Skip to main content
. 2021 Nov 25;124:108452. doi: 10.1016/j.patcog.2021.108452

Table 7.

Performance comparisons between different methods on UESTC-COVID-19. Bold black text represents the best results.

Methods Dice(%) Sens.(%) MAE(%)
U-Net [11] 85.48±0.125 75.90±3.466 0.47±0.060
UNet+ [9] 85.13±0.296 76.41±1.636 0.47±0.010
Attention U-Net [34] 84.62±0.539 77.04±2.484 0.50±0.035
PSPNet [17] 82.67±0.248 76.45±1.246 0.56±0.006
Deeplabv3 [18] 80.13±1.212 70.88±2.573 0.63±0.049
Inf-Net [6] 83.26±0.440 77.45±1.810 0.54±0.015
SCRN [22] 83.78±0.402 77.75±2.949 0.52±0.017
F3Net [20] 83.58±0.974 78.56±1.178 0.53±0.026
DANet [28] 85.40±0.745 79.40±2.175 0.47±0.026
ACFNet [39] 84.31±0.435 79.83±1.799 0.50±0.020
CE-Net [38] 85.45±0.420 77.38±7.962 0.47±0.016
CPFNet [40] 85.36±0.182 77.35±1.151 0.47±0.015
ResUNet_C2F(Ours) 85.52±0.081 79.46±2.286 0.47±0.006