Table 4 Models trained on ECG and art line, 12-h gap
From: Continuous sepsis trajectory prediction using tensor-reduced physiological signals
Model | Rank | F1 score | Recall | Specificity | AUROC |
|---|---|---|---|---|---|
LUCCK | 1 | 0.45 (0.08) | 0.73 (0.14) | 0.69 (0.14) | 0.72 (0.08) |
2 | 0.45 (0.07) | 0.72 (0.13) | 0.71 (0.11) | 0.72 (0.07) | |
3 | 0.47 (0.07) | 0.75 (0.11) | 0.71 (0.10) | 0.73 (0.07) | |
4 | 0.46 (0.08) | 0.73 (0.13) | 0.71 (0.11) | 0.71 (0.08) | |
None | 0.44 (0.06) | 0.77 (0.12) | 0.67 (0.10) | 0.72 (0.06) | |
RF | 1 | 0.42 (0.08) | 0.70 (0.13) | 0.68 (0.12) | 0.68 (0.09) |
2 | 0.44 (0.07) | 0.71 (0.12) | 0.69 (0.10) | 0.70 (0.08) | |
3 | 0.47 (0.07) | 0.74 (0.12) | 0.72 (0.10) | 0.72 (0.08) | |
4 | 0.48 (0.08) | 0.75 (0.10) | 0.72 (0.10) | 0.74 (0.07) | |
None | 0.44 (0.08) | 0.76 (0.12) | 0.67 (0.10) | 0.72 (0.07) | |
SVM | 1 | 0.32 (0.07) | 0.57 (0.17) | 0.61 (0.15) | 0.49 (0.10) |
2 | 0.32 (0.08) | 0.57 (0.18) | 0.62 (0.15) | 0.51 (0.12) | |
3 | 0.33 (0.08) | 0.57 (0.16) | 0.61 (0.16) | 0.51 (0.11) | |
4 | 0.33 (0.07) | 0.54 (0.15) | 0.61 (0.15) | 0.51 (0.12) | |
None | 0.36 (0.07) | 0.65 (0.16) | 0.61 (0.15) | 0.56 (0.12) |
