Table 5.
Comparison of the proposed method with previous studies.
| Study | Attributes of dataset | Feature extraction and classification methods | Type of Images | Accuracy (%) |
|---|---|---|---|---|
| [65] | 125 COVID-19 500 Healthy 500 Pneumonia |
DarkCovidNet | Xray Images | 87.02 |
| [66] | 25 COVID-19 25 Healthy |
COVIDX-Net | Xray Images | 90.0 |
| [67] | 224 COVID-19 504 Healthy 700 Pneumonia |
VGG-19 | Xray Images | 93.48 |
| [68] | 777 COVID-19 708 Healthy |
DRE-Net | CT Images | 86 |
| [69] | 25 COVID-19 25 Healthy |
ResNet50þ SVM | Xray Images | 95.38 |
| [70] | 313 COVID-19 229 Healthy |
UNetþ3D Deep Network | CT Images | 90.8 |
| [46] | 53 COVID-19 5526 Healthy |
COVID-Net | 92.4 | |
| [71] | 219 COVID-19 175 Healthy 224 Pneumonia |
ResNet þ Location Attention | CT Images | 86.7 |
| [30] | 195 COVID-19 258 Healthy |
M-Inception | CT Images | 82.9 |
| [72] | 449 patients with COVID-19, 425 normal ones, 98 with lung cancer, 98 with lung cancer | Multitask learning | CT images | 94.67 |
| [73] | 320 COVID-19 320 Healthy |
Deep convolutional neural network | CT images | 93.64 |
| [74] | 100 COVID-19 200 Healthy 322 Pneumonia |
Convolutional neural network | Xray Images | 95.74 |
| [75] | 80 COVID-19 72 Healthy 78 Pneumonia |
Convolutional neural network | CT images | 96.2 |
| Our Proposed method | 135 COVID-19, 150 Healthy 150 Pneumonia | F-transform, MKLBP and SVM | Xray Images | 97.01 |