Table 8.

Results of BCR-UNet and other methods on IOSTAR dataset.

Models	ACC	SEN	SPE	AUC	F1	IOU	MCC
NFN+ (37)	0.9683	0.7921	0.9812	0.9803	-	-	-
SCS-Net (38)	0.9706	0.8255	0.9830	0.9865	-	-	-
U-Net (5)	0.9714	0.7642	0.9894	0.9873	0.8104	0.6813	0.7967
Attention UNet (41)	0.9701	0.7711	0.9874	0.9865	0.8049	0.6735	0.7896
SD-UNet (24)	0.9717	0.7835	0.9881	0.9880	0.8159	0.6890	0.8014
MultiResUNet (27)	0.9712	0.7795	0.9879	0.9832	0.8125	0.6842	0.7978
DRNet (20)	0.9717	0.8191	0.9850	0.9880	0.8223	0.6983	0.8070
BCR-UNet	0.9727	0.7965	0.9880	0.9882	0.8234	0.6999	0.8091

Bold values is the highest scores for the metrics.