All-cause mortality | Accuracy | Sensitivity | Specificity | PPV | NPV | F1 | AUC | p-value* |
MLP model | 0.776 | 0.637 | 0.797 | 0.314 | 0.938 | 0.421 | 0.796 |  < 0.001 |
PHD score | 0.670 | 0.637 | 0.675 | 0.223 | 0.927 | 0.330 | 0.693 |
ICU admission | Accuracy | Sensitivity | Specificity | PPV | NPV | F1 | AUC | p-value |
MLP model | 0.809 | 0.521 | 0.827 | 0.161 | 0.964 | 0.246 | 0.743 | 0.084 |
PHD score | 0.671 | 0.521 | 0.681 | 0.094 | 0.957 | 0.160 | 0.641 |
- ICU Intensive care unit, AI Artificial intelligence, MLP Multilayer perceptron, PHD Predicting the hyperglycemic crisis death, PPV Positive predictive value, NPV Negative predictive value; F1, 2 × (precision × recall/precision + recall), AUC Area under the curve
- *The DeLong test was used to compare the AUC between MLP model and PHD score [27]. Note: We adjusted the classification threshold to approach the same level of sensitivity as the prediction using the PHD score