{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T05:03:41Z","timestamp":1789621421420,"version":"build-2803163510"},"reference-count":39,"publisher":"MDPI AG","issue":"18","license":[{"start":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T00:00:00Z","timestamp":1789430400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"crossref","award":["RS-2022-NR070859"],"award-info":[{"award-number":["RS-2022-NR070859"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"crossref"}]},{"award":["RS-2022-NR070859"],"award-info":[{"award-number":["RS-2022-NR070859"]}],"id":[{"id":"https:\/\/ror.org\/013aysd81","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background: Identification of ICU patients at increased risk of death may support clinical assessment and prioritization of care. However, many mortality prediction approaches incorporate extensive laboratory information, conventional severity scores, or computationally complex models that may limit straightforward interpretation at the bedside. This study investigated whether routinely recorded physiological measurements collected during the first 24 h of ICU care could support an interpretable machine learning approach to subsequent in-hospital mortality prediction. Methods: We performed a retrospective analysis of 51,598 adult first ICU admissions from MIMIC-IV using a strict 24 h landmark design. Patients with a missing ICU duration, a documented ICU stay shorter than 24 h, or a documented death occurring on or before the 24 h landmark were excluded. Physiological measurements recorded during the first 24 h were summarized to construct the predictor set. Five classification algorithms\u2014Logistic Regression, Random Forest, XGBoost, LightGBM, and CatBoost\u2014were developed using a stratified 80:20 train\u2013test split. Performance was evaluated using discrimination, precision\u2013recall, calibration, threshold-dependent classification measures, and decision-curve analysis. Bootstrap resampling was used to estimate 95% confidence intervals and pairwise performance differences. Shapley Additive Explanations (SHAPs) were used to interpret the XGBoost model. Results: The revised cohort included 46,259 survivors and 5339 patients with an in-hospital mortality outcome. XGBoost achieved the numerically highest AUROC of 0.803 (95% CI: 0.790\u20130.815) and AUPRC of 0.341 (95% CI: 0.313\u20130.369), with a Brier score of 0.080 (95% CI: 0.076\u20130.084). Paired bootstrap comparisons showed that XGBoost had a significantly higher AUROC than Random Forest and LightGBM, but not CatBoost. Its AUPRC was significantly higher than that of Random Forest but did not differ significantly from LightGBM or CatBoost, while its Brier score was significantly lower than that of Random Forest but did not differ significantly from LightGBM or CatBoost. At the default probability threshold of 0.50, XGBoost achieved high specificity (0.990) but low recall (0.091), whereas lowering the threshold increased mortality detection at the cost of additional false-positive classifications. Conclusions: Routinely recorded physiological information accumulated during the first 24 h of ICU care supported an interpretable machine learning framework with useful discrimination and probability estimation for subsequent in-hospital mortality. XGBoost achieved the highest numerical performance, although differences among the gradient-boosting models were generally small and not significant for all evaluated metrics. Threshold-sensitivity analysis demonstrated that classification behavior depended strongly on the selected operating threshold. Independent multicenter validation and prospective clinical evaluation are required before routine deployment.<\/jats:p>","DOI":"10.3390\/diagnostics16182982","type":"journal-article","created":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T10:32:01Z","timestamp":1789468321000},"page":"2982","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Interpretable Machine Learning for In-Hospital Mortality Prediction in ICU Patients Using First-24-Hour Routine Vital Signs: A SHAP-Based MIMIC-IV Study"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2190-7210","authenticated-orcid":false,"given":"Abdul","family":"Karim","sequence":"first","affiliation":[{"name":"Cerebrovascular Disease Research Center, Hallym University, Chuncheon 24252, Republic of Korea"},{"name":"Department of Artificial Intelligence Convergence, Hallym University, Chuncheon 24252, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"methodology","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"visualization","vocabulary":"credit"},{"role":"conceptualization","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-1207-9828","authenticated-orcid":false,"given":"Jinwon","family":"Kim","sequence":"additional","affiliation":[{"name":"Cerebrovascular Disease Research Center, Hallym University, Chuncheon 24252, Republic of Korea"},{"name":"Department of Artificial Intelligence Convergence, Hallym University, Chuncheon 24252, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"validation","vocabulary":"credit"},{"role":"visualization","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9314-5601","authenticated-orcid":false,"given":"In cheol","family":"Jeong","sequence":"additional","affiliation":[{"name":"Cerebrovascular Disease Research Center, Hallym University, Chuncheon 24252, Republic of Korea"},{"name":"Department of Artificial Intelligence Convergence, Hallym University, Chuncheon 24252, Republic of Korea"},{"name":"Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,15]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Darraj, A., Hudays, A., Hazazi, A., Hobani, A., and Alghamdi, A. (2023). The association between emergency department overcrowding and delay in