Table 1.
The general characteristics of the recruited studies.
| Study Number | First Author | Publication Year | Country | Study Type |
Dataset | Deep Learning Model | All Data | COVID | Non-COVID | Sensitivity | Specificity |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Borkowski [29] | 2020 | United States of America |
Case control |
COVID-19/non- COVID pneumonia/ COVID-19/non-COVID pneumonia/normal |
Microsoft CustomVision |
1000 | 500 | 500 | 100 | 95 |
| 2 | Zokaeinikoo [30] | 2021 | United States of America |
Case control |
COVID-19/non- COVID infection/ normal |
AIDCOV using VGG-16 |
5801 | 269 | 5532 | 99.3 | 99.98 |
| 3 | Keidar [31] | 2021 | Israel | Retrospective | COVID- 19/normal |
RetNet50 | 2427 | 360 | 2067 | 87.1 | 92.4 |
| 4 | Ahmed [32] | 2021 | Japan | Case control |
COVID/non- COVID |
HRNet | 1410 | 410 | 1000 | 98.53 | 98.52 |
| 5 | Kikkisetti [33] | 2020 | United States of America |
Retrospective | COVID/bacterial pneumonia/ viral pneumonia/ normal |
VGG-16 | 2031 | 445 | 1586 | 79 | 93 |
| 6 | Shibly [34] | 2020 | Bangladesh | Case control |
COVID/non- COVID |
Faster R-CNN | 19,250 | 283 | 18,967 | 97.65 | 95.48 |
| 7 | Gomes [35] | 2020 | Brazil | Case control |
COVID- 19/bacterial and viral pneumonia |
IKONOS | 6320 | 464 | 5856 | 97.7 | 99.3 |
| 8 | Ko [36] | 2020 | South Korea |
Case Control |
COVID- 19/pneumonia/ normal |
DarkNet-19 | 1125 | 125 | 1000 | 95.13 | 95.3 |
| 9 | Sharma [37] | 2020 | United States of America |
Case control |
COVID-19/non COVID-19 |
Residual Att Net |
239 | 120 | 119 | 100 | 96 |