Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 22, 2026Scientific ReportsOpen Access

Development of interpretable machine learning models for early diagnosis of sepsis-associated acute kidney injury

View Full Paper
Ask AI
Bookmark
Share

Authors

XZXiaoya ZhangZZZhenqi ZhangJYJili Yang

Discussion

Loading...

Member takes

Overview

Retrospective study demonstrates accurate prediction of sepsis-associated acute kidney injury in ICU patients using interpretable XGBoost, suggesting utility for early clinical risk stratification.

Key Points

  • To develop and validate an interpretable machine learning model using routine clinical predictors for the early diagnosis and risk stratification of sepsis-associated acute kidney injury.
  • Analyzed data from a retrospective dual-center cohort of 687 septic ICU patients for model development and 125 independent cases for external validation in Northwest China.
  • Handled missing data with MICE, addressed class imbalance using SMOTE, and selected 19 clinical predictors via LASSO regression (λ_min = 0.0066).
  • Trained and compared 10 machine learning algorithms, utilizing SHAP analysis to resolve model interpretability and evaluate predictor interactions.
  • XGBoost demonstrated balanced overall predictive performance with good calibration, achieving an AUC of 0.91 and an internal validation sensitivity of 0.86.
  • SHAP analysis identified blood urea nitrogen as the primary risk biomarker while uncovering nonlinear and synergistic interactions among predictors.
  • External validation and decision curve analysis confirmed model stability and sustained net clinical benefit.

Cite This Study

Zhang et al. (2026) studied this question.

synapsesocial.com/papers/6a895ec3ca7ade938187cda9https://doi.org/10.1038/s41598-026-67412-1
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Risk prediction of sepsis-associated acute kidney injury: development, validation of a machine learning model with multicenter data2026
  2. 2An interpretable machine-learning model for predicting in-hospital mortality in patients with sepsis-associated acute kidney injury2026 · 1 citations
  3. 3Interpretable Machine Learning for Early Prediction of Acute Kidney Disease (AKD) in Sepsis-Associated Acute Kidney Injury (SA-AKI): A Multicenter Cohort Study with External Validation 2025
  4. 4Development and validation of a machine-learning model for predicting the risk of death in sepsis patients with acute kidney injury2024 · 18 citations
  5. 5Machine learning for prediction of acute kidney injury in patients diagnosed with sepsis in critical care2024 · 9 citations