Table 5 Models trained on ECG, art line, and EHR data, 6-h gap.
From: Continuous sepsis trajectory prediction using tensor-reduced physiological signals
Model | Rank | F1 score | Recall | Specificity | AUROC |
|---|---|---|---|---|---|
LUCCK | 1 | 0.53 (0.06) | 0.74 (0.11) | 0.70 (0.09) | 0.74 (0.07) |
2 | 0.53 (0.06) | 0.74 (0.10) | 0.69 (0.09) | 0.74 (0.07) | |
3 | 0.53 (0.07) | 0.74 (0.10) | 0.70 (0.09) | 0.73 (0.07) | |
4 | 0.53 (0.07) | 0.74 (0.11) | 0.70 (0.08) | 0.73 (0.07) | |
None | 0.52 (0.06) | 0.73 (0.10) | 0.69 (0.08) | 0.72 (0.07) | |
RF | 1 | 0.57 (0.07) | 0.76 (0.09) | 0.73 (0.09) | 0.76 (0.06) |
2 | 0.57 (0.06) | 0.77 (0.08) | 0.72 (0.08) | 0.77 (0.06) | |
3 | 0.58 (0.07) | 0.77 (0.09) | 0.74 (0.08) | 0.77 (0.06) | |
4 | 0.58 (0.07) | 0.78 (0.09) | 0.73 (0.09) | 0.77 (0.06) | |
None | 0.53 (0.06) | 0.76 (0.09) | 0.69 (0.08) | 0.74 (0.06) | |
SVM | 1 | 0.49 (0.07) | 0.70 (0.11) | 0.67 (0.10) | 0.67 (0.08) |
2 | 0.45 (0.08) | 0.66 (0.13) | 0.63 (0.12) | 0.62 (0.10) | |
3 | 0.44 (0.07) | 0.65 (0.12) | 0.63 (0.12) | 0.60 (0.10) | |
4 | 0.43 (0.07) | 0.63 (0.12) | 0.63 (0.11) | 0.58 (0.09) | |
None | 0.52 (0.06) | 0.74 (0.10) | 0.69 (0.08) | 0.72 (0.06) |
