Table 5 Comparison of classification performance by hot spot size in “BUS+CEUS”. All units except AUROC are in percent. Values in parentheses are 95 % confidence interval
Hot spot size (mm) | 2 | 3 | 4 | 5 | Entire tumor |
|---|---|---|---|---|---|
Number of selected features | 7 | 13 | 12 | 12 | 13 |
AUROC | 0.875 (0.795 - 0.936) | 0.865 (0.795 - 0.926) | 0.862 (0.791 - 0.928) | 0.865 (0.791 - 0.934) | 0.868 (0.791 - 0.934) |
ACC | 0.816 (0.750 - 0.883) | 0.805 (0.750 - 0.867) | 0.803 (0.750 - 0.867) | 0.814 (0.750 - 0.867) | 0.809 (0.750 - 0.867) |
Precision (PPV) | 0.841 (0.764 - 0.905) | 0.838 (0.765 - 0.900) | 0.834 (0.764 - 0.897) | 0.842 (0.769 - 0.905) | 0.835 (0.774 - 0.889) |
Recall (Sensitivity) | 0.920 (0.814 - 1.000) | 0.907 (0.814 - 1.000) | 0.909 (0.814 - 1.000) | 0.915 (0.814 - 1.000) | 0.918 (0.814 - 1.000) |
F1 | 0.877 (0.828 - 0.921) | 0.869 (0.822 - 0.913) | 0.869 (0.824 - 0.911) | 0.876 (0.825 - 0.913) | 0.873 (0.828 - 0.913) |
Specificity | 0.550 (0.235 - 0.765) | 0.548 (0.235 - 0.765) | 0.536 (0.235 - 0.765) | 0.558 (0.294 - 0.765) | 0.535 (0.294 - 0.706) |
NPV | 0.749 (0.562 - 1.000) | 0.718 (0.550 - 1.000) | 0.720 (0.555 - 1.000) | 0.740 (0.556 - 1.000) | 0.737 (0.562 - 1.000) |