Integrating Inflammatory and Endothelial Biomarkers in Machine Learning Models to Predict Sepsis Development in Critically Ill Patients: A Combined Retrospective and Prospective Multicenter Study
Pathogens · 18.11.2026 · 15 (11) · 1214
DOI: 10.3390/pathogens15111214 ↗Kurz gefasst
Fachartikel in der Zeitschrift Pathogens, online erschienen am 18.11.2026. Verfasst von Ioannis Papapanagiotou und 22 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European Union Recovery and Resilience Facility (RRF), Greece 2.0.
Abstract (Originalsprache)
Sepsis is a severe, life-threatening condition in which timely recognition improves patient outcomes. Machine learning (ML) methods, usually relying on vital sign data, have shown promise in early sepsis prediction; however, they underrepresent sepsis biomarkers, which are widely used in clinical practice. We aimed to develop and validate interpretable ML models for early sepsis prediction using non-temporal data from the first 24 h of intensive care unit (ICU) admission, and to quantify the incremental value of inflammatory (Interleukin-6 (IL-6), procalcitonin (PCT), C-Reactive Protein (CRP)) and endothelial (soluble Endothelial Protein C Receptor (sEPCR)) biomarkers. A combined cohort of 1257 ICU patients from retrospective and prospective multicenter datasets was analyzed. We evaluated four tree-based models alongside Logistic Regression. We assessed discrimination (Area Under the Receiver Operating Characteristic Curve (AUC); Area Under the Precision-Recall Curve (AUPRC)), calibration, clinical utility (Decision Curve Analysis), and used structured feature ablation and Shapley additive explanations (SHAP) for interpretation. Light Gradient Boosting Machine (LGBM) achieved the best performance on the combined cohort, with an AUC of 0.86 and an AUPRC of 0.62. Moreover, LGBM achieved an AUC of 0.728 and an AUPRC of 0.288 during prospective multicenter validation from the retrospective to the prospective data. Feature ablation demonstrated that inclusion of IL-6, PCT, CRP, and sEPCR significantly improved model performance (AUC: p = 0.003; AUPRC: p = 0.001). SHAP analysis identified IL-6 and PCT as among the most influential predictors, both showing threshold-like nonlinear effects. These findings highlight the critical added value of biologically relevant biomarkers in ML-based clinical decision support.
Förderung
- European Union Recovery and Resilience Facility (RRF), Greece 2.0
Zitieren
APA
Papapanagiotou, I., Vassiliou, A. G., Vrettou, C. S., Theodorakopoulou, M., Mastora, Z., Karalis, A., Giannopoulou, V., Kampouropoulou, O., Charcharos, I., Poupouzas, G., Issaris, V., Dede, K., Christopoulou, H., Karkoulias, K., Raftopoulou, S., Pouriki, S., Keskinidou, C., Lotsios, N. S., Ganitidis, T., …, & Dimopoulou, I. (2026). Integrating Inflammatory and Endothelial Biomarkers in Machine Learning Models to Predict Sepsis Development in Critically Ill Patients: A Combined Retrospective and Prospective Multicenter Study. Pathogens, 15(11), 1214. https://doi.org/10.3390/pathogens15111214
Vancouver
Papapanagiotou I, Vassiliou AG, Vrettou CS, Theodorakopoulou M, Mastora Z, Karalis A, et al. Integrating Inflammatory and Endothelial Biomarkers in Machine Learning Models to Predict Sepsis Development in Critically Ill Patients: A Combined Retrospective and Prospective Multicenter Study. Pathogens. 2026;15(11):1214. doi:10.3390/pathogens15111214
BibTeX
@article{Papapanagiotou2026,
author = {Papapanagiotou, Ioannis and Vassiliou, Alice G. and Vrettou, Charikleia S. and Theodorakopoulou, Maria and Mastora, Zafeiria and Karalis, Apostolos and Giannopoulou, Vassiliki and Kampouropoulou, Olga and Charcharos, Ioannis and Poupouzas, Georgios and Issaris, Vasileios and Dede, Konstantina and Christopoulou, Helen and Karkoulias, Kyriakos and Raftopoulou, Sylvia and Pouriki, Sophia and Keskinidou, Chrysi and Lotsios, Nikolaos S. and Ganitidis, Theofanis and Golemati, Spyretta and Kotanidou, Anastasia and Papapanagiotou, Stavros and Dimopoulou, Ioanna},
title = {Integrating Inflammatory and Endothelial Biomarkers in Machine Learning Models to Predict Sepsis Development in Critically Ill Patients: A Combined Retrospective and Prospective Multicenter Study},
journal = {Pathogens},
year = {2026},
volume = {15},
number = {11},
pages = {1214},
doi = {10.3390/pathogens15111214}
}
EUVIMED