Table 6 Comparison of computational complexity in terms of flops and GPU across various segmentation models.

From: Multimodal feature-optimized approaches for cancer classification using microarray gene expression analysis

Method

FLOPs (G)

GPU (M)

UNet

102.27

3877

AttUNet

104.00

2027

TransUNet

50.39

4918

SegNeXt

2.33

2657

CariesAttNet

14.44

4535

AIMACGD-SFST

0.79

1200