{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T23:24:13Z","timestamp":1787009053092,"version":"build-2736575974"},"reference-count":55,"publisher":"Oxford University Press (OUP)","issue":"11","license":[{"start":{"date-parts":[[2022,8,22]],"date-time":"2022-08-22T00:00:00Z","timestamp":1661126400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,10,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>Chest pain is common, and current risk-stratification methods, requiring 12-lead electrocardiograms (ECGs) and serial biomarker assays, are static and restricted to highly resourced settings. Our objective was to predict myocardial injury using continuous single-lead ECG waveforms similar to those obtained from wearable devices and to evaluate the potential of transfer learning from labeled 12-lead ECGs to improve these predictions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We studied 10\u200a874 Emergency Department (ED) patients who received continuous ECG monitoring and troponin testing from 2020 to 2021. We defined myocardial injury as newly elevated troponin in patients with chest pain or shortness of breath. We developed deep learning models of myocardial injury using continuous lead II ECG from bedside monitors as well as conventional 12-lead ECGs from triage. We pretrained single-lead models on a pre-existing corpus of labeled 12-lead ECGs. We compared model predictions to those of ED physicians.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>A transfer learning strategy, whereby models for continuous single-lead ECGs were first pretrained on 12-lead ECGs from a separate cohort, predicted myocardial injury as accurately as models using patients\u2019 own 12-lead ECGs: area under the receiver operating characteristic curve 0.760 (95% confidence interval [CI], 0.721\u20130.799) and area under the precision\u2013recall curve 0.321 (95% CI, 0.251\u20130.397). Models demonstrated a high negative predictive value for myocardial injury among patients with chest pain or shortness of breath, exceeding the predictive performance of ED physicians, while attending to known stigmata of myocardial injury.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Deep learning models pretrained on labeled 12-lead ECGs can predict myocardial injury from noisy, continuous monitor data early in a patient\u2019s presentation. The utility of continuous single-lead ECG in the risk stratification of chest pain has implications for wearable devices and preclinical settings, where external validation of the approach is needed.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/jamia\/ocac135","type":"journal-article","created":{"date-parts":[[2022,8,22]],"date-time":"2022-08-22T10:38:26Z","timestamp":1661164706000},"page":"1908-1918","source":"Crossref","is-referenced-by-count":21,"title":["Transfer learning enables prediction of myocardial injury from continuous single-lead electrocardiography"],"prefix":"10.1093","volume":"29","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4810-5816","authenticated-orcid":false,"given":"Boyang Tom","family":"Jin","sequence":"first","affiliation":[{"name":"Department of Computer Science, Stanford University , Palo Alto, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Raj","family":"Palleti","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Stanford University , Palo Alto, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Siyu","family":"Shi","sequence":"additional","affiliation":[{"name":"Department of Medicine, Stanford University , Palo Alto, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrew Y","family":"Ng","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Stanford University , Palo Alto, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James V","family":"Quinn","sequence":"additional","affiliation":[{"name":"Department of Emergency Medicine, Stanford University , Palo Alto, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8030-3727","authenticated-orcid":false,"given":"Pranav","family":"Rajpurkar","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, Harvard University , Boston, Massachusetts, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0151-5121","authenticated-orcid":false,"given":"David","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Emergency Medicine, Stanford University , Palo Alto, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2022,8,22]]},"reference":[{"issue":"10","key":"2022101107583444500_ocac135-B1","doi-asserted-by":"crossref","first-page":"e146","DOI":"10.1161\/CIR.0000000000000485","article-title":"Heart disease and stroke statistics\u20142017 update: a report from the American Heart Association","volume":"135","author":"Benjamin","year":"2017","journal-title":"Circulation"},{"issue":"1","key":"2022101107583444500_ocac135-B2","doi-asserted-by":"crossref","first-page":"3966","DOI":"10.1038\/s41467-020-17804-2","article-title":"Machine learning-based prediction of acute coronary syndrome using only the pre-hospital 12-lead electrocardiogram","volume":"11","author":"Al-Zaiti","year":"2020","journal-title":"Nat Commun"},{"issue":"7","key":"2022101107583444500_ocac135-B3","doi-asserted-by":"crossref","first-page":"1029","DOI":"10.1001\/jamainternmed.2016.2498","article-title":"A national study of the prevalence of life-threatening diagnoses in patients with chest pain","volume":"176","author":"Hsia","year":"2016","journal-title":"JAMA Intern Med"},{"issue":"7","key":"2022101107583444500_ocac135-B4","doi-asserted-by":"crossref","first-page":"1207","DOI":"10.1001\/jamainternmed.2015.1674","article-title":"Risk for clinically relevant adverse cardiac events in patients with chest pain at hospital admission","volume":"175","author":"Weinstock","year":"2015","journal-title":"JAMA Intern Med"},{"issue":"18","key":"2022101107583444500_ocac135-B5","doi-asserted-by":"crossref","first-page":"1239","DOI":"10.1056\/NEJM199105023241803","article-title":"Ruling out acute myocardial infarction: a prospective multicenter validation of a 12-hour strategy for patients at low risk","volume":"324","author":"Lee","year":"1991","journal-title":"N Engl J Med"},{"issue":"2","key":"2022101107583444500_ocac135-B6","doi-asserted-by":"crossref","first-page":"181","DOI":"10.7326\/0003-4819-106-2-181","article-title":"Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24 hours of hospitalization","volume":"106","author":"Lee","year":"1987","journal-title":"Ann Intern Med"},{"key":"2022101107583444500_ocac135-B7","year":"2016"},{"issue":"6","key":"2022101107583444500_ocac135-B8","doi-asserted-by":"crossref","first-page":"423","DOI":"10.1097\/MEJ.0000000000000393","article-title":"How do patients with chest pain access Emergency Department care?","volume":"24","author":"Van Severen","year":"2017","journal-title":"Eur J Emerg Med"},{"issue":"6","key":"2022101107583444500_ocac135-B9","doi-asserted-by":"crossref","first-page":"586","DOI":"10.1093\/fampra\/18.6.586","article-title":"Chest pain in general practice or in the hospital emergency department: is it the same?","volume":"18","author":"Buntinx","year":"2001","journal-title":"Fam Pract"},{"issue":"10","key":"2022101107583444500_ocac135-B10","doi-asserted-by":"crossref","first-page":"1205","DOI":"10.1016\/j.amjmed.2017.03.032","article-title":"Troponin testing in patients without chest pain or electrocardiographic ischemic changes","volume":"130","author":"Shimoni","year":"2017","journal-title":"Am J Med"},{"issue":"7","key":"2022101107583444500_ocac135-B11","doi-asserted-by":"crossref","first-page":"869","DOI":"10.1016\/j.amjmed.2019.01.043","article-title":"Appropriateness of cardiac troponin testing: insights from the Use of TROPonin In Acute coronary syndromes (UTROPIA) study","volume":"132","author":"Sandoval","year":"2019","journal-title":"Am J Med"},{"issue":"2","key":"2022101107583444500_ocac135-B12","doi-asserted-by":"crossref","first-page":"212","DOI":"10.1001\/jamainternmed.2017.7360","article-title":"Noninvasive cardiac testing vs clinical evaluation alone in acute chest pain: a secondary analysis of the ROMICAT-II randomized clinical trial","volume":"178","author":"Reinhardt","year":"2018","journal-title":"JAMA Intern