{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T12:52:00Z","timestamp":1791377520112,"version":"4.3.2"},"reference-count":48,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T00:00:00Z","timestamp":1790035200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100017962","name":"Jiangsu Commission of Health","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100017962","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Cell. Infect. Microbiol."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Sepsis is a prevalent and life-threatening complication in ICU patients after intracranial hematoma evacuation. This study aimed to develop and externally validate an interpretable machine learning model for predicting sepsis risk in this high-risk neurocritical care population.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A multicenter retrospective study was designed. The MIMIC-IV was based on for model development and internal testing; MIMIC-III, eICU and an independent clinical cohort from Jiangsu Provincial Hospital of Integrated Traditional Chinese and Western Medicine for external validation. Sepsis during ICU stay was defined according to the operationalized Sepsis-3 criteria. Early ICU variables were selected using Boruta and LASSO, and six machine learning algorithms were developed and compared. Model performance was evaluated by discrimination, calibration, and decision curve analysis, and its interpretability was assessed using SHAP and LIME. A sensitivity analysis was conducted by retraining all six models after excluding baseline SOFA.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      A total of 1,729 patients were included, of whom 717 developed sepsis during ICU stay. The final model retained 13 predictors, including WBC, SOFA score, SpO\n                      <jats:sub>2<\/jats:sub>\n                      , glucose, chloride, pneumonia, aspirin use, mannitol use, vasopressor use, mechanical ventilation, RBC, sedative use, and GCS score. XGBoost showed the most balanced overall performance, with AUCs of 0.794 in the internal test set and 0.820, 0.837, and 0.796 in the MIMIC-III, eICU, and independent clinical validation cohorts, respectively. SHAP analysis identified SOFA score, vasopressor use, glucose, pneumonia, and sedative exposure as major contributors to model predictions, while LIME supported individual-level interpretability. A web-based calculator was implemented for individualized risk prediction. Sensitivity analysis showed that XGBoost retained discrimination after SOFA exclusion, with no significant AUC changes across the external cohorts.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This study developed and externally validated an interpretable XGBoost-based model for predicting sepsis risk during ICU stay among patients after intracranial hematoma evacuation. The model showed generally consistent AUC-based discrimination across multiple validation cohorts and may support individualized risk stratification. Further prospective validation and clinical impact assessment are required before routine clinical implementation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fcimb.2026.1895258","type":"journal-article","created":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T04:26:16Z","timestamp":1790051176000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Interpretable machine learning for postoperative sepsis prediction in ICU patients following intracranial hematoma evacuation"],"prefix":"10.3389","volume":"16","author":[{"given":"Shaoyang","family":"Yu","sequence":"first","affiliation":[{"name":"The Second School of Clinical Medicine, Nanjing Medical University","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yun","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Affiliated Hospital of Integrated Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine","place":["Nanjing, Jiangsu, China"]},{"name":"Department of Neurosurgery, Jiangsu Province Academy of Traditional Chinese Medicine","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lei","family":"Guo","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Affiliated Hospital of Integrated Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine","place":["Nanjing, Jiangsu, China"]},{"name":"Department of Neurosurgery, Jiangsu Province Academy of Traditional Chinese Medicine","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinchang","family":"He","sequence":"additional","affiliation":[{"name":"The Second School of Clinical Medicine, Nanjing Medical University","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuyuan","family":"Ge","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Jiangsu Province Academy of Traditional Chinese Medicine","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wan","family":"Yu","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Jiangsu Province Academy of Traditional Chinese