Skip to main content

Table 2 MIoU and dice of different models in three datasets

From: PyConvU-Net: a lightweight and multiscale network for biomedical image segmentation

 

kaggleLung

liver

ISBICell

MIoU

Dice

MIoU

Dice

MIoU

Dice

U-Net

0.7279

0.7834

0.6207

0.7386

0.7742

0.8639

UNet++

0.6078

0.6471

0.6504

0.7690

0.7878

0.8808

Resnet34_UNet

0.9494

0.9721

0.6623

0.7451

0.8398

0.9115

Attention U-Net

0.7723

0.8278

0.6989

0.8083

0.8269

0.8945

FCN8s

0.9545

0.9752

0.5139

0.6447

0.8345

0.9005

PyConvU-Net

0.9630

0.9339

0.7050

0.8227

0.8385

0.9117

  1. Bold numbers indicate the best performance
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us