{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T09:02:58Z","timestamp":1782291778576,"version":"3.54.5"},"reference-count":52,"publisher":"Oxford University Press (OUP)","issue":"3","license":[{"start":{"date-parts":[[2024,12,28]],"date-time":"2024-12-28T00:00:00Z","timestamp":1735344000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,3,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Objective<\/jats:title>\n                  <jats:p>Early detection of surgical complications allows for timely therapy and proactive risk mitigation. Machine learning (ML) can be leveraged to identify and predict patient risks for postoperative complications. We developed and validated the effectiveness of predicting postoperative complications using a novel surgical Variational Autoencoder (surgVAE) that uncovers intrinsic patterns via cross-task and cross-cohort presentation learning.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Materials and Methods<\/jats:title>\n                  <jats:p>This retrospective cohort study used data from the electronic health records of adult surgical patients over 4 years (2018-2021). Six key postoperative complications for cardiac surgery were assessed: acute kidney injury, atrial fibrillation, cardiac arrest, deep vein thrombosis or pulmonary embolism, blood transfusion, and other intraoperative cardiac events. We compared surgVAE\u2019s prediction performance against widely-used ML models and advanced representation learning and generative models under 5-fold cross-validation.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>89\u00a0246 surgeries (49% male, median [IQR] age: 57 [45-69]) were included, with 6502 in the targeted cardiac surgery cohort (61% male, median [IQR] age: 60 [53-70]). surgVAE demonstrated generally superior performance over existing ML solutions across postoperative complications of cardiac surgery patients, achieving macro-averaged AUPRC of 0.409 and macro-averaged AUROC of 0.831, which were 3.4% and 3.7% higher, respectively, than the best alternative method (by AUPRC scores). Model interpretation using Integrated Gradients highlighted key risk factors based on preoperative variable importance.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Discussion and Conclusion<\/jats:title>\n                  <jats:p>Our advanced representation learning framework surgVAE showed excellent discriminatory performance for predicting postoperative complications and addressing the challenges of data complexity, small cohort sizes, and low-frequency positive events. surgVAE enables data-driven predictions of patient risks and prognosis while enhancing the interpretability of patient risk profiles.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/jamia\/ocae316","type":"journal-article","created":{"date-parts":[[2024,12,28]],"date-time":"2024-12-28T12:11:22Z","timestamp":1735387882000},"page":"459-469","source":"Crossref","is-referenced-by-count":7,"title":["A novel generative multi-task representation learning approach for predicting postoperative complications in cardiac surgery patients"],"prefix":"10.1093","volume":"32","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-6259-7509","authenticated-orcid":false,"given":"Junbo","family":"Shen","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering, Washington University in St Louis , St Louis, MO 63130,","place":["United States"]},{"name":"AI for Health Institute, Washington University in St Louis , St Louis, MO 63130,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bing","family":"Xue","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Washington University in St Louis , St Louis, MO 63130,","place":["United States"]},{"name":"AI for Health Institute, Washington University in St Louis , St Louis, MO 63130,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4119-4836","authenticated-orcid":false,"given":"Thomas","family":"Kannampallil","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Washington University in St Louis , St Louis, MO 63130,","place":["United States"]},{"name":"AI for Health Institute, Washington University in St Louis , St Louis, MO 63130,","place":["United States"]},{"name":"Department of Anesthesiology, Washington University in St Louis , St Louis, MO 63110,","place":["United States"]},{"name":"Institute for Informatics, Data Science, and Biostatistics, Washington University in St Louis , St Louis, MO 63108,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chenyang","family":"Lu","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, Washington University in St Louis , St Louis, MO 63130,","place":["United States"]},{"name":"AI for Health Institute, Washington University in St Louis , St Louis, MO 63130,","place":["United States"]},{"name":"Department of Anesthesiology, Washington University in St Louis , St Louis, MO 63110,","place":["United