Medicine","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weijian","family":"Jia","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Jiangsu Province Academy of Traditional Chinese Medicine","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kexiang","family":"Xiong","sequence":"additional","affiliation":[{"name":"The Second School of Clinical Medicine, Nanjing Medical University","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zixiang","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Life Sciences, Xiamen University","place":["Xiamen, Fujian, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Gao","sequence":"additional","affiliation":[{"name":"Department of Critical Care Medicine, Jiangsu Province Academy of Traditional Chinese Medicine","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yong","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Neurosurgery, Jiangsu Province Academy of Traditional Chinese Medicine","place":["Nanjing, Jiangsu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,22]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1155\/2012\/629204","article-title":"Anemia and blood transfusions in critically ill patients","volume":"2012","author":"Athar","year":"2012","journal-title":"J. Blood Transfus."},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-026-03344-0","article-title":"Explainable AI for critical care: a systematic review of interpretable models for sepsis and ICU mortality prediction","volume":"26","author":"Athukorala","year":"2026","journal-title":"BMC Med. Inform. Decis. Mak"},{"key":"B3","doi-asserted-by":"publisher","first-page":"76","DOI":"10.1016\/j.jclinepi.2016.05.007","article-title":"Geographic and temporal validity of prediction models: different approaches were useful to examine model performance","volume":"79","author":"Austin","year":"2016","journal-title":"J. Clin. Epidemiol."},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"BMJ"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1136\/bmj-2023-074819","article-title":"Evaluation of clinical prediction models (part 1): from development to external validation","volume":"384","author":"Collins","year":"2024","journal-title":"BMJ"},{"key":"B6","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1038\/s41746-023-00832-9","article-title":"Perspectives on validation of clinical predictive algorithms","volume":"6","author":"de Hond","year":"2023","journal-title":"NPJ Digit Med."},{"key":"B7","doi-asserted-by":"publisher","first-page":"272","DOI":"10.1186\/s13054-016-1432-0","article-title":"The critical care management of spontaneous intracranial hemorrhage: a contemporary review","volume":"20","author":"de Oliveira Manoel","year":"2016","journal-title":"Crit. Care"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1136\/bmj-2023-078276","article-title":"Developing clinical prediction models: a step-by-step guide","volume":"386","author":"Efthimiou","year":"2024","journal-title":"BMJ"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1181","DOI":"10.1007\/s00134-021-06506-y","article-title":"Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021","volume":"47","author":"Evans","year":"2021","journal-title":"Intensive Care Med."},{"key":"B10","doi-asserted-by":"publisher","first-page":"127","DOI":"10.1186\/s12974-021-02177-0","article-title":"Stroke-induced immunosuppression: implications for the prevention and prediction of post-stroke infections","volume":"18","author":"Faura","year":"2021","journal-title":"J. Neuroinflamm."},{"key":"B11","doi-asserted-by":"publisher","first-page":"973","DOI":"10.1016\/S1474-4422(24)00369-7","article-title":"Global, regional, and national burden of stroke and its risk factors, 1990\u20132021: a systematic analysis for the Global Burden of Disease Study 2021","volume":"23","author":"Feigin","year":"2024","journal-title":"Lancet Neurol."},{"key":"B12","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1186\/s13613-018-0388-4","article-title":"Hyperchloraemia in sepsis","volume":"8","author":"Filis","year":"2018","journal-title":"Ann. Intensive Care"},{"key":"B13","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1016\/S1474-4422(18)30415-0","article-title":"Global, regional, and national burden of traumatic brain injury and spinal cord injury, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016","volume":"18","year":"2019","journal-title":"Lancet Neurol."},{"key":"B14","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1016\/j.cell.2020.03.022","article-title":"How machine learning will transform biomedicine","volume":"181","author":"Goecks","year":"2020","journal-title":"Cell."