Skip to main content

Table 1 Clinical characteristics of training and validation sets

From: T2-weighted MRI-based radiomics for discriminating between benign and borderline epithelial ovarian tumors: a multicenter study

Cohort

Training set (n = 309)

Internal validation set (n = 78)

External validation set (n = 30)

p value

Age (years)

45.49 ± 16.06

46.94 ± 17.46

44.70 ± 19.17

0.740

Menopausal status

   

0.581

 Postmenopausal

154 (49.8)

39 (50.0)

12 (40.0)

 

 Premenopausal

155 (50.2)

39 (50.0)

18 (60.0)

 

Parity

   

0.219

 Multipara

256 (82.8)

64 (82.1)

21 (70.0)

 

 Nullipara

53 (17.2)

14 (17.9)

9 (30.0)

 

Abdominal symptoms

   

0.362

 Pain and distention

135 (43.7)

37 (47.4)

17 (56.7)

 

 None

174 (56.3)

41 (52.6)

13 (43.3)

 

CA125 (U/mL)

   

0.118

 > 35

78 (25.2)

14 (17.9)

11 (36.7)

 

 ≥ 35

231 (74.8)

64 (82.1)

19 (63.3)

 

HE4a (pmol/L)

   

1.000

 Abnormal

20 (6.5)

5 (6.4)

2 (6.7)

 

 Normal

289 (93.5)

73 (93.6)

28 (93.3)

 

Final diagnosis

   

0.549

 Benign

224 (72.5)

57 (73.1)

19 (63.3)

 

 Borderline

85 (27.5)

21 (26.9)

11 (36.7)

 

Histopathology

   

0.080

 Serous

148 (47.9)

40 (51.3)

21 (70.0)

 

 Mucinous

141 (45.6)

33 (42.3)

6 (20.0)

 

 Others

20 (6.5)

5 (6.4)

3 (10.0)

 
  1. Data are presented as mean ± standard deviation for normally distributed continuous variables, median (interquartile range, IQR) for non-normally distributed continuous variables, or number (%) for categorical variables. HE4, human epididymis protein 4; CA125, carbohydrate antigen 125
  2. aNormal value of HE4: postmenopausal woman < 121 pmol/L or premenopausal woman < 92.1 pmol/L