Table 6 Results comparison of feature extraction models for skin disease diagnosis.
From: Skin disease diagnostics through federated transfer learning on heterogeneous data
Feature extraction model | Accuracy (%) | Loss | ||
|---|---|---|---|---|
Training | Testing | Training | Testing | |
DenseNet | 87.85 | 81.494 | 0.625 | 0.638 |
VGG19 | 88.713 | 82.28 | 0.645 | 0.658 |
Xception | 89.511 | 83.067 | 0.598 | 0.624 |
UNet | 90.338 | 83.854 | 0.587 | 0.612 |
