Table 1 Baseline characteristics of participants who provided images for the data sets.

From: Deep learning algorithms for detecting and visualising intussusception on plain abdominal radiography in children: a retrospective multicenter study

 

Positive images

Negative images

P-value

Set A (n = 3843)

Images, n

318

3525

 

Participants, n

161

1760

 

Age, months, mean [s.d.]

21.8 [12.0]

24.0 [17.6]

0.12

Sex, male, n (%)

101 (62.7)

1207 (68.6)

0.13

Set B (n = 5926)

Images, n

716

5210

 

Participants, n

361

2615

 

Age, months, mean [s.d.]

22.2 [17.9]

32.8 [17.9]

 < 0.001*

Sex, male

228 (63.2)

1461 (55.9)

0.01*

Set C (n = 1615)

Images, n

415

1200

 

Participants, n

208

602

 

Age, months, mean [s.d.]

20.9 [16.8]

31.1 [23.8]

 < 0.001*

Sex, male

136 (65.4)

305 (50.7)

 < 0.001*

All (n = 11,384)

Images, n

1449

9935

 

Participants, n

730

4977

 

Age, months, mean [s.d.]

21.7 [15.8]

29.5 [19.3]

 < 0.001*

Sex, male

465 (63.7)

2973 (59.7)

0.04*

  1. Continuous variables are presented by mean [standard deviation] and categorical variables are presented by N (%), p < 0.05.
  2. The independent t-test or the Kruskal–Wallis test were used to compare positive and negative groups according to normality. Categorical variables were presented as numbers and percentages and analysed using a chi-squared test.
  3. *P-values < 0.05 were considered statistically significant.
Back to article page