Table 2 Model performance for prediction of ESKD at 3 years
From: Machine learning models to predict end-stage kidney disease in chronic kidney disease stage 4
Model | Model performance metric | |||
|---|---|---|---|---|
Accuracy | Precision | AUROC | AUPRC | |
LASSO | 0.83 (0.82 to 0.85) | 0.58 (0.51 to 0.66) | 0.77 (0.75 to 0.79) | 0.45 (0.40 to 0.49) |
Random Forest | 0.82 (0.81 to 0.83) | 0.38 (0.35 to 0.42) | 0.76 (0.74 to 0.79) | 0.44 (0.39 to 0.48) |
XGBoost | 0.84 (0.83 to 0.85) | 0.38 (0.35 to 0.42) | 0.76 (0.74 to 0.78) | 0.43 (0.38 to 0.47) |
Neural network | 0.83 (0.82 to 0.85) | 0.73 (0.60 to 0.87) | 0.77 (0.75 to 0.79) | 0.44 (0.39 to 0.48) |