Skip to main content

Table 4 On the validation dataset of Brats2018, we applied Wilcoxon Signed-Rank Test to the dice coefficients obtained from the comparison model and the dice scores obtained from Swin Unet3D, respectively

From: Swin Unet3D: a three-dimensional medical image segmentation network combining vision transformer and convolution

model name

P-value

 

ET

TC

WT

3D U-Net

0.2852

0.0133

0.0003

V-Net

0

0

0

UnetR

0

0

0

TransBTS

0.0038

0.0001

0.0559

SwinBTS

0.3714

0.3304

0.6393

AttentionUnet

0

0

0

Swin Pure Unet3D

0

0

0