Table 6 Computational efficiency comparison showing FLOPs (in Giga operations), number of parameters, frames per second (FPS), and average F1 score across models.

From: Multi scale deep learning quantifies Ki67 index in breast cancer histopathology images

Model

FLOPs (G)

Parameters

FPS

Avg. F1 (%)

U-Net19

54.63

31,037,763

21

78.63

UNet++21

82.00

36,629,763

14.01

78.96

PathoNet17

5.24

3,228,603

85

82.26

TransUNe29

66.48

34,878,703

17

83.69

TransAttUnet30

49.80

17,864,948

23

84.47

Kpi-Net

22.20

15,894,955

68

85.79

  1. The best results in each column are highlighted in bold.
Back to article page