{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T01:48:19Z","timestamp":1780969699755,"version":"3.54.1"},"reference-count":57,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2021,1,26]],"date-time":"2021-01-26T00:00:00Z","timestamp":1611619200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"funder":[{"name":"Health and Medical Research Fund Project","award":["07180216"],"award-info":[{"award-number":["07180216"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,3,18]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Objective<\/jats:title>\n                  <jats:p>Accurate risk prediction is important for evaluating early medical treatment effects and improving health care quality. Existing methods are usually designed for dynamic medical data, which require long-term observations. Meanwhile, important personalized static information is ignored due to the underlying uncertainty and unquantifiable ambiguity. It is urgent to develop an early risk prediction method that can adaptively integrate both static and dynamic health data.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Materials and Methods<\/jats:title>\n                  <jats:p>Data were from 6367 patients with Peptic Ulcer Bleeding between 2007 and 2016. This article develops a novel End-to-end Importance-Aware Personalized Deep Learning Approach (eiPDLA) to achieve accurate early clinical risk prediction. Specifically, eiPDLA introduces a long short-term memory with temporal attention to learn sequential dependencies from time-stamped records and simultaneously incorporating a residual network with correlation attention to capture their influencing relationship with static medical data. Furthermore, a new multi-residual multi-scale network with the importance-aware mechanism is designed to adaptively fuse the learned multisource features, automatically assigning larger weights to important features while weakening the influence of less important features.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>Extensive experimental results on a real-world dataset illustrate that our method significantly outperforms the state-of-the-arts for early risk prediction under various settings (eg, achieving an AUC score of 0.944 at 1 year ahead of risk prediction). Case studies indicate that the achieved prediction results are highly interpretable.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>These results reflect the importance of combining static and dynamic health data, mining their influencing relationship, and incorporating the importance-aware mechanism to automatically identify important features. The achieved accurate early risk prediction results save precious time for doctors to timely design effective treatments and improve clinical outcomes.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/jamia\/ocaa306","type":"journal-article","created":{"date-parts":[[2020,11,21]],"date-time":"2020-11-21T12:13:11Z","timestamp":1605960791000},"page":"713-726","source":"Crossref","is-referenced-by-count":12,"title":["Importance-aware personalized learning for early risk prediction using static and dynamic health data"],"prefix":"10.1093","volume":"28","author":[{"given":"Qingxiong","family":"Tan","sequence":"first","affiliation":[{"name":"Department of Computer Science, Hong Kong Baptist University, Hong Kong, Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mang","family":"Ye","sequence":"additional","affiliation":[{"name":"School of Computer Science, Wuhan University, Wuhan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andy Jinhua","family":"Ma","sequence":"additional","affiliation":[{"name":"School of Data and Computer Science, Sun Yat-sen University, Guangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Terry Cheuk-Fung","family":"Yip","sequence":"additional","affiliation":[{"name":"Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Hong Kong, Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Grace Lai-Hung","family":"Wong","sequence":"additional","affiliation":[{"name":"Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Hong Kong, Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pong C","family":"Yuen","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Hong Kong Baptist University, Hong Kong, Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2021,1,26]]},"reference":[{"issue":"10094","key":"2021031906035539400_ocaa306-B1","doi-asserted-by":"crossref","first-page":"613","DOI":"10.1016\/S0140-6736(16)32404-7","article-title":"Peptic ulcer disease","volume":"390","author":"Lanas","year":"2017","journal-title":"Lancet"},{"issue":"9684","key":"2021031906035539400_ocaa306-B2","doi-asserted-by":"crossref","first-page":"119","DOI":"10.1016\/S0140-6736(09)61246-0","article-title":"Famotidine for the prevention of peptic ulcers and oesophagitis in patients taking low-dose