EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A machine learning model for robust prediction of sepsis-induced coagulopathy in critically ill patients with sepsis

Ein Machine-Learning-Modell zur robusten Vorhersage einer sepsisinduzierten Koagulopathie bei kritisch kranken Patient:innen mit Sepsis

Jia Sun, Lixin Zhang, Zhaotang Gong, Hongling Ma, Dan Wu, Rihan Wu, Guleng Siri

Frontiers in Cellular and Infection Microbiology · 06.06.2025 · 15

DOI: 10.3389/fcimb.2025.1579558 ↗

Auf einen Blick

In einer retrospektiven Kohorte sagte ein ML-Modell das Risiko einer sepsisinduzierten Koagulopathie (SIC) bei Patient:innen mit Sepsis gut voraus. Das GBM-Modell schnitt besser ab als SOFA-basierte Modelle.

Fragestellung

Kann ein praktisches ML-Modell das Risiko einer SIC bei kritisch kranken Patient:innen mit Sepsis vorhersagen?

Methode

Retrospektive Kohortenstudie mit Daten aus MIMIC-IV und einem Krankenhaus in der Inner Mongolia Autonomous Region; Vergleich von neun ML-Modellen anhand des SOFA-Score.

Ergebnisse

15.479 Patient:innen aus MIMIC-IV; 6.036 entwickelten SIC. Das GBM-Modell zeigte AUC 0,773 (95 %-KI 0,765–0,782) im Training, 0,730 (95 %-KI 0,715–0,745) intern und 0,966 (95 %-KI 0,938–0,994) extern.

Schlussfolgerung der Autor:innen

Die Autor:innen entwickelten ein optimales, praktikables ML-Modell, das das SIC-Risiko bei Patient:innen mit Sepsis besser vorhersagte als SOFA-Modelle.

Kernaussagen

  • SIC trat bei 6.036 von 15.479 Patient:innen auf.
  • GBM war das beste Modell unter neun ML-Modellen.
  • Wichtige Prädiktoren waren Bilirubin, RDW, SBP, Heparin und BUN.

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Cellular and Infection Microbiology, online erschienen am 06.06.2025. Verfasst von Jia Sun und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pharmacy, Inner Mongolia People's Hospital, Hohhot, Inner Mongolia Autonomous Region, China. Hauptthemen laut MEDLINE: Sepsis, Machine Learning, Blood Coagulation Disorders. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Sepsis-induced coagulopathy (SIC) is a common disease in patients with sepsis. It denotes higher mortality rates and a poorer prognosis in these patients. This study aimed to develop a practical machine learning (ML) model for the prediction of the risk of SIC in critically ill patients with sepsis.

Methods

In this retrospective cohort study, patients were extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database and the Inner Mongolia Autonomous Region People’s Hospital database. Sepsis and SIC were defined based on the Sepsis-3 criteria and the criteria developed based on the International Society of Thrombosis and Haemostasis (ISTH), respectively. We compared nine ML models using the Sequential Organ Failure Assessment (SOFA) score in terms of SIC prediction ability. Optimal model selection was based on the superior performance metrics exhibited by the model on the training dataset, the internal validation dataset, and the external validation dataset.

Results

Of the 15,479 patients in MIMIC-IV included in the final cohort, a total of 6,036 (38.9%) patients developed SIC during sepsis. We selected 17 features to construct ML prediction models. The gradient boosting machine (GBM) model was deemed optimal as it achieved high predictive accuracy and reliability across the training, internal, and external validation datasets. The areas under the curve of the GBM model were 0.773 (95%CI = 0.765–0.782) in the training dataset, 0.730 (95%CI = 0.715–0.745) in the internal validation dataset, and 0.966 (95%CI = 0.938–0.994) in the external validation dataset. The Shapley Additive Explanations (SHAP) values illustrated the prediction results, indicating that total bilirubin, red cell distribution width (RDW), systolic blood pressure (SBP), heparin, and blood urea nitrogen (BUN) were risk factors for progression to SIC in patients with sepsis.

Conclusions

We developed an optimal and operable ML model that was able to predict the risk of SIC in septic patients better than the SOFA scoring models.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SepsisMachine LearningBlood Coagulation DisordersHumansCritical IllnessRetrospective StudiesMaleFemaleMiddle AgedAgedOrgan Dysfunction ScoresPrognosisReproducibility of ResultsRisk Factors

Stichwörter der Autor:innen

machine learningpredictrisk factorsepsissepsis-induced coagulopathy

Autor:innen und Institutionen

  • Department of Pharmacy, Inner Mongolia People's Hospital, Hohhot, Inner Mongolia Autonomous Region, China.
  • Department of Pharmacy, Baotou Medical College, Baotou, Inner Mongolia Autonomous Region, China.
  • Real-World Research Center, Inner Mongolia Academy of Medical Sciences, Hohhot, Inner Mongolia Autonomous Region, China.

Zitieren

APA

Sun, J., Zhang, L., Gong, Z., Ma, H., Wu, D., Wu, R., & Siri, G. (2025). A machine learning model for robust prediction of sepsis-induced coagulopathy in critically ill patients with sepsis. Frontiers in Cellular and Infection Microbiology, 15. https://doi.org/10.3389/fcimb.2025.1579558

Vancouver

Sun J, Zhang L, Gong Z, Ma H, Wu D, Wu R, et al. A machine learning model for robust prediction of sepsis-induced coagulopathy in critically ill patients with sepsis. Frontiers in Cellular and Infection Microbiology. 2025;15. doi:10.3389/fcimb.2025.1579558

BibTeX
@article{Sun2025,
  author = {Sun, Jia and Zhang, Lixin and Gong, Zhaotang and Ma, Hongling and Wu, Dan and Wu, Rihan and Siri, Guleng},
  title = {A machine learning model for robust prediction of sepsis-induced coagulopathy in critically ill patients with sepsis},
  journal = {Frontiers in Cellular and Infection Microbiology},
  year = {2025},
  volume = {15},
  doi = {10.3389/fcimb.2025.1579558}
}
Kostenlos starten →