{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T02:04:09Z","timestamp":1789524249964,"version":"build-2803163510"},"reference-count":24,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,6,27]],"date-time":"2026-06-27T00:00:00Z","timestamp":1782518400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T00:00:00Z","timestamp":1789516800000},"content-version":"vor","delay-in-days":81,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Inform Decis Mak"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Prolonged postoperative intensive care unit (ICU) length of stay (LOS) after coronary artery bypass grafting (CABG) drives resource use yet remains difficult to anticipate at the 24-hour ICU evaluation moment when extended monitoring beyond 72\u00a0h is first considered. We developed and externally validated an interpretable machine-learning decision-support calculator for this deployment moment.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      Adult CABG patients were identified from MIMIC-IV 3.1 (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u20096,919; 7:3 stratified split for development) and eICU-CRD 2.0 (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u20095,972; external validation). The outcome was prolonged ICU LOS (&gt;\u20093 days). Elastic Net plus Boruta selected eight bedside features collected within the first 24\u00a0h of ICU admission: 24-hour fluid intake, Charlson Comorbidity Index (CCI), Sequential Organ Failure Assessment (SOFA) score, Simplified Acute Physiology Score II (SAPS-II), Glasgow Coma Scale (GCS), vasopressor use, congestive heart failure, and atrial fibrillation. Nine machine-learning algorithms were compared by 10-fold cross-validation paired t-tests with Bonferroni and Benjamini\u2013Hochberg correction. Calibration metrics included Hosmer\u2013Lemeshow test, Integrated Calibration Index (ICI), expected-to-observed ratio, intercept, and slope (full Methods). SHapley Additive exPlanations (SHAP) provided per-patient feature attribution.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      CatBoost was selected as the deployed model. On the MIMIC-IV internal test set (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u20092,076), the area under the receiver-operating-characteristic curve (AUC) was 0.7739 (95% confidence interval 0.7379\u20130.8099), Hosmer\u2013Lemeshow\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.224, calibration slope 0.973. On the eICU-CRD external cohort, AUC was 0.6452 (95% CI 0.6311\u20130.6602), calibration slope 0.998, ICI 0.023. At the prevalence-anchored threshold of 0.30, sensitivity was 0.55, specificity 0.65, positive predictive value 0.40, and negative predictive value 0.77. Decision Curve Analysis showed positive net benefit over treat-all and treat-none across t\u2009=\u20090.20\u20130.40. SHAP top-3 features were 24-hour fluid intake, CCI, and atrial fibrillation.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The model provides decision support at the post-CABG 24-hour ICU evaluation moment with modest discrimination and well-calibrated probabilities at deployment. The deployed online calculator renders per-patient feature contributions transparent at the bedside via SHAP; prospective validation is required before clinical deployment.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s12911-026-03659-y","type":"journal-article","created":{"date-parts":[[2026,6,27]],"date-time":"2026-06-27T12:15:54Z","timestamp":1782562554000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Development and external validation of an interpretable machine learning model for predicting prolonged postoperative ICU length of stay in coronary artery bypass grafting patients using MIMIC-IV 3.1 and eICU-CRD 2.0"],"prefix":"10.1186","volume":"26","author":[{"given":"Dayan","family":"Liu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-8930-0791","authenticated-orcid":false,"given":"Pengyu","family":"Lu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yulan","family":"Meng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xianglong","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Huang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,6,27]]},"reference":[{"key":"3659_CR1","doi-asserted-by":"crossref","unstructured":"Baris O, et al. Assessing the predictive impact of preoperative lactate dehydrogenase to albumin ratio on outcomes following coronary artery bypass graft surgery. J Clin Med. 2025;14(2).","DOI":"10.3390\/jcm14020554"},{"issue":"1","key":"3659_CR2","doi-asserted-by":"publisher","first-page":"57","DOI":"10.21470\/1678-9741-2019-0405","volume":"36","author":"M Zarrizi","year":"2021","unstructured":"Zarrizi M, et al. Predictors of length of stay in intensive care unit after coronary artery bypass grafting: development a risk scoring system. Braz J Cardiovasc Surg. 2021;36(1):57\u201363.","journal-title":"Braz J Cardiovasc Surg"},{"key":"3659_CR3","doi-asserted-by":"crossref","unstructured":"Huang YC, et al. Machine-learning techniques for feature selection and prediction of mortality in elderly CABG patients. Healthc (Basel). 