Researchers have developed and validated a machine learning tool to predict 28-day mortality in ICU patients with sepsis and acute respiratory failure using early clinical data.
Key Details
- 1Machine learning model predicts 28-day mortality risk for ICU patients with sepsis complicated by acute respiratory failure.
- 2Routinely collected clinical data from the first 24 hours of ICU admission were used as input.
- 3Model was trained using MIMIC-IV (v3.1) and externally validated on eICU-CRD (v2.0) databases.
- 4XGBoost outperformed other algorithms in mortality risk discrimination and generalizability.
- 5Model focused on interpretability using SHapley Additive exPlanations (SHAP) to highlight key clinical predictors.
- 6Study published in Journal of Intensive Medicine on January 10, 2026 (DOI: 10.1016/j.jointm.2025.10.010).
Why It Matters
This development demonstrates how interpretable AI can support early clinical risk stratification, guiding resource allocation and treatment in critical care settings. Such models, leveraging large clinical datasets and interpretability tools, are key for integrating AI decision support into real-world healthcare.
Source
EurekAlert
Related News
•EurekAlert
AI-Enhanced Imaging Breaks Barriers in Complex Media for Biomedical Use
Researchers developed a physics-based machine learning imaging system that enhances visibility through complex media, holding promise for biomedical imaging.
•EurekAlert
AI Model Accelerates Accurate Dose Prediction for CyberKnife Radiotherapy
A physics- and spatially informed diffusion model accelerates Monte Carlo–level dose estimation for CyberKnife radiotherapy using planning CT and anatomical data.
•EurekAlert
New G-AUDIT Tool Detects Hidden Bias in Medical AI Training Data
Johns Hopkins and FDA researchers unveil G-AUDIT, a tool to identify hidden biases in medical AI datasets before model training.