},{"key":"B15","doi-asserted-by":"publisher","first-page":"E215","DOI":"10.1161\/01.cir.101.23.e215","article-title":"PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals","volume":"101","author":"Goldberger","year":"2000","journal-title":"Circulation"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.3390\/s25216649","article-title":"eXplainable artificial intelligence (XAI): a systematic review for unveiling the black box models and their relevance to biomedical imaging and sensing","volume":"25","author":"Hettikankanamage","year":"2025","journal-title":"Sensors (Basel)"},{"key":"B17","doi-asserted-by":"publisher","first-page":"6447","DOI":"10.1038\/s41467-024-50805-z","article-title":"Ventilator-associated pneumonia: pathobiological heterogeneity and diagnostic challenges","volume":"15","author":"Howroyd","year":"2024","journal-title":"Nat. Commun."},{"key":"B18","doi-asserted-by":"publisher","first-page":"110","DOI":"10.1213\/ANE.0000000000005943","article-title":"Association between aspirin use and sepsis outcomes: a national cohort study","volume":"135","author":"Hsu","year":"2022","journal-title":"Anesth. Analg."},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.3390\/jcm12175658","article-title":"Machine learning-based early prediction of sepsis using electronic health records: a systematic review","volume":"12","author":"Islam","year":"2023","journal-title":"J. Clin. Med."},{"key":"B20","doi-asserted-by":"publisher","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":"B21","doi-asserted-by":"publisher","first-page":"1082","DOI":"10.1213\/ANE.0000000000005912","article-title":"Severe intraoperative hyperglycemia and infectious complications after elective brain neurosurgical procedures: prospective observational study","volume":"135","author":"Kulikov","year":"2022","journal-title":"Anesth. Analg."},{"key":"B22","doi-asserted-by":"publisher","first-page":"456","DOI":"10.3390\/epidemiologia5030032","article-title":"The global burden of sepsis and septic shock","volume":"5","author":"La Via","year":"2024","journal-title":"Epidemiologia (Basel)"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1159\/000547563","article-title":"The burden of traumatic brain injury,uture trend predictions in 204 countries and territories (1990\u20132021): results from the Global Burden of Disease Study 2021","volume":"60","author":"Liu","year":"2026","journal-title":"Neuroepidemiol."},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.2196\/71413","article-title":"Identification and validation of an explainable prediction model of sepsis in patients with intracerebral hemorrhage: multicenter retrospective study","volume":"27","author":"Liu","year":"2025","journal-title":"J. Med. Internet Res."},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0313132","article-title":"Explainable machine learning for early prediction of sepsis in traumatic brain injury: a discovery and validation study","volume":"19","author":"Liu","year":"2024","journal-title":"PloS One"},{"key":"B26","doi-asserted-by":"publisher","first-page":"633","DOI":"10.3949\/ccjm.87a.19117","article-title":"Hospital-acquired and ventilator-associated pneumonia: diagnosis, management, and prevention","volume":"87","author":"Modi","year":"2020","journal-title":"Cleve Clin. J. Med."},{"key":"B27","doi-asserted-by":"publisher","first-page":"1971","DOI":"10.1007\/s00134-025-08137-z","article-title":"ESICM guidelines on circulatory shock and hemodynamic monitoring 2025","volume":"51","author":"Monnet","year":"2025","journal-title":"Intensive Care Med."},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1136\/bmj-2024-082505","article-title":"PROBAST+AI: an updated quality, risk of bias, and applicability assessment tool for prediction models using regression or artificial intelligence methods","volume":"388","author":"Moons","year":"2025","journal-title":"BMJ"},{"key":"B29","doi-asserted-by":"publisher","first-page":"399","DOI":"10.1007\/s10877-023-01063-9","article-title":"Autonomic dysfunction as a predictor of infection in neurocritical care unit: a prospective cohort study","volume":"38","author":"Naik","year":"2024","journal-title":"J. Clin. Monit. Comput."},{"key":"B30","doi-asserted-by":"publisher","first-page":"R30","DOI":"10.1186\/cc8907","article-title":"Intensive care unit-acquired infection as a side effect of sedation","volume":"14","author":"Nseir","year":"2010","journal-title":"Crit. Care"},{"key":"B31","doi-asserted-by":"publisher","first-page":"113","DOI":"10.1186\/s13054-024-04860-z","article-title":"Use of artificial intelligence in critical care: opportunities and obstacles","volume":"28","author":"Pinsky","year":"2024","journal-title":"Crit. Care"},{"key":"B32","doi-asserted-by":"publisher","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":"B33","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.i3140","article-title":"External validation of clinical prediction models using big datasets from e-health records or IPD meta-analysis: opportunities and challenges","volume":"353","author":"Riley","year":"2016","journal-title":"BMJ"},{"key":"B34","doi-asserted-by":"publisher","first-page":"8761","DOI":"10.1021\/acs.jmedchem.9b01101","article-title":"Interpretation of compound activity predictions from complex machine learning models using local approximations and Shapley values","volume":"63","author":"Rodriguez-Perez","year":"2020","journal-title":"J. Med. Chem."