Research Article
Transfer Learning to Detect COVID-19 Automatically from X-Ray Images Using Convolutional Neural Networks
Table 7
General comparison of the best CNN obtained with state-of-the-art method.
| Study | Method used | Database size | Acc% |
| [10] | VGG-19 | 224 COVID-19, 504 healthy, and 700 pneumonia | 93.48 | [10] | MobileNet v2 | 224 COVID-19, 504 healthy, and 700 pneumonia | 94.72 | [21] | DenseNet201 | 423 COVID-19, 1341 healthy, and 1345 viral pneumonia | 97.94 | [38] | CapsNet | 231 COVID-19, 1050 healthy, and 1050 pneumonia | 84.22 | [37] | COVID-Net | 358 COVID-19, 8066 healthy, and 5538 pneumonia | 93.3 | [13] | DarkCOVIDNet | 157 COVID-19, 500 healthy, and 500 pneumonia | 87.02 | [22] | CoroNet | 157 COVID-19, 500 healthy, and 500 pneumonia [13] | 90.21 | [22] | CoroNet | 284 COVID-19, 310 healthy, and 657 pneumonia | 95 | Our work | VGG16 | 423 COVID-19, 1341 healthy, and 1345 viral pneumonia [21] | 96.79 | 1200 COVID-19, 1341 healthy, and 1345 viral pneumonia | 98.29 |
|
|