Skip to main content

Table 4 Summary of the changes in the decision-making of radiologists before and after AI assistance

From: Enhancing prediction of supraspinatus/infraspinatus tendon complex injuries through integration of deep visual features and clinical information: a multicenter two-round assessment study

 

Physician

True negative

True positive

Sensitivity (%)

Specificity (%)

Internal validation cohort (n = 89)

1

32 → 40

33 → 34

0.660 → 0.790*

0.821 → 0.869

2

36 → 41

18 → 33

0.581 → 0.804*

0.621 → 0.854*

3

37 → 42

30 → 32

0.741 → 0.820

0.787 → 0.840

4

45 → 44

29 → 31

0.878 → 0.961

0.804 → 0.830

5

42 → 40

28 → 33

0.800 → 0.785

0.778 → 0.851*

External validation cohort (n = 57)

6

22 → 28

15 → 18

0.577 → 0.750*

0.742 → 0.848*

7

19 → 25

17 → 17

0.531 → 0.654*

0.760 → 0.806

8

26 → 30

17 → 18

0.680 → 0.818*

0.812 → 0.857

9

28 → 29

18 → 18

0.750 → 0.783

0.848 → 0.853

10

25 → 25

15 → 19

0.625 → 0.678

0.757 → 0.862*

  1. *Indicates the p < 0.05