aspirin (FAMOUS): a phase III, randomised, double-blind, placebo-controlled trial","volume":"374","author":"Taha","year":"2009","journal-title":"Lancet"},{"issue":"1","key":"2021031906035539400_ocaa306-B3","doi-asserted-by":"crossref","first-page":"82","DOI":"10.1053\/j.gastro.2009.09.063","article-title":"Famotidine is inferior to pantoprazole in preventing recurrence of aspirin-related peptic ulcers or erosions","volume":"138","author":"Ng","year":"2010","journal-title":"Gastroenterology"},{"issue":"10","key":"2021031906035539400_ocaa306-B4","doi-asserted-by":"crossref","first-page":"1080","DOI":"10.1111\/apt.13591","article-title":"Independent risk factors of 30-day outcomes in 1264 patients with peptic ulcer bleeding in the USA: large ulcers do worse","volume":"43","author":"Camus","year":"2016","journal-title":"Aliment Pharmacol Ther"},{"issue":"7","key":"2021031906035539400_ocaa306-B5","doi-asserted-by":"crossref","first-page":"1084","DOI":"10.1093\/jamia\/ocaa074","article-title":"Dr. agent: clinical predictive model via mimicked second opinions","volume":"27","author":"Gao","year":"2020","journal-title":"J Am Med Inform Assoc"},{"issue":"10","key":"2021031906035539400_ocaa306-B6","doi-asserted-by":"crossref","first-page":"1419","DOI":"10.1093\/jamia\/ocy068","article-title":"Opportunities and challenges in developing deep learning models using electronic health records data: a systematic review","volume":"25","author":"Xiao","year":"2018","journal-title":"J Am Med Inform Assoc"},{"issue":"4","key":"2021031906035539400_ocaa306-B7","doi-asserted-by":"crossref","first-page":"1111","DOI":"10.1016\/j.cell.2016.10.018","article-title":"Host and environmental factors influencing individual human cytokine responses","volume":"167","author":"Ter Horst","year":"2016","journal-title":"Cell"},{"issue":"1","key":"2021031906035539400_ocaa306-B8","doi-asserted-by":"crossref","first-page":"87","DOI":"10.1093\/jamiaopen\/ooy011","article-title":"Predictive modeling in urgent care: a comparative study of machine learning approaches","volume":"1","author":"Tang","year":"2018","journal-title":"JAMIA Open"},{"issue":"5","key":"2021031906035539400_ocaa306-B9","doi-asserted-by":"crossref","first-page":"938","DOI":"10.1093\/jamia\/ocv032","article-title":"Automated methods for the summarization of electronic health records","volume":"22","author":"Pivovarov","year":"2015","journal-title":"J Am Med Inform Assoc"},{"issue":"1","key":"2021031906035539400_ocaa306-B10","doi-asserted-by":"crossref","first-page":"198","DOI":"10.1093\/jamia\/ocw042","article-title":"Opportunities and challenges in developing risk prediction models with electronic health records data: a systematic review","volume":"24","author":"Goldstein","year":"2017","journal-title":"J Am Med Inform Assoc"},{"issue":"4","key":"2021031906035539400_ocaa306-B11","doi-asserted-by":"crossref","first-page":"364","DOI":"10.1093\/jamia\/ocy173","article-title":"Natural language processing of symptoms documented in free-text narratives of electronic health records: a systematic review","volume":"26","author":"Koleck","year":"2019","journal-title":"J Am Med Inform Assoc"},{"issue":"10","key":"2021031906035539400_ocaa306-B12","doi-asserted-by":"crossref","first-page":"1056","DOI":"10.1093\/jamia\/ocz041","article-title":"Integration of genetic and clinical information to improve imputation of data missing from electronic health records","volume":"26","author":"Li","year":"2019","journal-title":"J Am Med Inform Assoc"},{"issue":"4","key":"2021031906035539400_ocaa306-B13","doi-asserted-by":"crossref","first-page":"1476","DOI":"10.1016\/j.eswa.2013.08.044","article-title":"Breast cancer diagnosis based on feature extraction using a hybrid of k-means and support vector machine algorithms","volume":"41","author":"Zheng","year":"2014","journal-title":"Expert Syst Appl"},{"key":"2021031906035539400_ocaa306-B14","doi-asserted-by":"crossref","first-page":"112","DOI":"10.1016\/j.jbi.2014.02.001","article-title":"Evolutionary strategy to develop learningbased decision systems. application to breast cancer and liver fibrosis stadialization","volume":"49","author":"Gorunescu","year":"2014","journal-title":"J Biomed Inform"},{"issue":"1","key":"2021031906035539400_ocaa306-B15","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1016\/j.jiph.2018.09.009","article-title":"A predictive method for hepatitis disease diagnosis using ensembles of neuro-fuzzy technique","volume":"12","author":"Nilashi","year":"2019","journal-title":"J Infect Public Health"},{"key":"2021031906035539400_ocaa306-B16","author":"Choi"},{"issue":"2","key":"2021031906035539400_ocaa306-B17","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1093\/jamia\/ocw112","article-title":"Using recurrent neural network models for early detection of heart failure