Table 1.
Training analysis (%) of the proposed IoT-enabled deep ensemble model for automated diagnosis of COVID-19 suspected cases on the four-class chest CT dataset when training to testing ratio is 65 : 35.
| Model | Accuracy | F-measure | AUC | Recall | Precision |
|---|---|---|---|---|---|
| JLM [16] | 97.88 | 98.27 | 98.07 | 98.97 | 97.83 |
| WSDL [17] | 98.11 | 98.56 | 98.33 | 99.14 | 97.89 |
| IPCNN [18] | 97.93 | 97.63 | 97.78 | 98.54 | 97.38 |
| DeCNN [19] | 98.45 | 97.53 | 97.99 | 98.89 | 97.82 |
| DLCRD [20] | 98.43 | 97.64 | 98.03 | 98.94 | 97.73 |
| PARL [22] | 98.67 | 97.44 | 98.05 | 98.76 | 98.12 |
| AGGDF [24] | 98.17 | 97.65 | 97.91 | 98.72 | 97.65 |
| GCNN [25] | 98.68 | 97.58 | 98.13 | 98.97 | 98.21 |
| GoogLeNet [53] | 98.16 | 98.35 | 98.25 | 99.08 | 97.62 |
| ResNet152V2 [44] | 98.55 | 98.33 | 98.44 | 99.27 | 97.85 |
| DenseNet201 [34] | 98.57 | 98.18 | 98.34 | 99.09 | 97.83 |
| IRNV2 [3] | 98.18 | 97.48 | 97.83 | 98.67 | 97.56 |
| Proposed | 99.12 | 98.91 | 98.79 | 99.28 | 99.08 |