Table 2 Comparison of evaluation metrics using advanced methods on the DRIVE dataset.
From: Retina image segmentation using the three-path Unet model
Model | Year | Dice | SE | SP | Acc |
|---|---|---|---|---|---|
Unet10 | 2015 | 0.8157 | 0.7924 | 0.9781 | 0.9544 |
Bukenya et al.15 | 2018 | Â | 0.7490 | 0.9740 | 0.9540 |
ReUnet11 | 2019 | 0.8155 | 0.7751 | 0.9816 | 0.9556 |
R2Unet11 | 2019 | 0.8171 | 0.7792 | 0.9813 | 0.9556 |
Zou et al.13 | 2021 | 0.8129 | 0.7761 | 0.9792 | 0.9519 |
Du et al.23 | 2021 | Â | 0.7814 | 0.9810 | 0.9556 |
ContextUnet24 | 2021 | 0.8135 | 0.7961 | 0.9778 | 0.9547 |
CSUnet24 | 2021 | 0.8251 | 0.8071 | 0.9782 | 0.9565 |
Tariq et al.25 | 2022 | Â | 0.8125 | 0.9763 | 0.9610 |
Li et al.26 | 2022 | 0.7994 | 0.7882 | 0.9815 | 0.9657 |
LSW-Net27 | 2022 | 0.8216 | 0.7876 | 0.9837 | 0.9565 |
ResDO-Unet28 | 2023 | 0.8229 | 0.7985 | 0.9791 | 0.9561 |
Wang et al.29 | 2023 | 0.8687 | 0.7380 | 0.9703 | 0.9403 |
Ours | Â | 0.8291 | 0.8184 | 0.9773 | 0.9571 |