States"]},{"name":"Institute for Informatics, Data Science, and Biostatistics, Washington University in St Louis , St Louis, MO 63108,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0235-1632","authenticated-orcid":false,"given":"Joanna","family":"Abraham","sequence":"additional","affiliation":[{"name":"AI for Health Institute, Washington University in St Louis , St Louis, MO 63130,","place":["United States"]},{"name":"Department of Anesthesiology, Washington University in St Louis , St Louis, MO 63110,","place":["United States"]},{"name":"Institute for Informatics, Data Science, and Biostatistics, Washington University in St Louis , St Louis, MO 63108,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2024,12,28]]},"reference":[{"key":"2025021811390412100_ocae316-B1","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1016\/j.athoracsur.2017.10.035","article-title":"The society of thoracic surgeons adult cardiac surgery database: 2018 update on outcomes and quality","volume":"105","author":"D'Agostino","year":"2018","journal-title":"Ann Thorac Surg"},{"key":"2025021811390412100_ocae316-B2","author":"Research iData","year":"2023"},{"key":"2025021811390412100_ocae316-B3","doi-asserted-by":"publisher","first-page":"500","DOI":"10.2215\/CJN.07830814","article-title":"AKI associated with cardiac surgery","volume":"10","author":"Thiele","year":"2015","journal-title":"Clin J Am Soc Nephrol"},{"key":"2025021811390412100_ocae316-B4","doi-asserted-by":"publisher","first-page":"1419","DOI":"10.1016\/j.athoracsur.2018.03.003","article-title":"The society of thoracic surgeons 2018 adult cardiac surgery risk models: part 2-statistical methods and results","volume":"105","author":"O'Brien","year":"2018","journal-title":"Ann Thorac Surg"},{"key":"2025021811390412100_ocae316-B5","doi-asserted-by":"publisher","first-page":"308","DOI":"10.1001\/jamasurg.2014.1700","article-title":"Changes over time in risk profiles of patients who undergo coronary artery bypass graft surgery: the veterans affairs surgical quality improvement program (VASQIP)","volume":"150","author":"Cornwell","year":"2015","journal-title":"JAMA Surg"},{"key":"2025021811390412100_ocae316-B6","doi-asserted-by":"publisher","first-page":"875","DOI":"10.1016\/j.jtcvs.2017.08.144","article-title":"Cost of individual complications following coronary artery bypass grafting","volume":"155","author":"Mehaffey","year":"2018","journal-title":"J Thorac Cardiovasc Surg"},{"key":"2025021811390412100_ocae316-B7","doi-asserted-by":"publisher","first-page":"697","DOI":"10.1038\/nrneph.2017.119","article-title":"Cardiac surgery-associated acute kidney injury: risk factors, pathophysiology and treatment","volume":"13","author":"Wang","year":"2017","journal-title":"Nat Rev Nephrol"},{"key":"2025021811390412100_ocae316-B8","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1097\/00000542-200007000-00023","article-title":"Analysis of risk factors for myocardial infarction and cardiac mortality after major vascular surgery","volume":"93","author":"Sprung","year":"2000","journal-title":"Anesthesiology"},{"key":"2025021811390412100_ocae316-B9","doi-asserted-by":"publisher","first-page":"2444","DOI":"10.1161\/CIRCULATIONAHA.108.800011","article-title":"Acute kidney injury is associated with increased long-term mortality after cardiothoracic surgery","volume":"119","author":"Hobson","year":"2009","journal-title":"Circulation"},{"key":"2025021811390412100_ocae316-B10","doi-asserted-by":"publisher","first-page":"e002204","DOI":"10.1161\/JAHA.115.002204","article-title":"Outcome reporting in cardiac surgery trials: systematic review and critical appraisal","volume":"4","author":"Goldfarb","year":"2015","journal-title":"J Am Heart Assoc"},{"key":"2025021811390412100_ocae316-B11","doi-asserted-by":"publisher","first-page":"e012447","DOI":"10.1161\/JAHA.119.012447","article-title":"Early versus delayed stroke after cardiac surgery: a systematic review and meta-analysis","volume":"8","author":"Gaudino","year":"2019","journal-title":"J Am Heart Assoc"},{"key":"2025021811390412100_ocae316-B12","doi-asserted-by":"publisher","first-page":"1770","DOI":"10.1016\/j.athoracsur.2021.03.043","article-title":"STS adult cardiac surgery database: 2021 update on outcomes, quality, and research","volume":"111","author":"Bowdish","year":"2021","journal-title":"Ann Thorac Surg"},{"issue":"3","key":"2025021811390412100_ocae316-B13","doi-asserted-by":"publisher","first-page":"1005","DOI":"10.1016\/j.athoracsur.2016.10.033","article-title":"The society of thoracic surgeons expert consensus for the resuscitation of patients who arrest after cardiac surgery","volume":"103","author":"Dunning","year":"2017","journal-title":"Ann Thorac Surg"},{"key":"2025021811390412100_ocae316-B14","doi-asserted-by":"publisher","first-page":"734","DOI":"10.1093\/ejcts\/ezs043","article-title":"EuroSCORE