{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,3]],"date-time":"2026-11-03T22:47:50Z","timestamp":1793746070889,"version":"4.6.1"},"reference-count":31,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T00:00:00Z","timestamp":1788134400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Cardiovasc. Med."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Machine learning (ML) models have been increasingly applied to surgical risk prediction; however, their comparative performance and temporal generalizability remain inadequately evaluated.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A retrospective cohort study of 9,956 adult patients undergoing coronary artery bypass grafting and\/or valve surgery was conducted using the MIMIC-IV v3.1 database. Logistic regression (LR), random forest (RF), and XGBoost models were developed without class weighting and tuned within nested cross-validation, trained on patients from 2008 to 2016 and temporally validated on patients from 2017 to 2019. Demographics, comorbidities, and 13 laboratory values available within 24\u2005h of admission served as predictors. A class-weighted variant with isotonic recalibration was examined as a sensitivity analysis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>In-hospital mortality was 2.1% in both the training and validation sets. In temporal validation, LR achieved the highest AUC (0.876), followed by RF (0.856) and XGBoost (0.831); the paired difference vs. XGBoost was +0.044 [95% CI (0.002, 0.084)], marginally excluding zero, whereas the difference vs. RF was not significant. Without class weighting, all three models were well calibrated (calibration intercepts near zero). Age, blood urea nitrogen, creatinine, and partial thromboplastin time were the most influential predictors.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Under a fair comparison with tuned, unweighted models, logistic regression achieved the highest numerical discrimination, but the advantage was small and at most marginally significant. Logistic regression provides an interpretable and well-calibrated model for early-admission mortality risk stratification, whereas these data do not establish that tree-based models are intrinsically less temporally stable.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fcvm.2026.1914633","type":"journal-article","created":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T04:28:05Z","timestamp":1788150485000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Predicting in-hospital mortality after cardiac surgery using machine learning: temporal validation in the MIMIC-IV database"],"prefix":"10.3389","volume":"13","author":[{"given":"Jianhui","family":"Zhou","sequence":"first","affiliation":[{"name":"Department of Cardiovascular Surgery, The First Affiliated Hospital of Anhui Medical University","place":["Hefei, Anhui, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chengxin","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Cardiovascular Surgery, The First Affiliated Hospital of Anhui Medical University","place":["Hefei, Anhui, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,31]]},"reference":[{"key":"B1","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":"B2","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":"B3","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"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1146","DOI":"10.1093\/ejcts\/ezs584","article-title":"Dynamic trends in cardiac surgery: why the logistic EuroSCORE is no longer suitable for contemporary cardiac surgery","volume":"43","author":"Hickey","year":"2013","journal-title":"Eur J Cardiothorac Surg"},{"key":"B5","doi-asserted-by":"publisher","first-page":"746","DOI":"10.1093\/ejcts\/ezr285","article-title":"Performance of the original EuroSCORE","volume":"41","author":"Siregar","year":"2012","journal-title":"Eur J Cardiothorac Surg"},{"key":"B6","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1093\/eurheartj\/ehs342","article-title":"Does EuroSCORE II perform better than its original versions? A multicentre validation study","volume":"34","author":"Barili","year":"2013","journal-title":"Eur Heart J"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1009","DOI":"10.1093\/ejcts\/ezt174","article-title":"The new EuroSCORE II does not improve prediction of mortality in high-risk patients","volume":"44","author":"Howell","year":"2013","journal-title":"Eur J Cardiothorac Surg"},{"key":"B8","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach Learn"},{"key":"B9","first-page":"785","article-title":"XGBoost: a scalable tree boosting system","author":"Chen","year":"2016"},{"key":"B10","doi-asserted-by":"publisher","first-page":"e0169772","DOI":"10.1371\/journal.pone.0169772","article-title":"A comparison of a machine learning model with EuroSCORE II in predicting mortality after elective cardiac surgery: a decision curve analysis","volume":"12","author":"Allyn","year":"2017","journal-title":"PLoS One"},{"key":"B11","doi-asserted-by":"publisher","first-page":"2075","DOI":"10.1016\/j.jtcvs.2020.07.105","article-title":"Machine learning improves mortality risk prediction after cardiac surgery","volume":"163","author":"Benedetto","year":"2022","journal-title":"J Thorac Cardiovasc Surg"},{"key":"B12","doi-asserted-by":"publisher","first-page":"1811","DOI":"10.1016\/j.athoracsur.2019.09.049","article-title":"Predictive