Skin Lesion Segmentation Based on Edge Attention Vnet with Balanced Focal Tversky Loss

Table 2

Performance scores of the proposed model with the latest models in the literature.

Methods	Performance metrics
	Dice	Jaccard	Acc	Sens	Spec
DA-net	0.88	0.80	0.95	0.94	0.95
DRU-net [37]	0.86	0.76	—	0.88	0.92
Deeplabv3+	0.87	0.79	0.95	0.94	0.94
CKDNet [38]	0.88	—	0.95	0.91	0.97
DAGAN [24]	0.89	0.83	0.93	0.95	0.91
FAT-net [31]	0.89	—	0.96	0.91	0.97
Attn_U-net + GN [18]	0.91	0.83	0.95	0.94	0.95
SE_U-net [18]	0.91	0.83	0.95	0.89	0.96
ET-vnet (ours)	0.91	0.83	0.95	0.95	0.96