2021;9(5).","DOI":"10.3390\/healthcare9050547"},{"key":"3659_CR4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.atherosclerosis.2023.01.003","volume":"367","author":"X Feng","year":"2023","unstructured":"Feng X, et al. Machine learning improves mortality prediction in three-vessel disease. Atherosclerosis. 2023;367:1\u20137.","journal-title":"Atherosclerosis"},{"issue":"1","key":"3659_CR5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41597-022-01899-x","volume":"10","author":"AEW Johnson","year":"2023","unstructured":"Johnson AEW, et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci Data. 2023;10(1):1.","journal-title":"Sci Data"},{"key":"3659_CR6","doi-asserted-by":"publisher","first-page":"180178","DOI":"10.1038\/sdata.2018.178","volume":"5","author":"TJ Pollard","year":"2018","unstructured":"Pollard TJ, et al. The eICU Collaborative Research Database, a freely available multi-center database for critical care research. Sci Data. 2018;5:180178.","journal-title":"Sci Data"},{"key":"3659_CR7","doi-asserted-by":"publisher","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","volume":"385","author":"GS Collins","year":"2024","unstructured":"Collins GS, et al. TRIPOD\u2009+\u2009AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ. 2024;385:e078378.","journal-title":"BMJ"},{"issue":"6973","key":"3659_CR8","doi-asserted-by":"publisher","first-page":"170","DOI":"10.1136\/bmj.310.6973.170","volume":"310","author":"JM Bland","year":"1995","unstructured":"Bland JM, Altman DG. Multiple significance tests: the Bonferroni method. BMJ. 1995;310(6973):170.","journal-title":"BMJ"},{"issue":"1","key":"3659_CR9","doi-asserted-by":"publisher","first-page":"289","DOI":"10.1111\/j.2517-6161.1995.tb02031.x","volume":"57","author":"Y Benjamini","year":"1995","unstructured":"Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B. 1995;57(1):289\u2013300.","journal-title":"J R Stat Soc B"},{"key":"3659_CR10","doi-asserted-by":"publisher","first-page":"3979630","DOI":"10.1155\/2020\/3979630","volume":"2020","author":"R Zahara","year":"2020","unstructured":"Zahara R, Santoso A, Barano AZ. Myocardial Fluid Balance and Pathophysiology of Myocardial Edema in Coronary Artery Bypass Grafting. Cardiol Res Pract. 2020;2020:3979630.","journal-title":"Cardiol Res Pract"},{"issue":"1","key":"3659_CR11","doi-asserted-by":"publisher","first-page":"162","DOI":"10.1186\/s13019-022-01906-7","volume":"17","author":"A Khalili","year":"2022","unstructured":"Khalili A, et al. In-hospital mortality of patients with severe left ventricular dysfunction undergoing coronary artery bypass grafting in Iranian population. J Cardiothorac Surg. 2022;17(1):162.","journal-title":"J Cardiothorac Surg"},{"issue":"6","key":"3659_CR12","doi-asserted-by":"publisher","first-page":"1109","DOI":"10.1007\/s00467-020-04489-4","volume":"35","author":"I Bellos","year":"2020","unstructured":"Bellos I, Iliopoulos DC, Perrea DN. Association of postoperative fluid overload with adverse outcomes after congenital heart surgery: a systematic review and dose-response meta-analysis. Pediatr Nephrol. 2020;35(6):1109\u201319.","journal-title":"Pediatr Nephrol"},{"issue":"2","key":"3659_CR13","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1053\/j.semtcvs.2022.07.002","volume":"35","author":"DP Fudulu","year":"2023","unstructured":"Fudulu DP, et al. Impact of Preoperative Atrial Fibrillation on In-Hospital Outcomes of Coronary Artery Bypass Grafting. Semin Thorac Cardiovasc Surg. 2023;35(2):261\u20136.","journal-title":"Semin Thorac Cardiovasc Surg"},{"key":"3659_CR14","doi-asserted-by":"crossref","unstructured":"Damarlapally N, et al. High vasoactive-inotropic score is associated with poor outcomes in patients undergoing coronary artery bypass grafting: a systematic review and meta-analysis. Perfusion. 2025:2676591251364482.","DOI":"10.1177\/02676591251364482"},{"key":"3659_CR15","doi-asserted-by":"publisher","first-page":"774935","DOI":"10.3389\/fcvm.2021.774935","volume":"8","author":"F Xu","year":"2021","unstructured":"Xu F, et al. Performance of Sequential Organ Failure Assessment and Simplified Acute Physiology Score II for Post-Cardiac Surgery Patients in Intensive Care Unit. Front Cardiovasc Med. 2021;8:774935.","journal-title":"Front Cardiovasc Med"},{"issue":"1","key":"3659_CR16","doi-asserted-by":"publisher","first-page":"374","DOI":"10.1186\/s13054-019-2663-7","volume":"23","author":"S Lambden","year":"2019","unstructured":"Lambden S, et al. The SOFA score-development, utility and challenges of accurate assessment in clinical trials. Crit Care. 2019;23(1):374.","journal-title":"Crit Care"},{"issue":"1","key":"3659_CR17","doi-asserted-by":"publisher","first-page":"76","DOI":"10.1186\/s43044-022-00315-4","volume":"74","author":"M Laimoud","year":"2022","unstructured":"Laimoud M, et al. Predictors and clinical outcomes of post-coronary artery bypass grafting cerebrovascular strokes. Egypt Heart J. 2022;74(1):76.","journal-title":"Egypt Heart J"},{"issue":"1","key":"3659_CR18","doi-asserted-by":"publisher","first-page":"8","DOI":"10.1159\/000521288","volume":"91","author":"ME Charlson","year":"2022","unstructured":"Charlson ME, et al. Charlson Comorbidity Index: A Critical Review of Clinimetric Properties. Psychother Psychosom. 