Interpretable machine learning-based prediction of 28-day mortality in ICU patients with sepsis: a multicenter retrospective study
Interpretierbares Machine-Learning zur Vorhersage der 28-Tage-Mortalität bei ICU-Patient:innen mit Sepsis: multizentrische retrospektive Studie
Frontiers in Cellular and Infection Microbiology · 08.01.2025 · 14
DOI: 10.3389/fcimb.2024.1500326 ↗Auf einen Blick
Ein Machine-Learning-Modell sagte die 28-Tage-Gesamtmortalität bei ICU-Patient:innen mit Sepsis vorher. XGBoost zeigte die stabilste Leistung.
Fragestellung
Lässt sich die 28-Tage-Gesamtmortalität bei ICU-Patient:innen mit Sepsis mit Machine-Learning-Modellen vorhersagen?
Methode
Retrospektive multizentrische Studie mit Daten aus der eICU-CRD 2.0; 5 ML-Algorithmen, 10-fold cross-validation; Bewertung mit AUC, Accuracy, Sensitivity, Specificity, Recall und F1-Score.
Ergebnisse
4564 Patient:innen; 568 starben innerhalb von 28 Tagen nach ICU-Aufnahme (12,4 %). XGBoost: AUC 0,821, Sensitivity 0,703, Specificity 0,798. Top-Prädiktoren: APACHE-Score, Laktat, AST.
Schlussfolgerung der Autor:innen
ML-Modelle sagten die 28-Tage-Mortalität bei kritisch kranken Sepsis-Patient:innen zuverlässig voraus. XGBoost zeigte die stabilste Leistung zur Identifikation von Patient:innen mit erhöhtem Mortalitätsrisiko.
Kernaussagen
- 4564 Patient:innen; 12,4 % 28-Tage-Mortalität
- XGBoost: AUC 0,821
- Wichtigste Prädiktoren: APACHE-Score, Laktat, AST
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Multizentrische Studie in der Zeitschrift Frontiers in Cellular and Infection Microbiology, online erschienen am 08.01.2025. Verfasst von Li Shen und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Clinical Pharmacy, Children's Hospital of Fudan University, National Children's Medical Center, Shanghai, China. Hauptthemen laut MEDLINE: Sepsis, Intensive Care Units, Machine Learning. Die Arbeit wurde bisher 36-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Themen: Künstliche Intelligenz in der Pflege, Intensivpflege, Sepsis
Abstract (Originalsprache)
Background
Sepsis is a major cause of mortality in intensive care units (ICUs) and continues to pose a significant global health challenge, with sepsis-related deaths contributing substantially to the overall burden on healthcare systems worldwide. The primary objective was to construct and evaluate a machine learning (ML) model for forecasting 28-day all-cause mortality among ICU sepsis patients.
Methods
Data for the study was sourced from the eICU Collaborative Research Database (eICU-CRD) (version 2.0). The main outcome was 28-day all-cause mortality. Predictor selection for the final model was conducted using the least absolute shrinkage and selection operator (LASSO) regression analysis and the Boruta feature selection algorithm. Five machine learning algorithms including logistic regression (LR), decision tree (DT), extreme gradient boosting (XGBoost), support vector machine (SVM), and light gradient boosting machine (lightGBM) were employed to construct models using 10-fold cross-validation. Model performance was evaluated using AUC, accuracy, sensitivity, specificity, recall, and F1 score. Additionally, we performed an interpretability analysis on the model that showed the most stable performance.
Results
The final study cohort comprised 4564 patients, among whom 568 (12.4%) died within 28 days of ICU admission. The XGBoost algorithm demonstrated the most reliable performance, achieving an AUC of 0.821, balancing sensitivity (0.703) and specificity (0.798). The top three risk predictors of mortality included APACHE score, serum lactate levels, and AST.
Conclusion
ML models reliably predicted 28-day mortality in critically ill sepsis patients. Of the models evaluated, the XGBoost algorithm exhibited the most stable performance in identifying patients at elevated mortality risk. Model interpretability analysis identified crucial predictors, potentially informing clinical decisions for sepsis patients in the ICU.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Clinical Pharmacy, Children's Hospital of Fudan University, National Children's Medical Center, Shanghai, China.
- Department of Pharmacy, Suzhou Hospital, Affiliated Hospital of Medical School, Nanjing University, Suzhou, Jiangsu, China.
- School of Life Sciences and Biopharmaceutical Science, Shenyang Pharmaceutical University, Shenyang, China.
- Department of Neonatology, Children's Hospital of Fudan University, National Children's Medical Center, Shanghai, China.
- Department of Neurology, Children's Hospital of Fudan University, National Children's Medical Center, Shanghai, China.
Zitieren
APA
Shen, L., Wu, J., Lan, J., Chen, C., Wang, Y., & Li, Z. (2025). Interpretable machine learning-based prediction of 28-day mortality in ICU patients with sepsis: a multicenter retrospective study. Frontiers in Cellular and Infection Microbiology, 14. https://doi.org/10.3389/fcimb.2024.1500326
Vancouver
Shen L, Wu J, Lan J, Chen C, Wang Y, Li Z. Interpretable machine learning-based prediction of 28-day mortality in ICU patients with sepsis: a multicenter retrospective study. Frontiers in Cellular and Infection Microbiology. 2025;14. doi:10.3389/fcimb.2024.1500326
BibTeX
@article{Shen2025,
author = {Shen, Li and Wu, Jiaqiang and Lan, Jianger and Chen, Chao and Wang, Yi and Li, Zhiping},
title = {Interpretable machine learning-based prediction of 28-day mortality in ICU patients with sepsis: a multicenter retrospective study},
journal = {Frontiers in Cellular and Infection Microbiology},
year = {2025},
volume = {14},
doi = {10.3389/fcimb.2024.1500326}
}
EUVIMED