Table 4 Contingency table for assessment of the deep-learning algorithm across 11 diagnostic classes on the test set (N = 326)

From: Development and validation of a smartphone-based deep-learning-enabled system to detect middle-ear conditions in otoscopic images

 

“Ground truth” label

Deep-learning prediction

Normal

Wax plug

Eardrum perforation

OME

CRC

Otitis externa

Tympano- sclerosis

AOM

Osteoma

Foreign body

Tympanic graft

TOTAL

Normal

98

0

1

1

0

0

3

1

1

4

0

109

Wax plug

0

67

0

1

1

0

0

0

2

2

0

73

Eardrum perforation

1

0

46

2

0

2

0

3

0

1

2

57

OME

0

0

0

7

1

0

3

0

1

0

1

13

CCR

0

0

2

0

24

0

0

0

0

0

3

29

Otitis externa

0

0

0

0

0

7

0

1

1

0

0

9

Tympanosclerosis

0

0

1

1

0

0

7

0

0

1

0

10

AOM

0

0

0

0

0

0

0

7

0

0

0

7

Osteoma

0

0

0

0

0

0

0

0

8

0

0

8

Foreign body

0

0

0

0

0

0

1

0

0

4

0

5

Tympanic graft

0

0

1

0

0

1

0

0

1

0

3

6

Total

99

67

51

12

26

10

14

12

14

12

9

326

  1. AOM acute otitis media, CCR cavity after cholesteatoma removal, OME otitis media with effusion.