Table 6 Experimental results and comparison on DRD dataset
Paper ID | dataset | Disease labels | Method used | Accuracy |
|---|---|---|---|---|
[52] | DRD | No DR, Mild NPDR, Moderate NPDR, Severe NPDR, PDR | ResNet50(Direct training) | 0.888 |
ResNet101(Direct training) | 0.889 | |||
ViT_B_16 | 0.905 | |||
ViT_L_32 | 0.914 | |||
This paper | DRD | No DR, Mild NPDR, Moderate NPDR, Severe NPDR, PDR | ResNet50(Transfer learning) | 0.916 |
ResNet101(Transfer learning) | 0.922 | |||
ResNet50 + ResNet101 + ID-SET | 0.927 |