II","volume":"41","author":"Nashef","year":"2012","journal-title":"Eur J Cardiothorac Surg"},{"key":"2025021811390412100_ocae316-B15","doi-asserted-by":"publisher","first-page":"S2","DOI":"10.1016\/j.athoracsur.2009.05.053","article-title":"The society of thoracic surgeons 2008 cardiac surgery risk models: part 1\u2014coronary artery bypass grafting surgery","volume":"88","author":"Shahian","year":"2009","journal-title":"Ann Thorac Surg"},{"key":"2025021811390412100_ocae316-B16","doi-asserted-by":"publisher","first-page":"S23","DOI":"10.1016\/j.athoracsur.2009.05.056","article-title":"The society of thoracic surgeons 2008 cardiac surgery risk models: part 2\u2014isolated valve surgery","volume":"88","author":"O'Brien","year":"2009","journal-title":"Ann Thorac Surg"},{"key":"2025021811390412100_ocae316-B17","doi-asserted-by":"publisher","first-page":"S43","DOI":"10.1016\/j.athoracsur.2009.05.055","article-title":"The society of thoracic surgeons 2008 cardiac surgery risk models: part 3\u2014valve plus coronary artery bypass grafting surgery","volume":"88","author":"Shahian","year":"2009","journal-title":"Ann Thorac Surg"},{"key":"2025021811390412100_ocae316-B18","doi-asserted-by":"publisher","first-page":"1484","DOI":"10.1016\/j.athoracsur.2012.11.077","article-title":"The society of thoracic surgeons risk model for operative mortality after multiple valve surgery","volume":"95","author":"Rankin","year":"2013","journal-title":"Ann Thorac Surg"},{"key":"2025021811390412100_ocae316-B19","doi-asserted-by":"publisher","first-page":"e212240","DOI":"10.1001\/jamanetworkopen.2021.2240","article-title":"Use of machine learning to develop and evaluate models using preoperative and intraoperative data to identify risks of postoperative complications","volume":"4","author":"Xue","year":"2021","journal-title":"JAMA Netw Open"},{"key":"2025021811390412100_ocae316-B20","author":"Wang","year":"2024"},{"key":"2025021811390412100_ocae316-B21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.55524\/ijircst.2024.12.2.1","article-title":"Research on the application of semantic network in disease diagnosis prompts based on medical corpus","volume":"12","author":"Li","year":"2024","journal-title":"IJIRCST"},{"key":"2025021811390412100_ocae316-B22","doi-asserted-by":"publisher","first-page":"100","DOI":"10.1080\/00295639.2018.1499279","article-title":"Gaussian process-based inverse uncertainty quantification for TRACE physical model parameters using steady-state PSBT benchmark","volume":"193","author":"Wang","year":"2019","journal-title":"Nucl Sci Eng"},{"key":"2025021811390412100_ocae316-B23","doi-asserted-by":"publisher","first-page":"322","DOI":"10.3390\/jcm7100322","article-title":"Derivation and validation of machine learning approaches to predict acute kidney injury after cardiac surgery","volume":"7","author":"Lee","year":"2018","journal-title":"J Clin Med"},{"key":"2025021811390412100_ocae316-B24","doi-asserted-by":"publisher","first-page":"735","DOI":"10.1053\/j.semtcvs.2020.09.028","article-title":"Machine learning algorithms for predicting and risk profiling of cardiac surgery-associated acute kidney injury","volume":"33","author":"Penny-Dimri","year":"2021","journal-title":"Semin Thorac Cardiovasc Surg"},{"key":"2025021811390412100_ocae316-B25","doi-asserted-by":"publisher","first-page":"752","DOI":"10.1002\/clc.23377","article-title":"A novel machine learning algorithm, Bayesian networks model, to predict the high-risk patients with cardiac surgery-associated acute kidney injury","volume":"43","author":"Li","year":"2020","journal-title":"Clin Cardiol"},{"key":"2025021811390412100_ocae316-B26","doi-asserted-by":"publisher","first-page":"478","DOI":"10.1186\/s13054-020-03179-9","article-title":"Prediction of the development of acute kidney injury following cardiac surgery by machine learning","volume":"24","author":"Tseng","year":"2020","journal-title":"Crit Care"},{"key":"2025021811390412100_ocae316-B27","doi-asserted-by":"publisher","first-page":"1533","DOI":"10.1016\/j.athoracsur.2022.06.055","article-title":"Development of a machine learning model to predict outcomes and cost after cardiac surgery","volume":"115","author":"Zea-Vera","year":"2023","journal-title":"Ann Thorac Surg"},{"key":"2025021811390412100_ocae316-B28","doi-asserted-by":"publisher","first-page":"3384","DOI":"10.1038\/s41598-021-83020-7","article-title":"Machine learning prediction models for prognosis of critically ill patients after open-heart surgery","volume":"11","author":"Zhong","year":"2021","journal-title":"Sci Rep"},{"key":"2025021811390412100_ocae316-B29","first-page":"4268","author":"Xue","year":"2022"},{"key":"2025021811390412100_ocae316-B30","doi-asserted-by":"publisher","first-page":"18","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 Digit Med"},{"key":"2025021811390412100_ocae316-B31","doi-asserted-by":"publisher","first-page":"26094","DOI":"10.1038\/srep26094","article-title":"Deep