A study in Scientific Reports evaluated an explainable machine learning model’s ability to predict 28-day in-hospital mortality in ICU patients with sepsis and atrial fibrillation. Based on 11,740 patients from MIMIC-IV, MIMIC-III, and eICU, GBM performed best, with an internal validation AUC of 0.770, and SHAP was used to explain variable contributions. The study shows that the model has moderate to good discriminatory ability, but further research is needed to verify its cross-scenario applicability and value in routine clinical decision-making.
Source:HealthManagement.org · Read original
Leave a Reply