Generative Adversarial Network Combined with SE-ResNet and Dilated Inception Block for Segmenting Retinal Vessels

Table 4

The results of SAD-GAN and other algorithms on DRIVE dataset.

Methods	Year	SE	SP	ACC	ROC_AUC	PR_AUC
Wu et al. [40]	2018	0.7844	0.9819	0.9567	0.9807	—
Wang et al. [41]	2019	0.7940	0.9816	0.9567	0.9772	—
Cheng et al. [42]	2019	0.7672	0.9834	0.9559	0.9793	—
Yue et al. [43]	2019	0.8199	0.9762	0.9561	0.9796	—
Li et al. [44]	2020	0.7890	0.9799	0.9557	0.9774	0.8519
Ma et al. [45]	2020	0.7875	0.9813	0.9566	0.9794	—
Yang et al. [46]	2020	0.8340	0.9820	0.9560	0.9786	0.8821
Samuel and Veeramalai [47]	2020	0.7827	0.9821	0.9627	0.9789
Li et al. [48]	2021	0.7931	0.9896	0.9698	0.9738	—
Xu et al. [49]	2021	0.8320	0.9885	0.9590	0.9713
Lin et al. [21]	2021	0.8361	0.9740	0.9563	0.9805
Yang et al. [22]	2022	0.8115	0.9780	0.9568	0.9810	—
Zhai et al. [23]	2022	0.7982	0.9818	0.9571	0.9811	—
Our proposed		0.8422	0.9825	0.9567	0.9813	0.8928