Development and validation of a novel blending machine learning model for hospital mortality prediction in ICU patients with Sepsis

BioData Mining

Zhixuan Zeng¹  Jianfei Zheng¹  Shuo Yao¹  Xun Gong¹ 

[1] Department of Emergency Medicine, Second Xiangya Hospital, Central South University, No.139, Middle Renmin Road, 410011, Changsha, Hunan Province, China;Emergency Medicine and Difficult Diseases Institute, Second Xiangya Hospital, Central South University, No.139, Middle Renmin Road, 410011, Changsha, Hunan Province, China;

关键词: Sepsis;    Intensive care unit;    Hospital mortality prediction;    Machine learning;    MIMIC-III;    eICU-CRD;   

DOI  :  10.1186/s13040-021-00276-5

来源: Springer

【 摘 要 】

BackgroundEarly prediction of hospital mortality is crucial for ICU patients with sepsis. This study aimed to develop a novel blending machine learning (ML) model for hospital mortality prediction in ICU patients with sepsis.MethodsTwo ICU databases were employed: eICU Collaborative Research Database (eICU-CRD) and Medical Information Mart for Intensive Care III (MIMIC-III). All adult patients who fulfilled Sepsis-3 criteria were identified. Samples from eICU-CRD constituted training set and samples from MIMIC-III constituted test set. Stepwise logistic regression model was used for predictor selection. Blending ML model which integrated nine sorts of basic ML models was developed for hospital mortality prediction in ICU patients with sepsis. Model performance was evaluated by various measures related to discrimination or calibration.ResultsTwelve thousand five hundred fifty-eight patients from eICU-CRD were included as the training set, and 12,095 patients from MIMIC-III were included as the test set. Both the training set and the test set showed a hospital mortality of 17.9%. Maximum and minimum lactate, maximum and minimum albumin, minimum PaO2/FiO2 and age were important predictors identified by both random forest and extreme gradient boosting algorithm. Blending ML models based on corresponding set of predictors presented better discrimination than SAPS II (AUROC, 0.806 vs. 0.771; AUPRC 0.515 vs. 0.429) and SOFA (AUROC, 0.742 vs. 0.706; AUPRC 0.428 vs. 0.381) on the test set. In addition, calibration curves showed that blending ML models had better calibration than SAPS II.ConclusionsThe blending ML model is capable of integrating different sorts of basic ML models efficiently, and outperforms conventional severity scores in predicting hospital mortality among septic patients in ICU.

【 授权许可】

CC BY