Table 13.
Three class (normal, nCOVID-19 and pneumonia) performance comparison of the proposed method with the state of art methods.
| Articles | Class | Algorithms/techniques | Total number of images | ACC (%) |
|---|---|---|---|---|
| Ozturk et al. (2020) | 3 | DarkNet | 87.02 | |
| L. Wang et al. (2020) | 3 | COVID-Net | 93.30 | |
| Abbas et al. (2020) | 3 | DeTraC | 196 | 95.12 |
| Chowdhury et al. (2020) | 3 | AlexNet, ResNet18, DenseNet201, SqueezeNet | 3487 | 97.94 |
| Ucar and Korkmaz (2020) | 3 | Deep Bayes-SqueezeNe | 5957 | 98.26 |
| Nour et al. (2020) | 3 | CNN, SVM, DT, KNN | 3670 | 98.97 |
| Toğaçar et al. (2020) | 3 | MobileNetV2, SqueezeNet, SVM | 458 | 99.27 |
| Proposed Method (Overall) | 3 | Majority vote based classifier ensemble | 2346 | 93.41 |