Skip to main content
. 2021 Feb 12;17(9):6528–6538. doi: 10.1109/TII.2021.3059023

TABLE VI. Performance of Experiments on Public Data (Mean±Standard Deviation).

Method (%) ACC PRE REC DSC
U-Net [32] 86.78±9.88 57.59±19.67 90.64±10.73 55.69±16.91
U-Net++ [33] 86.92±7.30 50.16±19.43 85.23±9.06 62.34±18.68
DeepLab [34] 82.89±5.29 62.83±7.12 87.49±6.05 43.93±7.58
DI-U-Net [35] 90.67±6.80 61.93±18.88 90.99±7.70 69.54±16.63
Dense U-Net [35] 92.79±4.49 69.80±12.51 93.93±4.88 73.60±11.31
Inf-Net [18] 89.51±6.93 58.90±14.44 87.95±9.01 69.88±12.58
Semi-Inf-Net [18] 92.01±4.71 65.74±12.96 91.46±6.35 74.11±10.87
Mini-Seg [19] 87.53±8.24 56.42±8.73 97.64±3.56 55.97±7.73
COVID-SegNet [20] 86.79±5.09 57.80±12.14 91.52±4.50 50.01±7.98
TV-UNet [21] 92.34±4.78 67.94±12.06 93.67±4.66 72.94±10.60
Proposed Method 93.92±3.87 78.26±9.94 96.39±2.92 75.22±11.25