Researchxgboosticucardiac surgerydelirium
Machine Learning Predicts ICU Delirium After Cardiac Surgery
|
8.2
Researchers developed and externally validated an XGBoost model to predict postoperative delirium using first 24-hour ICU data from MIMIC-IV 2.0 (n=2,124) and eICU-CRD (n=2,406). The model used 57 features and achieved an external AUC of 0.75, with top predictors including hospital length of stay, minimum GCS, mean blood pressure, and SOFA score. The tool could support real-time delirium alerts and risk stratification in cardiac surgery ICUs.
Scoring Rationale
External validation and practical predictors support high impact; modest novelty within existing clinical-ML literature limits breakthrough status.
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
