Skip to main content

Table 2 Characteristics of the training set and test set

From: Prenatal prediction of neonatal respiratory morbidity: a radiomics method based on imbalanced few-shot fetal lung ultrasound images

Characteristics

Training set (n = 167)

Test set (n = 43)

Normal

NRM

p value

Normal

NRM

p value

No. of images

127

40

–

32

11

 

GA*

36.49 ± 0.85

34.37 ± 2.42

< 0.0001

36.78 ± 1.64

34.53 ± 2.37

< 0.0001

Birth weight (g)*

3096 ± 385

2978 ± 490

< 0.0001

3145 ± 423

3024 ± 540

< 0.0001

GDM

  

< 0.005

  

0.06

Yes

48 (37.80%)

26 (65.00%)

–

10 (31.25%)

7 (63.64%)

–

No

79 (62.20%)

14 (35.00%)

–

22 (68.75%)

4 (36.36%)

–

Mode of delivery

  

0.35

  

0.94

Spontaneous vaginal delivery

56 (44.09%)

21 (52.50%)

–

15 (46.88%)

5 (45.45%)

–

Caesarean delivery

71 (55.91%)

19 (47.50%)

–

17 (53.12%)

6 (54.55%)

–

Sex of newborn

  

0.87

  

0.43

Female

59 (46.46%)

18 (45.00%)

–

16 (50.00%)

7 (63.64%)

–

Male

68 (53.54%)

22 (55.00%)

–

16 (50.00%)

4 (36.36%)

–

Apgar

  

–

  

–

5 min ≤ 7

0(0.00%)

4 (10.00%)

–

0 (0.00%)

0 (0.00%)

–

5 min > 7

1 27 (100.00%)

36 (90.00%)

–

32 (100.00%)

11 (100.00%)

–

  1. The  p value < 0.05 is shown in blod
  2. The t test was performed for continuous variables and the χ2 test was performed for categorical variables
  3. GA gestational age, GDM gestational diabetes mellitus
  4. *Data are means ± standard deviations