic and semantic annotations of the clinical narrative","volume":"20","author":"Albright","year":"2013","journal-title":"J Am Med Inform Assoc"},{"key":"2020110613081799800_ocaa001-B28","author":"Radford"},{"key":"2020110613081799800_ocaa001-B29","article-title":"Language models are unsupervised multitask learners","author":"Radford","year":"2019","journal-title":"OpenAI Blog"},{"key":"2020110613081799800_ocaa001-B30","author":"Yang","year":"2019"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/27\/4\/584\/34152239\/ocaa001.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/27\/4\/584\/34152239\/ocaa001.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2020,11,6]],"date-time":"2020-11-06T19:19:18Z","timestamp":1604690358000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/27\/4\/584\/5733888"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,2,11]]},"references-count":30,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2020,2,11]]},"published-print":{"date-parts":[[2020,4,1]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocaa001","relation":{},"ISSN":["1527-974X"],"issn-type":[{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2020,4]]},"published":{"date-parts":[[2020,2,11]]}}}