Table 1 Summary of demographics, clinical characteristics and image dataset information stratified by disease.

From: Large-scale investigation of deep learning approaches for ventilated lung segmentation using multi-nuclear hyperpolarized gas MRI

Disease

Total number of scans

Number of patients

Number of HP gas scans

Sex*

Median (range) age*

Mean ± SD ventilated lung volume (liters)*

3He

129Xe

Male

Female

Healthy

43

33

1

42

15

13

12 (9, 76)

3.78 ± 1.18

Asthma

169

81

4

165

28

52

50 (13, 73)

4.23 ± 1.03

Asthma/COPD overlap

11

5

0

11

0

5

56 (45, 67)

4.13 ± 0.68

Bronchiectasis

3

3

1

2

1

1

15 (9, 29)

3.76 ± 1.00

CF

247

58

134

113

29

28

16 (6, 48)

3.65 ± 1.05

COPD

62

23

56

6

4

5

64 (52, 80)

4.43 ± 0.71

Non-IPF ILD**

77

41

0

77

25

16

69 (39, 83)

3.78 ± 0.80

Investigation for possible airways disease

38

21

5

33

2

16

49 (36, 69)

3.89 ± 1.05

IPF

46

20

45

1

17

3

72 (52, 80)

3.87 ± 0.71

Lung cancer

22

16

14

8

10

6

69 (34, 85)

4.12 ± 0.86

Preterm birth

41

40

4

37

15

25

12 (9, 14)

2.75 ± 0.55

  1. HP hyperpolarized, CF cystic fibrosis, COPD chronic obstructive pulmonary disease, ILD interstitial lung disease, IPF idiopathic pulmonary fibrosis, SD standard deviation.
  2. *Data for 25 patients was unavailable.
  3. **Contains connective tissue disease-associated ILD (CTD-ILD), hypersensitivity pneumonitis and drug-induced ILD (DI-ILD).
Back to article page