Skip to main content

Table 1 Patients characteristics in the training, validation and test sets

From: Deep learning for segmentation of the cervical cancer gross tumor volume on magnetic resonance imaging for brachytherapy

 

Training

Validation

Test

Total

117

39

39

Age (years)

   

Mean

53

52

56

Standard deviation

14

15

17

FIGO stage

   

FIGO I

12 (10.2%)

6 (15.4%)

6 (15.4%)

FIGO II

70 (59.8%)

22 (56.4%)

22 (56.4%)

FIGO III

21 (18.0%)

8 (20.5%)

8 (20.5%)

FIGO IV

9 (7.7%)

3 (7.7%)

3 (7.7%)

Unknown

5 (4.3%)

0

0

Volume at first BT fraction

   

Less than 2.8 cc

22 (18.8%)

8 (12.8%)

5 ( 20.5%)

between 2.8 and 4.3 cc

7 (6.0%)

7 (17.9%)

7 (17.9%)

between 4.3 and 12.1 cc

53 (42.3%)

12 (30.8%)

13 (33.3%)

More than 12.1 cc

35 (29.9%)

12 (30.8%)

14 (35.9%)

Histopathological type

   

Squamous cell carcinoma

97 (82.9%)

31 (79.5%)

33 (84.6%)

Adenocarcinoma

16 (13.7%)

6 (15.4%)

4 (10.3%)

Adeno-squamous cell carcinoma

1(0. %)

1 (2.6%)

1 (2.6%)

Non specified/unknown

3 (2.5%)

1 (2.5%)

1 (2.5%)

External beam radiotherapy scheme

   

23 × 2 Gy

95 (81.2%)

29 (74.4%)

32 (82.0%)

25 × 1.8 Gy

22 (18.8%)

10 (25.6%)

7 (18.0%)