onset","volume":"24","author":"Choi","year":"2017","journal-title":"J Am Med Inform Assoc"},{"key":"2021031906035539400_ocaa306-B18","author":"Aczon","year":"2017"},{"key":"2021031906035539400_ocaa306-B19","author":"Xu"},{"issue":"1","key":"2021031906035539400_ocaa306-B20","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41598-018-24271-9","article-title":"Recurrent neural networks for multivariate time series with missing values","volume":"8","author":"Che","year":"2018","journal-title":"Sci Rep"},{"key":"2021031906035539400_ocaa306-B21","author":"Tan"},{"key":"2021031906035539400_ocaa306-B22","author":"Tan"},{"key":"2021031906035539400_ocaa306-B23","author":"Shukla","year":"2019"},{"key":"2021031906035539400_ocaa306-B24","author":"Baytas"},{"key":"2021031906035539400_ocaa306-B25","author":"Che"},{"key":"2021031906035539400_ocaa306-B26","doi-asserted-by":"crossref","first-page":"103361","DOI":"10.1016\/j.jbi.2019.103361","article-title":"Identifying sub-phenotypes of acute kidney injury using structured and unstructured electronic health record data with memory networks","volume":"102","author":"Xu","year":"2020","journal-title":"J Biomed Inform"},{"key":"2021031906035539400_ocaa306-B27","first-page":"371","article-title":"Interpretable deep models for icu outcome prediction","volume":"2016","author":"Che","year":"2016","journal-title":"AMIA Annu Symp Proc"},{"key":"2021031906035539400_ocaa306-B28","author":"Esteban"},{"key":"2021031906035539400_ocaa306-B29","author":"Wang","year":"2016"},{"key":"2021031906035539400_ocaa306-B30","first-page":"59","article-title":"Shortfuse: biomedical time series representations in the presence of structured information","volume":"68","author":"Fiterau","year":"2017","journal-title":"Proc Mach Learn Res"},{"key":"2021031906035539400_ocaa306-B31","author":"He"},{"key":"2021031906035539400_ocaa306-B32","author":"Lipton","year":"2015"},{"key":"2021031906035539400_ocaa306-B33","first-page":"998","article-title":"A hybrid residual network and long short-term memory method for peptic ulcer bleeding mortality prediction","author":"Tan","year":"2018","journal-title":"AMIA Annu Symp Proc"},{"key":"2021031906035539400_ocaa306-B34","author":"Ioffe","year":"2015"},{"issue":"3","key":"2021031906035539400_ocaa306-B35","doi-asserted-by":"crossref","first-page":"358","DOI":"10.1007\/s10115-004-0154-9","article-title":"Exact indexing of dynamic time warping","volume":"7","author":"Keogh","year":"2005","journal-title":"Knowl Inf Syst"},{"key":"2021031906035539400_ocaa306-B36","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1016\/j.patcog.2017.08.015","article-title":"Time-series averaging using constrained dynamic time warping with tolerance","volume":"74","author":"Morel","year":"2018","journal-title":"Pattern Recogn"},{"key":"2021031906035539400_ocaa306-B37","author":"Xi"},{"issue":"2","key":"2021031906035539400_ocaa306-B38","doi-asserted-by":"crossref","first-page":"279","DOI":"10.1109\/TPAMI.2015.2414429","article-title":"Generalized canonical time warping","volume":"38","author":"Zhou","year":"2016","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"4","key":"2021031906035539400_ocaa306-B39","doi-asserted-by":"crossref","first-page":"1074","DOI":"10.1007\/s10618-018-0565-y","article-title":"Optimizing dynamic time warpings window width for time series data mining applications","volume":"32","author":"Dau","year":"2018","journal-title":"Data Min Knowl Disc"},{"key":"2021031906035539400_ocaa306-B40","author":"Gharghabi"},{"key":"2021031906035539400_ocaa306-B41","author":"Bhattacharya"},{"issue":"3","key":"2021031906035539400_ocaa306-B42","doi-asserted-by":"crossref","first-page":"262","DOI":"10.1093\/jamia\/ocy157","article-title":"Classifying relations in clinical narratives using segment graph convolutional and recurrent neural networks (seggcrns)","volume":"26","author":"Li","year":"2019","journal-title":"J Am Med Inform Assoc"},{"issue":"6","key":"2021031906035539400_ocaa306-B43","doi-asserted-by":"crossref","first-page":"577","DOI":"10.1093\/jamia\/ocz013","article-title":"Comment on:deep learning for pharmacovigilance: recurrent neural network architectures for labeling adverse drug reactions in twitter posts","volume":"26","author":"Magge","year":"2019","journal-title":"J Am Med Inform Assoc"},{"issue":"1","key":"2021031906035539400_ocaa306-B44","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Machine Learn"},{"issue":"1","key":"2021031906035539400_ocaa306-B45","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41746-018-0029-1","article-title":"Scalable and accurate deep learning with electronic health records","volume":"1","author":"Rajkomar","year":"2018","journal-title":"NPJ Digital Med"},{"issue":"1","key":"2021031906035539400_ocaa306-B46","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41598-019-39071-y","article-title":"Predictive modeling of the