2022;91(1):8\u201335.","journal-title":"Psychother Psychosom"},{"key":"3659_CR19","doi-asserted-by":"crossref","unstructured":"Baris O, et al. The efficacy of the charlson comorbidity index and its age-adjusted version in forecasting mortality and postoperative outcomes following isolated coronary artery bypass grafting. J Clin Med. 2025;14(2).","DOI":"10.3390\/jcm14020395"},{"issue":"1 Suppl","key":"3659_CR20","doi-asserted-by":"publisher","first-page":"S2","DOI":"10.1016\/j.athoracsur.2009.05.053","volume":"88","author":"DM Shahian","year":"2009","unstructured":"Shahian DM, et al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 1\u2013coronary artery bypass grafting surgery. Ann Thorac Surg. 2009;88(1 Suppl):S2\u201322.","journal-title":"Ann Thorac Surg"},{"issue":"3","key":"3659_CR21","doi-asserted-by":"publisher","first-page":"784","DOI":"10.1016\/j.jtcvs.2024.11.009","volume":"170","author":"AM Wisniewski","year":"2025","unstructured":"Wisniewski AM, et al. Multi-institutional model to predict intensive care unit length of stay after cardiac surgery. J Thorac Cardiovasc Surg. 2025;170(3):784\u201392.","journal-title":"J Thorac Cardiovasc Surg"},{"issue":"11","key":"3659_CR22","doi-asserted-by":"publisher","first-page":"2951","DOI":"10.1053\/j.jvca.2020.06.015","volume":"34","author":"C Dominici","year":"2020","unstructured":"Dominici C, et al. A nomogram for predicting long length of stay in the intensive care unit in patients undergoing CABG: results from the multicenter E-CABG registry. J Cardiothorac Vasc Anesth. 2020;34(11):2951\u201361.","journal-title":"J Cardiothorac Vasc Anesth"},{"issue":"21","key":"3659_CR23","doi-asserted-by":"publisher","first-page":"e017847","DOI":"10.1161\/JAHA.120.017847","volume":"9","author":"LY Sun","year":"2020","unstructured":"Sun LY, et al. Derivation and validation of a clinical model to predict intensive care unit length of stay after cardiac surgery. J Am Heart Assoc. 2020;9(21):e017847.","journal-title":"J Am Heart Assoc"},{"issue":"14","key":"3659_CR24","doi-asserted-by":"publisher","first-page":"e48224","DOI":"10.1097\/MD.0000000000048224","volume":"105","author":"P Zhang","year":"2026","unstructured":"Zhang P, et al. Development of a prediction model for the risk of prolonged ICU length of stay in patients undergoing coronary artery bypass surgery: a retrospective analysis based on the MIMIC-IV database. Med (Baltim). 2026;105(14):e48224.","journal-title":"Med (Baltim)"}],"container-title":["BMC Medical Informatics and Decision Making"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12911-026-03659-y","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-026-03659-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-026-03659-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T01:58:58Z","timestamp":1789523938000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s12911-026-03659-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,27]]},"references-count":24,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["3659"],"URL":"https:\/\/doi.org\/10.1186\/s12911-026-03659-y","relation":{},"ISSN":["1472-6947"],"issn-type":[{"value":"1472-6947","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,27]]},"assertion":[{"value":"2 February 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"22 June 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"27 June 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"This study was conducted using de-identified data from publicly available databases (MIMIC-IV 3.1 and eICU-CRD 2.0). All data were de-identified in accordance with the Health Insurance Portability and Accountability Act (HIPAA) standards and institutional privacy protocols. This retrospective analysis of de-identified data was exempt from institutional review board approval. The study was conducted in accordance with the Declaration of Helsinki and all applicable ethical guidelines for research with secondary data sources.","order":1,"name":"Ethics","label":"Ethics approval and consent to participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable. This study used only de-identified data from public databases; no individual-level identifying information is presented in the manuscript.","order":2,"name":"Ethics","label":"Consent for publication","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":3,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"353"}}