utility of a machine learning algorithm in estimating mortality risk in cardiac surgery","volume":"109","author":"Kilic","year":"2020","journal-title":"Ann Thorac Surg"},{"key":"B13","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1016\/j.jclinepi.2019.02.004","article-title":"A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models","volume":"110","author":"Christodoulou","year":"2019","journal-title":"J Clin Epidemiol"},{"key":"B14","doi-asserted-by":"publisher","first-page":"95","DOI":"10.1016\/j.jclinepi.2020.03.005","article-title":"Machine learning algorithms performed no better than regression models for prognostication in traumatic brain injury","volume":"122","author":"Gravesteijn","year":"2020","journal-title":"J Clin Epidemiol"},{"key":"B15","doi-asserted-by":"publisher","first-page":"1052","DOI":"10.1093\/jamia\/ocx030","article-title":"Calibration drift in regression and machine learning models for acute kidney injury","volume":"24","author":"Davis","year":"2017","journal-title":"J Am Med Inform Assoc"},{"key":"B16","first-page":"381","article-title":"Feature robustness in non-stationary health records: caveats to deployable model performance in common clinical machine learning tasks","volume":"106","author":"Nestor","year":"2019","journal-title":"Proc Mach Learn Res"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1577","DOI":"10.1016\/S0140-6736(19)30037-6","article-title":"Reporting of artificial intelligence prediction models","volume":"393","author":"Collins","year":"2019","journal-title":"Lancet"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41597-022-01899-x","article-title":"MIMIC-IV, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","year":"2023","journal-title":"Sci Data"},{"key":"B19","doi-asserted-by":"publisher","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD\u2009+\u2009AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"Br Med J"},{"key":"B20","first-page":"4765","article-title":"A unified approach to interpreting model predictions","volume-title":"Advances in Neural Information Processing Systems","author":"Lundberg","year":"2017"},{"key":"B21","volume-title":"Python 3 Reference Manual","author":"Van Rossum","year":"2009"},{"key":"B22","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"B23","doi-asserted-by":"publisher","first-page":"128","DOI":"10.1097\/EDE.0b013e3181c30fb2","article-title":"Assessing the performance of prediction models: a framework for traditional and novel measures","volume":"21","author":"Steyerberg","year":"2010","journal-title":"Epidemiology"},{"key":"B24","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1016\/j.jclinepi.2022.01.011","article-title":"Population median imputation was noninferior to complex approaches for imputing missing values in cardiovascular prediction models in clinical practice","volume":"145","author":"Berkelmans","year":"2022","journal-title":"J Clin Epidemiol"},{"key":"B25","doi-asserted-by":"publisher","first-page":"162","DOI":"10.1681\/ASN.2004040331","article-title":"A clinical score to predict acute renal failure after cardiac surgery","volume":"16","author":"Thakar","year":"2005","journal-title":"J Am Soc Nephrol"},{"key":"B26","doi-asserted-by":"publisher","first-page":"1502","DOI":"10.1016\/j.surg.2011.07.056","article-title":"Albumin is a better predictor of outcomes than body mass index following coronary artery bypass grafting","volume":"91","author":"Bhamidipati","year":"2011","journal-title":"Ann Thorac Surg"},{"key":"B27","doi-asserted-by":"publisher","first-page":"762","DOI":"10.1016\/j.athoracsur.2010.09.004","article-title":"Hypoalbuminemia and long-term survival after coronary artery bypass","volume":"91","author":"de La Cruz","year":"2011","journal-title":"Ann Thorac Surg"},{"key":"B28","doi-asserted-by":"publisher","first-page":"866","DOI":"10.1016\/S0022-5223(99)70056-5","article-title":"Impact of body mass index and albumin on morbidity and mortality after cardiac surgery","volume":"118","author":"Engelman","year":"1999","journal-title":"J Thorac Cardiovasc Surg"},{"key":"B29","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1056\/NEJMc2104626","article-title":"The clinician and dataset shift in artificial intelligence","volume":"385","author":"Finlayson","year":"2021","journal-title":"N Engl J Med"},{"key":"B30","doi-asserted-by":"publisher","first-page":"1317","DOI":"10.1001\/jama.2017.18391","article-title":"Big data and machine learning in health care","volume":"319","author":"Beam","year":"2018","journal-title":"JAMA"},{"key":"B31","doi-asserted-by":"crossref","first-page":"595","DOI":"10.1175\/1520-0450(1973)012<0595:ANVPOT>2.0.CO;2","article-title":"A new vector partition of the probability score","volume":"12","author":"Murphy","year":"1973","journal-title":"J Appl Meteorol"}],"container-title":["Frontiers in Cardiovascular Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fcvm.2026.1914633\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T04:28:07Z","timestamp":1788150487000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fcvm.2026.1914633\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,31]]},"references-count":31,"alternative-id":["10.3389\/fcvm.2026.1914633"],"URL":"https:\/\/doi.org\/10.3389\/fcvm.2026.1914633","relation":{},"ISSN":["2297-055X"],"issn-type":[{"value":"2297-055X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,31]]},"article-number":"1914633"}}