Table 5.
Results of the calculation parameters
Number | Image classes included | Sensitivity (%) | Specificity (%) | Precision (%) | ACC | AUC |
---|---|---|---|---|---|---|
35 | COVID/normal | 96 | 79 | 96 | 86 | - |
36 | COVID/influenza pneumonia | 91.54 | 81.19 | 94.92 | - | 0.9524 |
37 | COVID/normal | 100 | 95.23 | 98.70 | 99.02 | 0.9770 |
38 | COVID/non-COVID | 72.59 | 79.76 | 78.67(P)/74.74(N) * | 76.18 | - |
39 | COVID/non-COVID | 98.45 | 99.90 | 98.45 | 99.81 | - |
40 | COVID/normal | 97.8 | 94.7 | 96.3 | 97.4 | 0.969 |
*P means true positives, and N means true negatives.