Table 1 Baseline characteristics of the datasets.

From: Development and validation of a deep learning image quality feedback system for infant fundus photography

Characteristics

Internal training set

Internal validation set

Internal test set

External validation set

Gestational age (w)

39.1 ± 1.4

39.1 ± 1.2

38.1 ± 2.2

38.6 ± 1.5

Birth weight (g)

3446.8 ± 473.0

3449.5 ± 482.6

3038.5 ± 570.5

3205.5 ± 516.5

Subjects

774

209

209

152

Poor quality

532

146

144

380

 DC

138

19

35

99

 OF

122

50

35

138

 IPD

272

77

74

143

Adequate quality

2531

628

655

414

 DC

776

191

226

158

 OF

671

157

190

127

 IPD

1084

280

239

129

Excellent quality

4678

1298

1288

678

  1. Values are presented as the means ± standard deviations at baseline in different groups. DC, defective contact; OF, out of focus; IPD, inadequate pupil dilatation.