Developing an ensemble machine learning model for early prediction of sepsis-associated acute kidney injury
Luming Zhang,
Zichen Wang,
Zhenyu Zhou,
Shaojin Li,
Tao Huang,
Haiyan Yin,
Jun Lyu
Affiliations
Luming Zhang
Department of Intensive Care Unit, The First Affiliated Hospital of Jinan University, Guangzhou, Guangdong Province 510630, China; Department of Clinical Research, The First Affiliated Hospital of Jinan University, Guangzhou, Guangdong Province 510630, China
Zichen Wang
Department of Clinical Research, The First Affiliated Hospital of Jinan University, Guangzhou, Guangdong Province 510630, China; Department of Public Health, University of California, Irvine, CA 92697, USA
Zhenyu Zhou
Warshel Institute for Computational Biology, School of Life and Health Sciences, The Chinese University of Hong Kong (Shenzhen), Shenzhen, Guangdong 518172, China
Shaojin Li
Department of Orthopaedics, The First Affiliated Hospital of Jinan University, Guangzhou, Guangdong Province 510630, China
Tao Huang
Department of Clinical Research, The First Affiliated Hospital of Jinan University, Guangzhou, Guangdong Province 510630, China
Haiyan Yin
Department of Intensive Care Unit, The First Affiliated Hospital of Jinan University, Guangzhou, Guangdong Province 510630, China; Corresponding author
Jun Lyu
Department of Clinical Research, The First Affiliated Hospital of Jinan University, Guangzhou, Guangdong Province 510630, China; Corresponding author
Summary: Sepsis-associated acute kidney injury (S-AKI) is very common and early prediction is beneficial. This study aiming to develop an accurate ensemble model to predict the risk of S-AKI based on easily available clinical information. Patients with sepsis from the United States (US) database Medical Information Mart for Intensive Care-IV were used as a modeling cohort to predict the occurrence of AKI by combining Support Vector Machine, Random Forest, Neural Network, and Extreme Gradient Boost as four first-level learners via stacking algorithm. The external validation databases were the eICU Collaborative Research Database from US and Critical Care Database comprising infection patients at Zigong Fourth People’s Hospital from China, whose AUROC values for the ensemble model 48–12 h before the onset of AKI were 0.774–0.788 and 0.756–0.813, respectively. In this study, an ensemble model for early prediction of S-AKI onset was developed and it demonstrated good performance in multicenter external datasets.
