Skip to main content

Table 2 Slide-level accuracy of the test set of the different models

From: Implementing deep learning models for the classification of Echinococcus multilocularis infection in human liver tissue

Slide_ID

Ground trutha

DL neural models

VGG19_bn

Squeezenet

Resnet18

0

0

0.974

0.991

1.000

1

0

1.000

1.000

1.000

2

1

0.818

0.819

0.839

3

1

0.956

0.932

0.946

4

1

0.937

0.919

0.901

5

1

0.798

0.683

0.792

6

1

0.806

0.799

0.823

7

0

0.998

0.997

0.997

8

1

0.487

0.505

0.475

9

1

0.817

0.808

0.850

10

1

0.822

0.818

0.833

11

0

0.997

0.999

1.000

12

0

0.999

0.999

1.000

13

1

0.948

0.934

0.943

14

1

0.593

0.581

0.578

15

0

0.995

0.995

0.998

16

1

0.375

0.392

0.371

17

1

0.717

0.718

0.703

18

0

1.000

0.999

1.000

  1. a0: normal; 1: echinococcus