Table 2.
The Prognostic Value Comparison Between Logisic Regression and Xgboost Algorism
| AUC | Sensitivity | Specificity | Youden Index | Accuracy | FPR | FNR | PPV | NPV | F Score | |
|---|---|---|---|---|---|---|---|---|---|---|
| Mortality | ||||||||||
| Logisic regression | 0.767 | 0.837 | 0.750 | 0.587 | 0.830 | 0.116 | 0.400 | 0.545 | 0.905 | 0.860 |
| Xgboost algorithm | 0.950 | 1.000 | 0.900 | 0.900 | 0.981 | 0.000 | 0.100 | 1.000 | 0.977 | 0.990 |
| GOS (1–3) | ||||||||||
| Logisic regression | 0.829 | 0.836 | 0.756 | 0.592 | 0.764 | 0.098 | 0.422 | 0.813 | 0.743 | 0.803 |
| Xgboost algorithm | 0.958 | 0.984 | 0.933 | 0.917 | 0.962 | 0.016 | 0.067 | 0.977 | 0.952 | 0.976 |
Abbreviations: AUC, area under the receiver operating characteristics curve; FPR, false positive rate; FNR, false negative rate; PPV, positive predictive value; NPV, negative predictive value; GOS, Glasgow outcome scale.