Table 2. Classification results: COVID-19 vs. non-COVID-19 CT slices.
Test | TrainCOV | TrainCOV | TrainCOV | TrainCOV & TrainPUB | TrainCOV & TrainPUB | |
---|---|---|---|---|---|---|
- | TrainCOV | TrainCOV & TrainPUB | - | TrainCOV & TrainPUB | ||
TestCOV & TestPUB | ACC | 99.18% | 99.52% | 99.83% | 99.88% | 99.89% |
Sens | 0.8980 | 0.8980 | 0.9286 | 0.9490 | 0.9388 | |
Spec | 0.9928 | 0.9963 | 0.9991 | 0.9993 | 0.9996 | |
PPV | 0.5789 | 0.7273 | 0.9192 | 0.9394 | 0.9583 | |
TestCOV | ACC | 89.33% | 91.33% | 94.67% | 92.67% | 93.33% |
Sens | 0.8980 | 0.8980 | 0.9286 | 0.9490 | 0.9388 | |
Spec | 0.8846 | 0.9423 | 0.9808 | 0.8846 | 0.9231 | |
PPV | 0.9362 | 0.9670 | 0.9891 | 0.9394 | 0.9583 | |
TestPUB | Spec | 0.9935 | 0.9966 | 0.9992 | 1.0000 | 1.0000 |
ACC: accuracy, Sens: sensitivity, Spec: specificity, PPV: positive predictive value