The prediction of in-hospital mortality in chronic kidney disease patients with coronary artery disease using machine learning models

Abstract

Objective: Chronic kidney disease (CKD) patients with coronary artery disease (CAD) in the intensive care unit (ICU) have higher in-hospital mortality and poorer prognosis than patients with either single condition. The objective of this study is to develop a novel model that can predict the in-hospital mortality of that kind of patient in the ICU using machine learning methods.

Methods: Data of CKD patients with CAD were extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. Boruta algorithm was conducted for the feature selection process. Eight machine learning algorithms, such as logistic regression (LR), random forest (RF), Decision Tree, K-nearest neighbors (KNN), Gradient Boosting Decision Tree Machine (GBDT), Support Vector Machine (SVM), Neural Network (NN), and Extreme Gradient Boosting (XGBoost), were conducted to construct the predictive model for in-hospital mortality and performance was evaluated by average precision (AP) and area under the receiver operating characteristic curve (AUC). Shapley Additive Explanations (SHAP) algorithm was applied to explain the model visually. Moreover, data from the Telehealth Intensive Care Unit Collaborative Research Database (eICU-CRD) were acquired as an external validation set.

Results: 3590 and 1657 CKD patients with CAD were acquired from MIMIC-IV and eICU-CRD databases, respectively. A total of 78 variables were selected for the machine learning model development process. Comparatively, GBDT had the highest predictive performance according to the results of AUC (0.946) and AP (0.778). The SHAP method reveals the top 20 factors based on the importance ranking. In addition, GBDT had good predictive value and a certain degree of clinical value in the external validation according to the AUC (0.865), AP (0.672), decision curve analysis, and calibration curve.

Conclusion: Machine learning algorithms, especially GBDT, can be reliable tools for accurately predicting the in-hospital mortality risk for CKD patients with CAD in the ICU. This contributed to providing optimal resource allocation and reducing in-hospital mortality by tailoring precise management and implementation of early interventions.

Keywords: Chronic kidney disease; Coronary artery disease; In-hospital mortality; MIMIC-IV database; Machine learning.

Fig. 1

Flowchart of patient selection from MIMIC-IV and eICU-CRD database. MIMIC Medical Information Mort for Intensive Care, eICU-CRD Telehealth Intensive Care Unit Collaborative Research Database

Fig. 2

Feature selection analyzed by Boruta algorithm. The horizontal axis is the name of each variable, and the vertical axis is the Z-value of each variable. The box plot shows the Z-value of each variable in the model calculation. The green boxes represent the 76 important variables, the yellow represents tentative attributes, and the red represents unimportant variables. los_icu length of stay in intensive care unit, scr serum creatinine, eGFR estimated glomerular filtration rate, CKD chronic kidney disease, ACS acute coronary syndrome, HT hypertension, PCI percutaneous coronary intervention, CABG coronary artery bypass grafting, NOAC Non-vitamin K Antagonist Oral Anticoagulant, CRRT continuous renal replacement therapy, max maximum, min minimum, WBC white blood cell, RBC red blood cell, ALT alanine aminotransferase, AST aspartate aminotransferase, ALP alkaline phosphatase, BUN blood urea nitrogen, INR International Normalized Ratio, PT prothrombin time, PTT partial thromboplastin time, SOFA sequential organ failure assessment, sbp systolic blood pressure, dbp diastolic blood pressure, mbp mean blood pressure, HR heart rate, spo2 oxyhemoglobin saturation

Fig. 3

Discrimination performance of eight machine learning models. A ROC of eight machine learning models. B P-R curves of eight machine learning models. The GBDT algorism exhibited the best performance both in ROC and P-R curves. ROC Receiver Operating Characteristic, P-R curve precision/recall curve, SVM support vector machine, GBDT Gradient Boosting Decision Tree Machine, KNN k-nearest neighbors, NN neural network, XGBoost Extreme Gradient Boosting, AUC area under the curve

Fig. 4

SHAP analysis result. A Bar charts that rank the importance of the top 20 significant variables most correlated to in-hospital death in GBDT model. B Impact of each feature on the in-hospital mortality in GBDT model by SHAP values. GBDT Gradient Boosting Decision Tree Machine, SHAP Shapley Additive Explanations, spo2 oxyhemoglobin saturation, HR heart rate, WBC white blood cell, CABG coronary artery bypass grafting, SOFA sequential organ failure assessment, sbp systolic blood pressure, BUN blood urea nitrogen, PTT partial thromboplastin time, ALT alanine aminotransferase, AST aspartate aminotransferase, PT prothrombin time

Fig. 5

External validation for the GBDT model in the eICU-CRD dataset. A DCA curve of the GBDT model in external validation. B calibration curve of the GBDT model in external validation. C ROC of the GBDT model in external validation. D P-R curves of the GBDT models in external validation. DCA showed the GBDT model had some net benefit compared with the “treat-none” or “treat-all” strategies with a certain degree of clinical utility. The AUC (0.865) and AP (0.672) results demonstrated the GBDT model had good predictive values in external validation. DCA decision curve analysis, ROC Receiver Operating Characteristic, P-R curve precision/recall curve, GBDT Gradient Boosting Decision Tree Machine, eICU-CRD Telehealth Intensive Care Unit Collaborative Research Database