Research Article
A Medical Image Classification Model Based on Adversarial Lesion Enhancement
Table 2
The experimental results.
| Backbone | With LE | Accuracy | Precision | Recall | F1-score | Kappa |
| VGG-19 | | 82.8 | 76.82 | 94.47 | 84.73 | 65.5 | 85.59 | 81.11 | 93.19 | 86.73 | 71.13 | ResNet-50 | | 81.72 | 77.78 | 89.36 | 83.17 | 63.38 | 83.23 | 78.97 | 91.06 | 84.59 | 66.39 | DenseNet-121 | | 81.51 | 78.54 | 87.23 | 82.66 | 62.96 | 83.01 | 82.77 | 83.83 | 83.3 | 66.01 | Inception-v3 | | 82.15 | 78.36 | 89.36 | 83.5 | 64.24 | 83.01 | 78.68 | 91.06 | 84.42 | 65.96 |
|
|