Skip to main content
. 2022 Aug 29;22(17):6501. doi: 10.3390/s22176501

Table 2.

Complexity of models used in comparison process.

Methods Training Time (h) Prediction Time (s) Parameters
Our method 26.4 55.7 52.3 mln
3D U-Net 15.2 76.2 19 mln
3D Res-Net 24.4 65.1 46.4 mln
SwinBTS 13.2 58.7 16.3 mln
VT U-Net 15.8 59.8 18.4 mln
3D U-Net++ 25.6 64.5 51.6 mln
Attention 3D U-Net 18.3 121.4 22.3 mln
3D Res-U-Net 29.2 69.3 58.6 mln
3D DeepLabV3 35.8 78.6 76.1 mln
3D PSP-Net 36.5 96.9 72.4. mln