Table 1 Clinical and pathological characteristics of all datasets

From: Deep learning CT model for stratified diagnosis of pancreatic cystic neoplasms: multicenter development, validation, and real-world clinical impact

Characteristics

Training set

Validation set

Test 1

Test 2

Test 3

 

(n = 999)

(n = 250)

(n = 173)

(n = 289)

(n = 124)

Age (years), Median (IQR)

56.0(23.0)

55.0(23.3)

57.0(25.0)

56.0(20.5)

63.0(12.0)

Sex, n(%)

Female

560(56)

147(58.8)

96(55.4)

172(59.5)

53(42.7)

Male

439(44)

103(41.2)

77(44.6)

117(40.5)

71(57.3)

BMI (kg/m2), Mean ± SD

23.0 ± 3.3

22.7 ± 3.2

22.7 ± 3.5

23.2 ± 3.7

22.4 ± 3.2

CA19-9 μmol/L, Median (IQR)

9.1(18.4)

9.3(17.0)

7.9(20.5)

10.8(15.4)

13.4(19.6)

CEA (ng/ml), Median (IQR)

2.0(2.0)

1.9(1.8)

1.7(2.2)

0.8(1.4)

2.2(2.1)

Abdominal pain, n(%)

350(35.0)

99(39.6)

58(33.5)

84(29.0)

27(21.7)

Jaundice, n(%)

41(4.1)

9(3.6)

7(4.0)

5(1.7)

4(3.2)

Diabetes, n(%)

145(14.5)

19(7.6)

26(15)

30(10.3)

8(6.4)

Alcohol consumption, n(%)

80(8.0)

22(8.8)

15(8.6)

33(11.4)

7(5.6)

Family history, n(%)

18(1.8)

3(1.2)

1(0.5)

12(4.1)

0(0.0)

Pathological classification, n(%)

Precancerous IPMN

299(29.9)

90(36.0)

50(28.9)

74(25.6)

57(46.0)

Malignant IPMN

214(21.4)

39(15.6)

40(23.1)

30(10.4)

37(29.8)

Precancerous MCN

72(7.2)

18(7.2)

13(7.5)

47(16.3)

6(4.8)

Malignant MCN

7(0.7)

1(0.4)

1(0.6)

1(0.3)

1(0.8)

SCN

212 (21.2)

53(21.2)

37(21.3)

84(29.0)

10(8.0)

SPN

195 (19.5)

49(19.6)

32(18.4)

53(18.3)

13(10.4)

  1. Note: Categorical variables are expressed as the number of cases (%), and continuous variables as the median (interquartile range).
  2. IPMN intraductal papillary mucinous cystic neoplasms, MCN mucinous cystic neoplasms, SCN serous cystic neoplasms, SPN solid pseudopapillary neoplasms.