Table 3 A comparison of the LAug method was carried out while keeping a consistent architecture.

From: Multi-view hybrid encoder U-Net for 3D renal vascular medical image segmentation

Encoder Arch.

Lower encoder

Loss

FPR

Recall

DSC

MSD

CNNs + CNNs

Ese Vovnet39b32

0.4478

0.043

0.894

0.923

0.826

Convnext Small34

0.4350

0.032

0.890

0.926

0.829

DM Nfnet F035

0.4313

0.048

0.916

0.933

0.837

CNNs + ViTNets

Sam2 Hiera Small38

0.4367

0.042

0.887

0.920

0.831

Swin Tiny39

0.4334

0.033

0.905

0.934

0.836

 

Proposed (CoaT Lite Small)41

0.4323

0.028

0.891

0.928

0.832

  1. Significant values are in bold.
Back to article page