Research Article
Improved Classification of White Blood Cells with the Generative Adversarial Network and Deep Convolutional Neural Network
Table 8
10-fold cross-validation accuracy of the DNN models initialized using pretrained weights.
| Model | GAN_aug1 (100 inst./class) (%) | GAN_aug2 (150 inst./class) (%) | GAN_aug3 (200 inst./class) (%) |
| VGG-16 | 92.3 | 93.0 | 94.1 | VGG-19 | 93.3 | 93.7 | 95.0 | ResNet-18 | 92.9 | 93.7 | 94.2 | ResNet-50 | 94.7 | 95.5 | 95.8 | DenseNet-121 | 95.4 | 96.2 | 97.2 | DenseNet-169 | 96.1 | 96.9 | 97.2 |
|
|