Table 2 Patient characteristics.

From: Integrating radiological and clinical data for clinically significant prostate cancer detection with machine learning techniques

Variable

Development cohort n = 3839

Validation cohort n = 960

CsPCa n = 1607 (41.9%)

Non-CsPCa n = 2232 (58.1%)

p-value

CsPCa n = 394 (41.1%)

Non-CsPCa n = 566 (58.9%)

p-value

Age

71 (65,76)

66 (60,71)

< 0.001

70 (65,77)

66 (61,72)

< 0.001

PSA

8.0 (5.6,13.5)

6.3 (4.7,9.0)

< 0.001

8.7 (5.7 14.9)

6.2 (4.5,9.0)

< 0.001

PV

45 (34, 62)

62 (44,85)

< 0.001

47 (35,65)

64 (46,86)

< 0.001

PSAD

0.18 (0.12,0.32)

0.10 (0.07,0.16)

< 0.001

0.19 (0.12,0.33)

0.10 (0.07,0.15)

< 0.001

DRE (suspicious)

699 (43.5%)

368 (16.5%)

< 0.001

186 (47.2%)

99 (17.5%)

< 0.001

FH (yes)

125 (7.8%)

148 (6.6%)

0.172

36 (9.1%)

32 (5.7%)

0.038

PNB (yes)

392 (24.4%)

768 (34.4%)

< 0.001

87 (22.1%)

205 (36.2%)

< 0.001

PI-RADS

  

< 0.001

  

< 0.001

1

51(3.2%)

393 (17.6%)

 

9 (2.3%)

96 (17.0%)

 

2

21 (1.3%)

117 (5.2%)

 

2 (0.5%)

28 (4.9%)

 

3

163 (10.1%)

801 (35.9%)

 

36 (9.1%)

207 (36.6%)

 

4

746 (46.4%)

782 (35.0%)

 

190 (48.2%)

207 (36.6%)

 

5

626 (39.0%)

139 (6.3%)

 

157 (39.9%)

28 (4.9%)

< 0.001

  1. CsPCa clinically significant prostate cancer, PSA prostate specific antigen, PV prostate volume, PSAD PSA density, DRE digital rectal exam findings, FH family history of PCa. PNB previous negative biopsy.
Back to article page