Table 5.
Accuracy, precision, recall, and F1 score of tested models for AKI prediction.
| AKI prediction | ||||
|---|---|---|---|---|
| Model | Accuracy (%) | Precision (%) | Recall (%) | F1 score |
| FTTransformer31 | 82.12 | 51.37 | 39.59 | 0.4156 |
| TabTransformer30 | 84.39 | 56.91 | 61.20 | 0.5890 |
| TabNet40 | 82.05 | 45.37 | 14.40 | 0.2071 |
| XGBoost32 | 88.05 | 68.40 | 64.79 | 0.6653 |
| LightGBM33 | 87.91 | 75.75 | 50.80 | 0.6053 |
| CatBoost34 | 87.17 | 67.79 | 57.20 | 0.6200 |
| Random forest39 | 86.15 | 69.79 | 43.20 | 0.5334 |
| Logistic regression | 81.97 | 51.73 | 21.19 | 0.3001 |
| XGBoost with22 tunning | 83.07 | 54.35 | 42.79 | 0.4777 |
| Random forest with22 tunning | 83.29 | 58.35 | 31.60 | 0.4090 |
Significant values are in bold.