patient: an unsupervised representation to predict the future of patients from the electronic health records","volume":"6","author":"Miotto","year":"2016","journal-title":"Sci Rep"},{"key":"2025021811390412100_ocae316-B32","author":"Kingma"},{"key":"2025021811390412100_ocae316-B33","doi-asserted-by":"publisher","first-page":"e020124","DOI":"10.1136\/bmjopen-2017-020124","article-title":"Using machine learning techniques to develop forecasting algorithms for postoperative complications: protocol for a retrospective study","volume":"8","author":"Fritz","year":"2018","journal-title":"BMJ Open"},{"key":"2025021811390412100_ocae316-B34","author":"Rong","year":"2014"},{"key":"2025021811390412100_ocae316-B35","doi-asserted-by":"publisher","first-page":"2","DOI":"10.1186\/s44158-022-00033-y","article-title":"Machine learning in perioperative medicine: a systematic review","volume":"2","author":"Bellini","year":"2022","journal-title":"J Anesth Analg Crit Care"},{"key":"2025021811390412100_ocae316-B36","first-page":"785","author":"Chen","year":"2016"},{"key":"2025021811390412100_ocae316-B37","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"2025021811390412100_ocae316-B38","volume-title":"Python 3 Reference Manual","author":"Van Rossum","year":"2009"},{"key":"2025021811390412100_ocae316-B39","first-page":"8024","article-title":"Pytorch: an imperative style, high-performance deep learning library","volume":"32","author":"Paszke","year":"2019","journal-title":"Adv Neural Inf Process Syst"},{"key":"2025021811390412100_ocae316-B40","first-page":"31:2615-2625","article-title":"Isolating sources of disentanglement in variational autoencoders","author":"Chen","year":"2018","journal-title":"Adv Neural Inf Process Syst"},{"key":"2025021811390412100_ocae316-B41","first-page":"2649","author":"Kim","year":"2018"},{"key":"2025021811390412100_ocae316-B42","first-page":"4080","article-title":"Prototypical networks for few-shot learning","volume":"30","author":"Snell","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"2025021811390412100_ocae316-B43","first-page":"1126","author":"Finn","year":"2017"},{"key":"2025021811390412100_ocae316-B44","author":". AAAI Press; 2019:5885-5892."},{"key":"2025021811390412100_ocae316-B45","first-page":"3319","author":"Sundararajan","year":"2017"},{"key":"2025021811390412100_ocae316-B46","first-page":"2579","article-title":"Visualizing data using t-SNE","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"J Mach Learn Res"},{"key":"2025021811390412100_ocae316-B47","author":"Brown"},{"key":"2025021811390412100_ocae316-B48","first-page":"4768","author":"Lundberg","year":"2017"},{"key":"2025021811390412100_ocae316-B49","doi-asserted-by":"publisher","first-page":"104270","DOI":"10.1016\/j.jbi.2022.104270","article-title":"Integrating machine learning predictions for perioperative risk management: towards an empirical design of a flexible-standardized risk assessment tool","volume":"137","author":"Abraham","year":"2023","journal-title":"J Biomed Inform"},{"key":"2025021811390412100_ocae316-B50","doi-asserted-by":"publisher","first-page":"104458","DOI":"10.1016\/j.ijmedinf.2021.104458","article-title":"An observational study of postoperative handoff standardization failures","volume":"151","author":"Abraham","year":"2021","journal-title":"Int J Med Inform"},{"key":"2025021811390412100_ocae316-B51","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1055\/s-0040-1721780","article-title":"Ascertaining design requirements for postoperative care transition interventions","volume":"12","author":"Abraham","year":"2021","journal-title":"Appl Clin Inform"},{"key":"2025021811390412100_ocae316-B52","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1136\/bmjqs-2020-012474","article-title":"Systematic review and meta-analysis of interventions for operating room to intensive care unit handoffs","volume":"30","author":"Abraham","year":"2021","journal-title":"BMJ Qual Saf"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/jamia\/advance-article-pdf\/doi\/10.1093\/jamia\/ocae316\/61289374\/ocae316.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jamia\/advance-article-pdf\/doi\/10.1093\/jamia\/ocae316\/61289374\/ocae316.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,18]],"date-time":"2025-02-18T11:39:21Z","timestamp":1739878761000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/32\/3\/459\/7934032"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,12,28]]},"references-count":52,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2024,12,28]]},"published-print":{"date-parts":[[2025,3,1]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocae316","relation":{},"ISSN":["1067-5027","1527-974X"],"issn-type":[{"value":"1067-5027","type":"print"},{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,3]]},"published":{"date-parts":[[2024,12,28]]}}}