Table 7.
The influences of different algorithms on the evaluation indices.
| Method | Acc | Recall | Specificity |
|---|---|---|---|
| Kermany (Inception V3) (Kermany, 2018) | 0.966 | 0.978 | 0.974 |
| VGG16 (Simonyan and Zisserman, 2015) | 0.939 | 1 | 0.908 |
| ResNet50 (He et al., 2016) | 0.967 | 0.996 | 0.948 |
| Hassan (Hassan et al., 2021) | 0.986 | 0.983 | 0.993 |
| Kaymak (Kaymak and Serener, 2018) | 0.971 | 0.996 | 0.984 |
| Hwang (Hwang et al., 2019) | 0.969 | Not Mentioned | Not Mentioned |
| Hard vote | 0.979 | 0.979 | 0.993 |
| Soft vote | 0.98 | 0.98 | 0.994 |
| FN-F1-OCT | 0.98 | 0.98 | 0.994 |
| FN-Weight-OCT | 0.984 | 0.984 | 0.995 |
| FN-Auto-OCT | 0.987 | 0.987 | 0.996 |