Researchxgboostcritical caresepsisinterpretability
Interpretable XGBoost Predicts 28-Day Mortality in Sepsis ARF
|
8.2
Researchers led by Dr. Jian Liu published on January 10, 2026 an interpretable XGBoost model that predicts 28-day mortality for ICU patients with sepsis complicated by acute respiratory failure. They trained the model on MIMIC‑IV v3.1 and externally validated it on eICU‑CRD v2.0, selecting 20 routinely available features and using SHAP for explanation, showing stable discrimination across cohorts.
Scoring Rationale
Strong external validation and interpretability raise impact, but modest methodological novelty and niche clinical scope limit broader disruption.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
