Skip to main content

Table 1 Demographics distributions for the corpus under study in terms of patients

From: A radiographic, deep transfer learning framework, adapted to estimate lung opacities from chest x-rays

Patients

All Data

Train

Test

Validation

OOB Test

Patient Count N (%)

17,418 (100)

13,808 (79.27)

1726 (9.99)

1725 (9.90)

159 (0.91)

Image Count N (%)

38,365 (100)

30,586 (79.72)

3660 (9.53)

3833 (9.99)

286 (0.74)

Age Mean (stdev)

60.88 (16.98)

60.89 (16.98)

60.81 (17.30)

60.78 (16.70)

62.44 (15.77)

Sex

 Male (%)

8983 (51.57)

7123 (51.58)

903 (52.31)

873 (50.60)

84 (52.83)

 Female (%)

8435 (48.42)

6685 (48.41)

823 (47.68)

852 (49.39)

75 (47.16)

Patient Status

 IP (%)

10,246 (58.82)

8096 (58.63)

1003 (58.11)

1043 (60.46)

104 (65.4)

 OP (%)

197 (1.13)

158 (1.14)

18 (1.04)

17 (0.98)

4 (2.51)

 ED (%)

6975 (40.04)

5554 (40.22)

705 (40.84)

665 (38.55)

51 (32.07)

COVID-19 Status

 Positive (%)

6692 (38.42)

5259 (38.08)

672 (38.93)

698 (40.46)

63 (39.62)

 Negative (%)

10,725 (61.57)

8548 (61.90)

1054 (61.06)

1027 (59.53)

96 (60.37)

 Unknown (%)

1 (0.005)

1 (0.007)

0 (0.00)

0 (0.00)

0 (0.00)

Race

 Black (%)

3153 (18.10)

2476 (17.93)

322 (18.65)

329 (19.07)

26 (16.35)

 White (%)

7942 (45.59)

6264 (45.36)

802 (46.46)

801 (46.43)

75 (47.16)

 Asian (%)

1349 (7.74)

1074 (7.77)

129 (7.47)

136 (7.88)

10 (6.28)

 Other (%)

4120 (23.65)

3304 (23.92)

391 (22.65)

380 (22.02)

45 (28.30)

 Unknown (%)

854 (4.90)

690 (4.99)

82 (4.75)

79 (4.57)

3 (1.88)

  1. Patient-wise demographics for age, sex, patient status, COVID-19 status, and race.