Table 4 Comparisons with state-of-the-art models on complexity analysis.

From: Multi-scale conv-attention U-Net for medical image segmentation

Method

FLOPs (M)

Params (M)

Inference time (ms)

U-net5

300

31.04

287

Attention U-Net7

1100

33.90

312

DualNet19

2500

39.9

502

MedT26

1710

20.23

732

TransUNet22

1590

25.31

577

Unext12

10

1.77

25

EGE-UNet8

2

0.05

351

WRANet13

1100

34.4

305

SeTformer32

2100

22.71

610

Ours

850

7.32

343

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