Tao Zhang,1,* Zi-Han Nan,1,* Xiao-Xuan Fan,1 Jing-Xiao Pang,1 Cong-Cong Zhao,1 Yan Xin,2 Zhen-Jie Hu,1 Shao-Han Guo1 

1Department of Intensive Care Unit, The Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei Province, 050000, People’s Republic of China; 2Department of Intensive Care Unit, The Third Hospital of Shijiazhuang, Shijiazhuang, Hebei Province, 050000, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Shao-Han Guo, Department of Intensive Care Unit, The Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei Province, 050000, People’s Republic of China, Email 48602779@hebmu.edu.cn

Purpose: This study aimed to develop and validate an interpretable machine learning (ML) model to predict 28-day all-cause mortality in critically ill patients undergoing continuous renal replacement therapy (CRRT), facilitating early risk stratification and clinical decision-making.
Patients and Methods: Data from 1362 CRRT patients were analyzed, including 1224 from the Medical Information Mart for Intensive Care IV database (training cohort) and 138 from a Chinese hospital (external validation cohort). Feature selection was performed using least absolute shrinkage and selection operator, support vector machine-recursive feature elimination, and Boruta algorithms. Nine machine learning models were constructed and compared, including Gaussian process (GP), ensemble methods (gradient boosting machine and eXtreme gradient boosting), and other classifiers. Model performance was assessed via the area under the receiver operating characteristic curve (AUC), decision curve analysis, and other metrics. The SHapley Additive exPlanation (SHAP) method was used to interpret the ML models.
Results: The GP model demonstrated consistent predictive performance across all cohorts, with training (AUC=0.841, accuracy=76.8%, sensitivity=65.5%), internal validation (AUC=0.794, accuracy=73.4%, sensitivity=60.0%), and external validation (AUC=0.780, accuracy=63.8%, sensitivity=39.0%) sets. Key predictors included red cell distribution width, age, lactate, septic shock, and vasoactive drug use. SHAP analysis provided transparent insights into feature contributions.
Conclusion: The GP-based model accurately predicts 28-day mortality in CRRT patients and demonstrates strong generalizability. By integrating SHAP explanations, it offers clinicians an interpretable tool to identify high-risk patients early, potentially improving outcomes.

Keywords: MIMIC-IV database, critical illness, mortality prediction, machine learning, Gaussian process, SHAP, interpretability, external validation