Skip to main content

Table 2 Comparisons with the state-of-the-art baselines on the CVC-clinicDB dataset

From: P-TransUNet: an improved parallel network for medical image segmentation

Method

mDice

mIou

Recall

Precision

U-Net [5]

0.8781

0.7881

0.7865

0.9329

Deeplabv3+  [37]

0.8897

0.8706

0.9251

0.9366

PraNet [30]

0.8990

0.8490

–

–

U-Net++  [6]

0.9035

0.8637

0.9175

0.8564

ResUNet++  [31]

0.9199

0.8892

0.9391

0.8445

TransUNet [15]

0.9350

0.8870

–

–

FANet [34]

0.9355

0.8937

0.9339

0.9401

DS-TransUNet [20]

0.9422

0.8939

0.9500

0.9369

FCBFormer [36]

0.9461

0.9020

0.9502

0.9412

Our method

0.9593

0.9142

0.9564

0.9537

  1. Terms. The “−” denotes that the corresponding result is not provided. For each column, the best results are highlighted
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us