Med"},{"issue":"5","key":"2022101107583444500_ocac135-B13","doi-asserted-by":"crossref","first-page":"336","DOI":"10.1161\/CIRCULATIONAHA.120.052501","article-title":"Impact of high-deductible health plans on emergency department patients with nonspecific chest pain and their subsequent care","volume":"144","author":"Chou","year":"2021","journal-title":"Circulation"},{"issue":"22","key":"2022101107583444500_ocac135-B14","doi-asserted-by":"crossref","first-page":"2456","DOI":"10.1161\/CIRCULATIONAHA.118.036528","article-title":"Safely identifying emergency department patients with acute chest pain for early discharge: HEART pathway accelerated diagnostic protocol","volume":"138","author":"Mahler","year":"2018","journal-title":"Circulation"},{"issue":"4","key":"2022101107583444500_ocac135-B15","doi-asserted-by":"crossref","first-page":"419","DOI":"10.1080\/10903127.2017.1399185","article-title":"Detection of STEMI using prehospital serial 12-lead electrocardiograms","volume":"22","author":"Tanguay","year":"2018","journal-title":"Prehosp Emerg Care"},{"issue":"4","key":"2022101107583444500_ocac135-B16","doi-asserted-by":"crossref","first-page":"448","DOI":"10.1016\/j.jelectrocard.2014.03.010","article-title":"Difficult ECGs in STEMI","volume":"47","author":"Ayer","year":"2014","journal-title":"J Electrocardiol"},{"key":"2022101107583444500_ocac135-B17","first-page":"557","article-title":"ECG motion artefact reduction improvements of a chest-based wireless patient monitoring system","volume":"37","author":"Catherwood","year":"2010","journal-title":"Comput Cardiol"},{"issue":"3","key":"2022101107583444500_ocac135-B18","doi-asserted-by":"crossref","first-page":"128","DOI":"10.12968\/bjca.2018.13.3.128","article-title":"Continuous ECG monitoring in hospital: part 2, practical issues","volume":"13","author":"Sampson","year":"2018","journal-title":"Br J Card Nurs"},{"issue":"13","key":"2022101107583444500_ocac135-B19","doi-asserted-by":"crossref","first-page":"1582","DOI":"10.1016\/j.jacc.2020.01.046","article-title":"Wearable devices for ambulatory cardiac monitoring","volume":"75","author":"Sana","year":"2020","journal-title":"J Am Coll Cardiol"},{"issue":"16","key":"2022101107583444500_ocac135-B20","doi-asserted-by":"crossref","first-page":"1189","DOI":"10.1161\/CIRCULATIONAHA.116.025282","article-title":"Telemedicine helps cardiologists extend their reach","volume":"134","author":"Kuehn","year":"2016","journal-title":"Circulation"},{"issue":"1","key":"2022101107583444500_ocac135-B21","doi-asserted-by":"crossref","first-page":"154","DOI":"10.1038\/s41597-020-0495-6","article-title":"PTB-XL, a large publicly available electrocardiography dataset","volume":"7","author":"Wagner","year":"2020","journal-title":"Sci Data"},{"key":"2022101107583444500_ocac135-B22","first-page":"1","article-title":"mice: Multivariate imputation by chained equations in R","volume":"45","author":"van Buuren","year":"2011","journal-title":"J Stat Soft"},{"issue":"20","key":"2022101107583444500_ocac135-B23","doi-asserted-by":"crossref","first-page":"2525","DOI":"10.1093\/eurheartj\/ehm355","article-title":"Universal definition of myocardial infarction: Kristian Thygesen, Joseph S. Alpert and Harvey D. White on behalf of the Joint ESC\/ACCF\/AHA\/WHF Task Force for the Redefinition of Myocardial Infarction","volume":"28","author":"Thygesen","year":"2007","journal-title":"Eur Heart J"},{"key":"2022101107583444500_ocac135-B24","author":"Vaswani"},{"key":"2022101107583444500_ocac135-B25","author":"Natarajan"},{"issue":"12","key":"2022101107583444500_ocac135-B26","doi-asserted-by":"crossref","first-page":"124003","DOI":"10.1088\/1361-6579\/abc960","article-title":"Classification of 12-lead ECGs: the PhysioNet\/Computing in Cardiology Challenge 2020","volume":"41","author":"Perez Alday","year":"2021","journal-title":"Physiol Meas"},{"issue":"1","key":"2022101107583444500_ocac135-B27","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1002\/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3","article-title":"Index for rating diagnostic