Table 3.
Quantitative segmentation results on Coronacases (Ma et al., 2021). The best, second best, and third best results are marked by italic, boldface, and underline, respectively.
Method | DSC | SEN | SPE | PPV | VOE | RVD |
---|---|---|---|---|---|---|
U-Net (Ronneberger et al., 2015) | 0.5184 | 0.3925 | 0.9848 | 0.8829 | 0.6501 | 0.5463 |
U-Net+ (Zhou et al., 2018) | 0.5371 | 0.4231 | 0.9850 | 0.8470 | 0.6328 | 0.4788 |
UNet (Qin et al., 2020) | 0.5311 | 0.4134 | 0.9851 | 0.8690 | 0.6384 | 0.5129 |
U-Net3+ (Huang et al., 2020) | 0.6097 | 0.4869 | 0.9871 | 0.8883 | 0.5614 | 0.4441 |
CE-Net (Gu et al., 2019) | 0.6712 | 0.5501 | 0.9885 | 0.8905 | 0.4948 | 0.3903 |
nCoVSegNet (Ours) | 0.6894 | 0.5918 | 0.9886 | 0.8949 | 0.4739 | 0.3270 |