treatment: A systematic review. Healthcare, 11.","DOI":"10.3390\/healthcare11030385"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Sartini, M., Carbone, A., Demartini, A., Giribone, L., Oliva, M., Spagnolo, A.M., Cremonesi, P., Canale, F., and Cristina, M.L. (2022). Overcrowding in emergency department: Causes, consequences, and solutions\u2014A narrative review. Healthcare, 10.","DOI":"10.3390\/healthcare10091625"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Kim, Y.E., and Lee, H.Y. (2022). The effects of an emergency department length-of-stay management system on severely ill patients\u2019 treatment outcomes. BMC Emerg. Med., 22.","DOI":"10.1186\/s12873-022-00760-z"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Liu, Z., Shu, W., Liu, H., Zhang, X., and Chong, W. (2025). Development and validation of interpretable machine learning models for triage patients admitted to the intensive care unit. PLoS ONE, 20.","DOI":"10.1371\/journal.pone.0317819"},{"key":"ref_5","first-page":"e85667","article-title":"Clinical impact of artificial intelligence-based triage systems in emergency departments: A systematic review","volume":"17","author":"Abdalhalim","year":"2025","journal-title":"Cureus"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Ren, B., Zhang, Y., Chen, S., Dai, J., Chong, J., Zhong, Y., Deng, M., Jiang, S., and Chang, Z. (2025). Interpretable prediction of hospital mortality in bleeding critically ill patients based on machine learning and SHAP. BMC Med. Inform. Decis. Mak., 25.","DOI":"10.1186\/s12911-025-03101-9"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"104058","DOI":"10.1016\/j.iccn.2025.104058","article-title":"The role of AI in emergency department triage: An integrative systematic review","volume":"89","year":"2025","journal-title":"Intensive Crit. Care Nurs."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Hou, Y., Shi, Q., and Wang, C. (2026). Risk stratification and determinant identification of high-need, high-cost ICU patients using machine learning: A large-scale retrospective study from a multi-specialty ICU in a tertiary hospital. Front. Public Health, 14.","DOI":"10.3389\/fpubh.2026.1783334"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"406","DOI":"10.1097\/SHK.0000000000002800","article-title":"Multi-model fusion for 28-day mortality prediction in ICU patients: A comprehensive retrospective cohort study with subgroup validation and clinical risk stratification","volume":"66","author":"Long","year":"2026","journal-title":"Shock"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"21592","DOI":"10.1038\/s41598-025-05027-8","article-title":"Ensemble learning for biomedical signal classification: A high-accuracy framework using spectrograms from percussion and palpation","volume":"15","author":"Karim","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"67","DOI":"10.1007\/s00134-021-06546-4","article-title":"Variation in severity-adjusted resource use and outcome in intensive care units","volume":"48","author":"Takala","year":"2022","journal-title":"Intensive Care Med."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1186\/s40635-025-00810-3","article-title":"Development and validation of a machine learning model for early prediction of intensive care unit acquired weakness","volume":"13","author":"Nakano","year":"2025","journal-title":"Intensive Care Med. Exp."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"302","DOI":"10.1186\/s12967-025-06308-6","article-title":"Artificial intelligence-driven translational medicine: A machine learning framework for predicting disease outcomes and optimizing patient-centric care","volume":"23","author":"Abualigah","year":"2025","journal-title":"J. Transl. Med."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Van Poucke, S., Zhang, Z., Schmitz, M., Van Poucke, S., Vukicevic, M., Laenen, M.V., Celi, L.A., and De Deyne, C. (2016). Scalable predictive analysis in critically ill patients using a visual open data analysis platform. PLoS ONE, 11.","DOI":"10.1371\/journal.pone.0145791"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Li, H., Ashrafi, N., Kang, C., Zhao, G., Chen, Y., and Pishgar, M. (2024). A machine learning-based prediction of hospital mortality in mechanically ventilated ICU patients. PLoS ONE, 19.","DOI":"10.1101\/2024.07.12.24310325"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Moor, M., Rieck, B., Horn, M., Jutzeler, C.R., and Borgwardt, K. (2021). Early prediction of sepsis in the ICU using machine learning: A systematic review. Front. Med., 8.","DOI":"10.3389\/fmed.2021.607952"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1186\/s40708-024-00222-1","article-title":"Interpreting artificial intelligence models: A systematic review on the application of LIME and SHAP in Alzheimer\u2019s disease detection","volume":"11","author":"Vimbi","year":"2024","journal-title":"Brain Inform."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"106584","DOI":"10.1016\/j.cmpb.2021.106584","article-title":"Explanation of machine learning models using SHapley Additive exPlanations and application for real data in hospital","volume":"214","author":"Nohara","year":"2022","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_19","first-page":"345","article-title":"From development to deployment: Dataset shift, causality, and shift-stable models in health AI","volume":"21","author":"Subbaswamy","year":"2020","journal-title":"Biostatistics"},{"key":"ref_20","first-page":"8052","article-title":"Generalizing to unseen domains: A survey on domain generalization","volume":"35","author":"Wang","year":"2023","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"2474172","DOI":"10.1080\/07853890.2025.2474172","article-title":"Applying machine learning to predict bowel preparation adequacy in elderly patients for colonoscopy: Development and validation of a web-based prediction tool","volume":"57","author":"Liu","year":"2025","journal-title":"Ann. Med."