tests","volume":"3","author":"Youden","year":"1950","journal-title":"Cancer"},{"key":"2022101107583444500_ocac135-B28","doi-asserted-by":"publisher","author":"Paszke","DOI":"10.48550\/arXiv.1912.01703"},{"key":"2022101107583444500_ocac135-B29","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"issue":"7825","key":"2022101107583444500_ocac135-B30","doi-asserted-by":"crossref","first-page":"357","DOI":"10.1038\/s41586-020-2649-2","article-title":"Array programming with NumPy","volume":"585","author":"Harris","year":"2020","journal-title":"Nature"},{"key":"2022101107583444500_ocac135-B31","doi-asserted-by":"publisher","author":"McKinney","DOI":"10.25080\/Majora-92bf1922-00a"},{"issue":"3","key":"2022101107583444500_ocac135-B32","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1038\/s41592-019-0686-2","article-title":"SciPy 1.0: fundamental algorithms for scientific computing in Python","volume":"17","author":"Virtanen","year":"2020","journal-title":"Nat Methods"},{"issue":"3","key":"2022101107583444500_ocac135-B33","doi-asserted-by":"crossref","first-page":"295","DOI":"10.1002\/(SICI)1521-4036(199906)41:3<295::AID-BIMJ295>3.0.CO;2-2","article-title":"Proof of safety in toxicology based on the ratio of two means for normally distributed data","volume":"41","author":"Hauschke","year":"1999","journal-title":"Biom J"},{"issue":"12","key":"2022101107583444500_ocac135-B34","doi-asserted-by":"crossref","first-page":"1809","DOI":"10.1016\/j.amjcard.2020.03.024","article-title":"Comparison between anatomical and functional imaging modalities for evaluation of chest pain in the emergency department","volume":"125","author":"Kargoli","year":"2020","journal-title":"Am J Cardiol"},{"issue":"1","key":"2022101107583444500_ocac135-B35","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1016\/j.ahj.2006.10.004","article-title":"Global Registry of Acute Coronary Events (GRACE) hospital discharge risk score accurately predicts long-term mortality post acute coronary syndrome","volume":"153","author":"Tang","year":"2007","journal-title":"Am Heart J"},{"issue":"6","key":"2022101107583444500_ocac135-B36","doi-asserted-by":"crossref","first-page":"191","DOI":"10.1007\/BF03086144","article-title":"Chest pain in the emergency room: value of the HEART score","volume":"16","author":"Six","year":"2008","journal-title":"NHJL"},{"issue":"8","key":"2022101107583444500_ocac135-B37","doi-asserted-by":"crossref","first-page":"671","DOI":"10.1111\/acem.13922","article-title":"Myocardial infarction can be safely excluded by high-sensitivity troponin I testing 3 hours after emergency department presentation","volume":"27","author":"Peacock","year":"2020","journal-title":"Acad Emerg Med"},{"issue":"2","key":"2022101107583444500_ocac135-B38","doi-asserted-by":"crossref","first-page":"253","DOI":"10.1016\/j.annemergmed.2021.02.007","article-title":"A methodological appraisal of the HEART score and its variants","volume":"78","author":"Green","year":"2021","journal-title":"Ann Emerg Med"},{"issue":"2","key":"2022101107583444500_ocac135-B39","doi-asserted-by":"crossref","first-page":"231","DOI":"10.1016\/j.annemergmed.2021.03.024","article-title":"A prospective evaluation of clinical HEART score agreement, accuracy, and adherence in emergency department chest pain patients","volume":"78","author":"Soares","year":"2021","journal-title":"Ann Emerg Med"},{"issue":"8","key":"2022101107583444500_ocac135-B40","doi-asserted-by":"crossref","first-page":"595","DOI":"10.1136\/emermed-2014-203798","article-title":"HEART score and clinical gestalt have similar diagnostic accuracy for diagnosing ACS in an unselected population of patients with chest pain presenting in the ED","volume":"32","author":"Visser","year":"2015","journal-title":"Emerg Med J"},{"issue":"7","key":"2022101107583444500_ocac135-B41","doi-asserted-by":"crossref","first-page":"442","DOI":"10.1016\/j.tcm.2019.10.010","article-title":"How useful is the smartwatch ECG?","volume":"30","author":"Isakadze","year":"2020","journal-title":"Trends Cardiovasc Med"},{"issue":"3","key":"2022101107583444500_ocac135-B42","doi-asserted-by":"crossref","first-page":"305","DOI":"10.1016\/j.artmed.2006.07.006","article-title":"Comparison between neural networks