From: Machine learning for the prediction of acute kidney injury in patients with sepsis
Models | AUC | Recall | Accuracy | F1 score | Sensitivity | Specificity |
---|---|---|---|---|---|---|
LR | 0.737 | 0.796 | 0.765 | 0.858 | 0.834 | 0.878 |
KNN | 0.664 | 0.798 | 0.742 | 0.840 | 0.886 | 0.857 |
SVM | 0.735 | 0.797 | 0.788 | 0.874 | 0.833 | 0.926 |
Decision tree | 0.749 | 0.834 | 0.793 | 0.870 | 0.910 | 0.882 |
Random forest | 0.779 | 0.809 | 0.794 | 0.876 | 0.935 | 0.923 |
XGBoost | 0.817 | 0.852 | 0.832 | 0.895 | 0.943 | 0.913 |
ANN | 0.755 | 0.778 | 0.783 | 0.875 | 0.824 | 0.899 |
SOFA | 0.646 | 0.755 | 0.723 | 0.781 | 0.633 | 0.712 |
SAPS II | 0.702 | 0.774 | 0.762 | 0.814 | 0.811 | 0.845 |