Machine Learning for the Prediction of Acute Kidney Injury in Critically Ill Patients With Coronary Heart Disease: Algorithm Development and Validation

Abstract

Background: Acute kidney injury (AKI) frequently occurs in critically ill patients with coronary heart disease (CHD), and its development markedly elevates mortality rates and prolongs hospitalization duration. Early AKI prediction is crucial for timely intervention and amelioration of patient outcomes.

Objective: This study aimed to develop and verify a clinical prediction model for the occurrence of AKI upon admission in the critically ill population with CHD through machine learning (ML).

Methods: Data from the MIMIC-IV (Medical Information Mart for Intensive Care IV) version 2.2 database were gathered and included information about critically ill individuals with CHD in the intensive care unit (ICU). The dataset was randomized into a training set (70%) and a testing set (30%). Least absolute shrinkage and selection operator (LASSO) regression was used for feature variable selection. ML models, including logistic regression (LR), decision tree (DT), naive Bayes (NB), random forest (RF), extreme gradient boosting (XGBoost), and support vector machine (SVM), were constructed using 13 variables in the training set. The 6 models were compared in the testing set to identify the best-performing model. Subsequently, the model was assessed using calibration curve analysis and decision curve analysis (DCA). External validation was conducted using data from the Second Affiliated Hospital of Zhengzhou University. Ultimately, the predictive model was interpreted via Shapley Additive Explanation (SHAP) values.

Results: In total, 2711 patients with CHD admitted to the ICU were selected, with 1809 (66.7%) having AKI. XGBoost exhibited the best performance regarding discrimination (area under the receiver operating characteristic curve [AUROC]=0.765, 95% CI 0.731-0.800), accuracy (0.725), and sensitivity (0.759). External validation using a cohort of 226 patients confirmed the strong generalizability of the XGBoost model (AUROC=0.835, 95% CI 0.782-0.887). Feature importance analyses derived from SHAP values, DT, RF, and XGBoost consistently identified 5 key predictors associated with the development of AKI: mechanical ventilation, use of antiplatelet agents, age, N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels, and acute physiology score III (APSIII).

Conclusions: ML models can serve as reliable tools for forecasting AKI in the critically ill population with CHD. The XGBoost model is highly accurate and may aid doctors in identifying high-risk individuals for early intervention to lower mortality.

Keywords: MIMIC-IV database; acute kidney injury; coronary artery disease; coronary heart disease; machine learning.

Figure 1

Participant-screening process diagram. AKI: acute kidney injury; ICU: intensive care unit; MIMIC-IV: Medical Information Mart for Intensive Care IV.

Figure 2

Comparison of AUC values across 6 models: (a) training set and (b) testing set. AUC: area under the curve; KNB: kernel naive Bayes; SVM: support vector machine; XGBoost: extreme gradient boosting.

Figure 3

Calibration curve of the XGBoost model: (a) training set and (b) testing set. DCA of the XGBoost model: (c) training set and (d) testing set. DCA: decision curve analysis; XGBoost: extreme gradient boosting.

Figure 4

(a) Comparison of AUC values of the 6 models in the external validation set and (b) the calibration curve and (c) the DCA curve of the XGBoost model. AUC: area under the curve; DCA: decision curve analysis; KNB: kernel naive Bayes; SVM: support vector machine; XGBoost: extreme gradient boosting.

Figure 5

Feature importance derived from (a) XGBoost, (b) DT, and (c) RF models, along with (d) SHAP analysis summary diagram based on XGBoost. Each row represents a feature; a point represents a sample; yellow represents a high feature value; and purple represents a low feature value. A further distance from a point to the baseline SHAP value of 0 indicates a greater impact on the output. AMI: acute myocardial infarction; APSIII: acute physiology score III; NLR: neutrophil-to-lymphocyte ratio; NT-proBNP: N-terminal pro–B-type natriuretic peptide.

Figure 6

SHAP force plot of (a) a patient with AKI and (b) a patient without AKI. Features contributing to an increase in the predicted value are shown in yellow, whereas those contributing to a decrease are shown in red. The length of the arrows represents the magnitude of each feature’s influence on the model output. The scale on the x-axis indicates the extent to which each feature increases or decreases the predicted value. AKI: acute kidney injury; AMI: acute myocardial infarction; NT-proBNP: N-terminal pro–B-type natriuretic peptide; SHAP: Shapley Additive Explanation.