MHA-Net: A Multibranch Hybrid Attention Network for Medical Image Segmentation

Table 1

AUC, Se, and Ac Values on the DRIVE dataset by difference methods.

Method	Year	AUC	Se	Ac
U-Net [5]	2015	0.9834	0.8059	0.9627
JSPL-Net [25]	2018	0.9752	0.7653	0.9542
Ce-Net [16]	2019	0.9831	0.8330	0.9542
IterNet [26]	2020	0.9813	0.7791	0.9574
AGAC-Net [27]	2020	0.9847	0.7941	0.9558
DAP [28]	2021	0.9788	0.8227	0.9545
VSSC-Net [10]	2021	0.9789	0.7827	0.9627
Our approach	2022	0.9864	0.8412	0.9636