Table 6.

Performance comparisons between different methods on MosMedData. Bold black text represents the best results.

Methods	$Dice(\%)\uparrow$	$Sens.(\%)\uparrow$	$MAE(\%)\downarrow$
U-Net [11]	$80.39\pm 9.865$	$64.32\pm 19.197$	$8.26\pm 10.959$
UNet+ [9]	$87.27\pm 2.102$	$74.80\pm 7.508$	$1.60\pm 2.223$
Attention U-Net [34]	$87.42\pm 0.395$	$81.92\pm 3.566$	$0.30\pm 0.059$
PSPNet [17]	$82.38\pm 1.048$	$79.11\pm 2.323$	$0.37\pm 0.040$
Deeplabv3 [18]	$83.70\pm 1.166$	$80.01\pm 0.350$	$0.34\pm 0.026$
Inf-Net [6]	$78.64\pm 0.390$	$78.17\pm 4.277$	$0.44\pm 0.015$
SCRN [22]	$87.13\pm 0.070$	$82.42\pm 3.957$	$0.26\pm 0.008$
F3Net [20]	$83.66\pm 0.337$	$82.58\pm 1.121$	$0.33\pm 0.006$
DANet [28]	$84.84\pm 4.435$	$70.65\pm 8.948$	$2.94\pm 4.268$
ACFNet [39]	$79.40\pm 1.114$	$79.07\pm 3.978$	$0.42\pm 0.023$
CE-Net [38]	$86.11\pm 0.371$	$76.17\pm 0.121$	$0.28\pm 0.012$
CPFNet [40]	$86.79\pm 0.266$	$73.50\pm 4.910$	$0.42\pm 0.101$
ResUNet_C${}_2$F(Ours)	$\mathbf{87}.\mathbf{43}\pm \mathbf{0}.\mathbf{165}$	$\mathbf{85}.\mathbf{64}\pm \mathbf{3}.\mathbf{013}$	$\mathbf{0}.\mathbf{253}\pm \mathbf{0}.\mathbf{006}$