and multiple logistic regression to predict acute coronary syndrome in the emergency room","volume":"38","author":"Green","year":"2006","journal-title":"Artif Intell Med"},{"issue":"1","key":"2022101107583444500_ocac135-B43","doi-asserted-by":"crossref","first-page":"58","DOI":"10.1016\/j.jelectrocard.2008.07.010","article-title":"In search of the best method to predict acute coronary syndrome using only the electrocardiogram from the emergency department","volume":"42","author":"Forberg","year":"2009","journal-title":"J Electrocardiol"},{"issue":"1","key":"2022101107583444500_ocac135-B44","doi-asserted-by":"crossref","first-page":"e0210103","DOI":"10.1371\/journal.pone.0210103","article-title":"Artificial intelligence to predict needs for urgent revascularization from 12-leads electrocardiography in emergency patients","volume":"14","author":"Goto","year":"2019","journal-title":"PLoS One"},{"issue":"1","key":"2022101107583444500_ocac135-B45","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1016\/j.cvdhj.2020.06.002","article-title":"Survey of current perspectives on consumer-available digital health devices for detecting atrial fibrillation","volume":"1","author":"Ding","year":"2020","journal-title":"Cardiovasc Digit Health J"},{"issue":"17","key":"2022101107583444500_ocac135-B46","doi-asserted-by":"crossref","first-page":"417","DOI":"10.21037\/atm.2019.06.79","article-title":"Apple watch, wearables, and heart rhythm: where do we stand?","volume":"7","author":"Raja","year":"2019","journal-title":"Ann Transl Med"},{"issue":"11","key":"2022101107583444500_ocac135-B47","doi-asserted-by":"crossref","first-page":"1461","DOI":"10.1001\/jamainternmed.2020.3989","article-title":"Accuracy of physicians\u2019 electrocardiogram interpretations: a systematic review and meta-analysis","volume":"180","author":"Cook","year":"2020","journal-title":"JAMA Intern Med"},{"issue":"4","key":"2022101107583444500_ocac135-B48","doi-asserted-by":"crossref","first-page":"149","DOI":"10.1007\/s11897-004-0002-y","article-title":"Atrial fibrillation post-myocardial infarction: frequency, consequences, and management","volume":"1","author":"Bhatia","year":"2004","journal-title":"Curr Heart Fail Rep"},{"key":"2022101107583444500_ocac135-B49","first-page":"54","article-title":"Study of rhythm disturbances in acute myocardial infarction","volume":"66","author":"Chiwhane","year":"2018","journal-title":"J Assoc Physicians India"},{"key":"2022101107583444500_ocac135-B50","author":"Warrick"},{"key":"2022101107583444500_ocac135-B51","doi-asserted-by":"publisher","first-page":"546","DOI":"10.1109\/ICMLA.2017.0-104","author":"Bulbul","year":"2017"},{"key":"2022101107583444500_ocac135-B52","volume-title":"Pattern Recognition","author":"Theodoridis","year":"2009"},{"issue":"3","key":"2022101107583444500_ocac135-B53","doi-asserted-by":"publisher","DOI":"10.1161\/JAHA.120.017871","article-title":"In search of an optimal subset of ECG features to augment the diagnosis of acute coronary syndrome at the Emergency Department","volume":"10","author":"Bouzid","year":"2021","journal-title":"JAHA"},{"key":"2022101107583444500_ocac135-B54","author":"Bodini"},{"key":"2022101107583444500_ocac135-B55","doi-asserted-by":"publisher","author":"Brown","year":"2020","DOI":"10.48550\/arXiv.2005.14165"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/29\/11\/1908\/46379618\/ocac135.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/29\/11\/1908\/46379618\/ocac135.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,10,11]],"date-time":"2022-10-11T04:01:47Z","timestamp":1665460907000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/29\/11\/1908\/6673193"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,22]]},"references-count":55,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2022,8,22]]},"published-print":{"date-parts":[[2022,10,7]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocac135","relation":{},"ISSN":["1067-5027","1527-974X"],"issn-type":[{"value":"1067-5027","type":"print"},{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022,11,1]]},"published":{"date-parts":[[2022,8,22]]}}}