Skip to main content

Table 1 Visual assessment of CT for test set participants

From: Deep learning parametric response mapping from inspiratory chest CT scans: a new approach for small airway disease screening

 

Total

GT PRMfSAD > 15%

GT PRMfSAD ≤ 15%

No. of subjects

76

46

30

CLE

   

 Total

68 (63.2)

33 (71.7)

15 (50)

 Trace

26 (34.2)

16 (34.8)

10 (33.4)

 Mild

16(21.1)

12 (26)

4 (13.3)

 Moderate

5 (6.6)

5 (10.9)

0 (0)

 Confluent

1 (1.3)

0 (0)

1 (3.3)

 ADE

0 (0)

0 (0)

0 (0)

Panlobular emphysema

   

 No

73 (96.1)

44 (95.7)

29 (96.7)

 Yes

3 (3.9)

2 (4.3)

1 (3.3)

Paraseptal emphysema

   

 Total

21 (27.7)

16 (34.8)

5 (16.7)

 Mild

11 (14.5)

9 (19.6)

2 (6.7)

 Substantial

10 (13.2)

7 (15.2)

3 (10)

Bronchial wall thickening

   

 No

72 (94.7)

44 (95.7)

28 (93.3)

 Yes

4 (5.3)

2 (4.3)

2 (6.7)

Bronchial dilation

   

 Total

24 (31.6)

17 (37)

7 (23.3)

 Single Lung Segment

19 (25)

13 (28.3)

6 (20)

 Multiple Lung Segments

5 (6.6)

4 (8.7)

1 (3.3)

 Extensive Lung Segments

0 (0)

0 (0)

0 (0)

Inflammatory SAD

   

 Total

29 (38.1)

25 (54.3)

4 (13.4)

 Single Lung Segment

19 (25)

17 (36.9)

2 (6.7)

 Multiple Lung Segments

9 (11.8)

7 (15.2)

2 (6.7)

 Extensive Lung Segments

1 (1.3)

1 (2.2)

0 (0)

  1. Data are presented as numbers, with percentages in parentheses
  2. PRM: parametric response mapping; GT: ground truth; fSAD: functional small airway disease; SAD: small airway disease; CLE: centrilobular emphysema; ADE: advanced destructive emphysema