},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2023.29729","article-title":"Sepsis prediction model for determining sepsis vs SIRS, qSOFA, and SOFA","volume":"6","author":"Schertz","year":"2023","journal-title":"JAMA Netw. Open"},{"key":"B36","doi-asserted-by":"publisher","first-page":"2235","DOI":"10.1056\/NEJMoa1703058","article-title":"Time to treatment and mortality during mandated emergency care for sepsis","volume":"376","author":"Seymour","year":"2017","journal-title":"N. Engl. J. Med."},{"key":"B37","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1136\/svn-2017-000123","article-title":"Stroke-induced immunosuppression and poststroke infection","volume":"3","author":"Shi","year":"2018","journal-title":"Stroke Vasc. Neurol."},{"key":"B38","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1038\/nrneurol.2017.13","article-title":"The far-reaching scope of neuroinflammation after traumatic brain injury","volume":"13","author":"Simon","year":"2017","journal-title":"Nat. Rev. Neurol."},{"key":"B39","doi-asserted-by":"publisher","first-page":"801","DOI":"10.1001\/jama.2016.0287","article-title":"The third international consensus definitions for sepsis and septic shock (Sepsis-3)","volume":"315","author":"Singer","year":"2016","journal-title":"JAMA"},{"key":"B40","doi-asserted-by":"publisher","first-page":"1624","DOI":"10.1097\/01.ccm.0000170106.61978.d8","article-title":"Acute hyperglycemia and the innate immune system: clinical, cellular, and molecular aspects","volume":"33","author":"Turina","year":"2005","journal-title":"Crit. Care Med."},{"key":"B41","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1227\/01.NEU.0000255497.26440.01","article-title":"Surgery of cerebral trauma and associated critical care","volume":"61","author":"Valadka","year":"2007","journal-title":"Neurosurgery"},{"key":"B42","doi-asserted-by":"publisher","first-page":"380","DOI":"10.1016\/S2213-2600(14)70061-X","article-title":"Assessment of the worldwide burden of critical illness: the intensive care over nations (ICON) audit","volume":"2","author":"Vincent","year":"2014","journal-title":"Lancet Respir. Med."},{"key":"B43","doi-asserted-by":"publisher","DOI":"10.2196\/15791","article-title":"Improving clinical translation of machine learning approaches through clinician-tailored visual displays of black box algorithms: development and validation","volume":"8","author":"Wongvibulsin","year":"2020","journal-title":"JMIR Med. Inform."},{"key":"B44","doi-asserted-by":"publisher","first-page":"24765","DOI":"10.1038\/s41598-025-10119-6","article-title":"A predictive model for sepsis risk in patients with non-traumatic cerebral hemorrhage based on the MIMIC-IV database","volume":"15","author":"Wu","year":"2025","journal-title":"Sci. Rep."},{"key":"B45","doi-asserted-by":"publisher","DOI":"10.1016\/j.phrs.2025.107917","article-title":"Organ-targeted biomarkers of sepsis: a systematic review reveals the value of inflammation and lipid metabolic dysregulation","volume":"219","author":"Xu","year":"2025","journal-title":"Pharmacol. Res."},{"key":"B46","doi-asserted-by":"publisher","first-page":"399","DOI":"10.1186\/s12911-025-03240-z","article-title":"Development and validation of a machine-learning model for predicting prognosis in critically ill patients undergoing major surgery","volume":"25","author":"Zhang","year":"2025","journal-title":"BMC Med. Inform. Decis. Mak"},{"key":"B47","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijmedinf.2025.106169","article-title":"Early prediction of septic shock in ICU patients using machine learning: development, external validation, and explainability with SHAP","volume":"206","author":"Zhou","year":"2026","journal-title":"Int. J. Med. Inform."},{"key":"B48","doi-asserted-by":"publisher","DOI":"10.3389\/fpubh.2022.895991","article-title":"Incidence, risk factors and outcomes of sepsis in critically ill post-craniotomy patients: a single-center prospective cohort study","volume":"10","author":"Zhou","year":"2022","journal-title":"Front. Public Health"}],"container-title":["Frontiers in Cellular and Infection Microbiology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fcimb.2026.1895258\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T11:54:12Z","timestamp":1791374052000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fcimb.2026.1895258\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,22]]},"references-count":48,"alternative-id":["10.3389\/fcimb.2026.1895258"],"URL":"https:\/\/doi.org\/10.3389\/fcimb.2026.1895258","relation":{},"ISSN":["2235-2988"],"issn-type":[{"value":"2235-2988","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,22]]},"article-number":"1895258"}}