Table 2 Model performance on datasets with different magnifications.

From: Enhanced digital pathology image recognition via multi-attention mechanisms: the MACC-Net approach

Model

Accuracy

Recall

DSC

10×

20×

40×

10×

20×

40×

10×

20×

40×

U-Net

0.932

0.955

0.968

0.791

0.846

0.889

0.698

0.750

0.783

Att-Unet

0.955

0.862

0.976

0.826

0.835

0.901

0.661

0.796

0.825

TransUnet

0.953

0.968

0.973

0.883

0.904

0.929

0.758

0787

0.809

Swin-Unet

0.956

0.961

0.976

0.876

0.900

0.923

0.782

0.807

0.822

ChannelNet

0.951

0.966

0.979

0.878

0.899

0.921

0.801

0.819

0.841

MACC-Net (Our)

0.961

0.975

0.981

0.887

0.908

0.931

0.803

0.826

0.847

Back to article page