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"105874","DOI":"10.1016\/j.ijmedinf.2025.105874","article-title":"Development and validation of an interpretable machine learning model for predicting in-hospital mortality for ischemic stroke patients in ICU","volume":"198","author":"Luo","year":"2025","journal-title":"Int. J. Med. Inform."},{"key":"ref_23","first-page":"313","article-title":"Triage and ongoing care for critically ill patients in the emergency department: Results from a national survey of emergency physicians","volume":"21","author":"Vu","year":"2020","journal-title":"West. J. Emerg. Med."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"774","DOI":"10.1007\/s12630-022-02231-2","article-title":"A framework for critical care triage during a major surge in critical illness","volume":"69","author":"Downar","year":"2022","journal-title":"Can. J. Anaesth."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"e00507","DOI":"10.14309\/ctg.0000000000000507","article-title":"Simplified machine learning models can accurately identify high-need, high-cost patients with inflammatory bowel disease","volume":"13","author":"Nguyen","year":"2022","journal-title":"Clin. Transl. Gastroenterol."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"e073485","DOI":"10.1136\/bmjopen-2023-073485","article-title":"Development and validation of a tool to predict high-need, high-cost patients hospitalised with ischaemic heart disease","volume":"13","author":"Moussa","year":"2023","journal-title":"BMJ Open"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"55","DOI":"10.7326\/M14-0697","article-title":"Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD): The TRIPOD statement","volume":"162","author":"Collins","year":"2015","journal-title":"Ann. Intern. Med."},{"key":"ref_28","doi-asserted-by":"crossref","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":"ref_29","doi-asserted-by":"crossref","first-page":"927","DOI":"10.3171\/2024.1.JNS232173","article-title":"Development and validation of machine learning models to predict postoperative infarction in moyamoya disease","volume":"141","author":"Fuse","year":"2024","journal-title":"J. Neurosurg."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Huang, Q., Jiang, Z., Shi, B., Meng, J., Shu, L., Hu, F., and Mi, J. (2025). Characterisation of cardiovascular disease incidence and machine learning risk prediction in middle-aged and elderly populations: Data from the China Health and Retirement Longitudinal Study (CHARLS). BMC Public Health, 25.","DOI":"10.1186\/s12889-025-21609-7"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"33096","DOI":"10.1038\/s41598-025-16790-z","article-title":"SMOTE-augmented machine learning model predicts recurrent and metastatic breast cancer from microbiome analysis","volume":"15","author":"Hong","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"6086","DOI":"10.1038\/s41598-024-56706-x","article-title":"Evaluation metrics and statistical tests for machine learning","volume":"14","author":"Rainio","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Aikodon, N., Ortega-Martorell, S., and Olier, I. (2024). Predicting decompensation risk in intensive care unit patients using machine learning. Algorithms, 17.","DOI":"10.3390\/a17010006"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"e893","DOI":"10.1016\/S2589-7500(22)00154-6","article-title":"Leveraging electronic health records for data science: Common pitfalls and how to avoid them","volume":"4","author":"Sauer","year":"2022","journal-title":"Lancet Digit. Health"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Ortega-Martorell, S., Olier, I., Ohlsson, M., and Lip, G.Y.H. (2024). Advancing personalised care in atrial fibrillation and stroke: The potential impact of AI from prevention to rehabilitation. Trends Cardiovasc. Med., Advance online publication.","DOI":"10.1016\/j.tcm.2024.12.003"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"707","DOI":"10.1007\/BF01709751","article-title":"The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction\/failure","volume":"22","author":"Vincent","year":"1996","journal-title":"Intensive Care Med."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"180178","DOI":"10.1038\/sdata.2018.178","article-title":"The eICU Collaborative Research Database, a freely available multi-center database for critical care research","volume":"5","author":"Pollard","year":"2018","journal-title":"Sci. Data"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"e563","DOI":"10.1097\/CCM.0000000000004916","article-title":"Sharing ICU patient data responsibly under the Society of Critical Care Medicine\/European Society of Intensive Care Medicine Joint Data Science Collaboration: The Amsterdam University Medical Centers Database (AmsterdamUMCdb) example","volume":"49","author":"Thoral","year":"2021","journal-title":"Crit. Care Med."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"557","DOI":"10.1177\/0885066616678394","article-title":"Complexity analysis of decision-making in the critically ill","volume":"33","author":"Daly","year":"2018","journal-title":"J. Intensive Care Med."}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/18\/2982\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T04:29:25Z","timestamp":1789619365000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/18\/2982"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,15]]},"references-count":39,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["diagnostics16182982"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16182982","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,15]]}}}