Machine Learning Models to Predict the Risk of Rapidly Progressive Kidney Disease and the Need for Nephrology Referral in Adult Patients with Type 2 Diabetes

Abstract

Early detection of rapidly progressive kidney disease is key to improving the renal outcome and reducing complications in adult patients with type 2 diabetes mellitus (T2DM). We aimed to construct a 6-month machine learning (ML) predictive model for the risk of rapidly progressive kidney disease and the need for nephrology referral in adult patients with T2DM and an initial estimated glomerular filtration rate (eGFR) ≥ 60 mL/min/1.73 m². We extracted patients and medical features from the electronic medical records (EMR), and the cohort was divided into a training/validation and testing data set to develop and validate the models on the basis of three algorithms: logistic regression (LR), random forest (RF), and extreme gradient boosting (XGBoost). We also applied an ensemble approach using soft voting classifier to classify the referral group. We used the area under the receiver operating characteristic curve (AUROC), precision, recall, and accuracy as the metrics to evaluate the performance. Shapley additive explanations (SHAP) values were used to evaluate the feature importance. The XGB model had higher accuracy and relatively higher precision in the referral group as compared with the LR and RF models, but LR and RF models had higher recall in the referral group. In general, the ensemble voting classifier had relatively higher accuracy, higher AUROC, and higher recall in the referral group as compared with the other three models. In addition, we found a more specific definition of the target improved the model performance in our study. In conclusion, we built a 6-month ML predictive model for the risk of rapidly progressive kidney disease. Early detection and then nephrology referral may facilitate appropriate management.

Keywords: diabetic kidney disease; machine learning; nephrology referral; type 2 diabetes.

Figure 1

Time frame of our study design and label definition.

Figure 2

The architecture of our prediction models.

Figure 3

Confusion matrix and predictive probabilities histogram of the XGBoost model in experiment 1 (persistent eGFR < 30 mL/min/1.73 m²). The green in the histogram represents the referral group, and the medium slate blue represents the non-referral group.

Figure 4

SHAP summary plot of the top 15 features for the XGBoost model in experiment 1.

Figure 5

Confusion matrix and predictive probabilities histogram of the XGBoost model in experiment 2 (persistent eGFR < 45 mL/min/1.73 m²). The green in the histogram represents the referral group, and the medium slate blue represents the non-referral group.

Figure 6

SHAP summary plot of the top 15 features of the XGBoost model in experiment 2.