hospital readmission risk from patients\u2019 claims data using machine learning: a case study on COPD","volume":"9","author":"Min","year":"2019","journal-title":"Sci Rep"},{"issue":"01","key":"2021031906035539400_ocaa306-B47","doi-asserted-by":"crossref","first-page":"16","DOI":"10.1055\/s-0039-1677908","article-title":"AI in health: State of the art, challenges, and future directions","volume":"28","author":"Wang","year":"2019","journal-title":"Yearb Med Inform"},{"issue":"9","key":"2021031906035539400_ocaa306-B48","doi-asserted-by":"crossref","first-page":"1337","DOI":"10.1038\/s41591-019-0548-6","article-title":"Do no harm: a roadmap for responsible machine learning for health care","volume":"25","author":"Wiens","year":"2019","journal-title":"Nat Med"},{"issue":"3","key":"2021031906035539400_ocaa306-B49","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1001\/jamainternmed.2018.7117","article-title":"Deep learning in medicine\u2014promise, progress, and challenges","volume":"179","author":"Wang","year":"2019","journal-title":"JAMA Intern Med"},{"issue":"1","key":"2021031906035539400_ocaa306-B50","doi-asserted-by":"crossref","first-page":"59","DOI":"10.7326\/M19-2548","article-title":"Should health care demand interpretable artificial intelligence or accept \u201cblack box\u201d medicine?","volume":"172","author":"Wang","year":"2020","journal-title":"Ann Intern Med"},{"issue":"2","key":"2021031906035539400_ocaa306-B51","doi-asserted-by":"crossref","first-page":"403","DOI":"10.1053\/j.gastro.2019.04.041","article-title":"Pantoprazole to prevent gastroduodenal events in patients receiving rivaroxaban and\/or aspirin in a randomized, double-blind, placebo-controlled trial","volume":"157","author":"Moayyedi","year":"2019","journal-title":"Gastroenterology"},{"issue":"4","key":"2021031906035539400_ocaa306-B52","doi-asserted-by":"crossref","first-page":"548","DOI":"10.14309\/ajg.0000000000000584","article-title":"Optimal timing of feeding after endoscopic hemostasis in patients with peptic ulcer bleeding: a randomized, noninferiority trial (CRIS KCT0001019)","volume":"115","author":"Gong","year":"2020","journal-title":"Am J Gastroenterol"},{"issue":"11","key":"2021031906035539400_ocaa306-B53","doi-asserted-by":"crossref","first-page":"1464","DOI":"10.1111\/apt.14652","article-title":"Contribution of Helicobacter pylori infection to the risk of peptic ulcer bleeding in patients on nonsteroidal anti-inflammatory drugs, antiplatelet agents, anticoagulants, corticosteroids and selective serotonin reuptake inhibitors","volume":"47","author":"Venerito","year":"2018","journal-title":"Aliment Pharmacol Ther"},{"issue":"4","key":"2021031906035539400_ocaa306-B54","doi-asserted-by":"crossref","first-page":"316","DOI":"10.1016\/j.jfma.2017.07.006","article-title":"Hypoalbuminemia is a predictor of mortality and rebleeding in peptic ulcer bleeding under proton pump inhibitor use","volume":"117","author":"Cheng","year":"2018","journal-title":"J Formos Med Assoc"},{"issue":"5","key":"2021031906035539400_ocaa306-B55","doi-asserted-by":"crossref","first-page":"936","DOI":"10.1016\/j.gie.2016.08.049","article-title":"Relationship between timing of endoscopy and mortality in patients with peptic ulcer bleeding: a nationwide cohort study","volume":"85","author":"Laursen","year":"2017","journal-title":"Gastrointestinal Endosc"},{"issue":"6","key":"2021031906035539400_ocaa306-B56","doi-asserted-by":"crossref","first-page":"1022","DOI":"10.1016\/j.gie.2017.03.1533","article-title":"Association between an increase in blood urea nitrogen at 24 hours and worse outcomes in acute nonvariceal upper GI bleeding","volume":"86","author":"Kumar","year":"2017","journal-title":"Gastrointest Endosc"},{"key":"2021031906035539400_ocaa306-B57","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1155\/2013\/382643","article-title":"Differences in blood urea and creatinine concentrations in earthed and unearthed subjects during cycling exercise and recovery","volume":"2013","author":"Sokal","year":"2013","journal-title":"Evid-Based Complement Altern Med"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/28\/4\/713\/36642113\/ocaa306.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/28\/4\/713\/36642113\/ocaa306.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,3,19]],"date-time":"2021-03-19T06:05:47Z","timestamp":1616133947000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/28\/4\/713\/6120416"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,1,26]]},"references-count":57,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2021,1,26]]},"published-print":{"date-parts":[[2021,3,18]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocaa306","relation":{},"ISSN":["1527-974X"],"issn-type":[{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2021,4,1]]},"published":{"date-parts":[[2021,1,26]]}}}