Skip to main content

Table 1 Continuous variables contained within the train and test sets

From: Machine learning algorithms to estimate 10-Year survival in patients with bone metastases due to prostate cancer: toward a disease-specific survival estimation tool

Variable by Time Point

Median (IQR)

P value*

Bayes Factor

Whole Cohort

Train Set

Test Set

Proximal PSA

 1-Year

33.4 (200)

38.6 (202)

23.7 (153)

0.37

0.15

 2-Year

35.7 (205)

30.5 (147)

51.7 (294)

0.59

0.14

 3-Year

36.7 (211)

34.0 (205)

46.2 (238)

0.38

0.15

 4-Year

36.8 (217)

36.8 (232)

37.4 (202)

0.43

0.15

 5-Year

39.6 (236)

42.8 (232)

31.5 (242)

0.54

0.14

 10-Year

42.8 (252)

44.5 (275)

40.7 (131)

0.21

0.17

Age

 1-Year

71.0 (12.7)

71.1 (12.6)

70.1 (12.9)

0.41

0.19

 2-Year

71.0 (12.7)

71.2 (12.6)

70.6 (12.4)

0.96

0.13

 3-Year

71.0 (12.7)

71.0 (12.9)

71.1 (11.7)

0.95

0.13

 4-Year

71.0 (12.8)

71.0 (12.7)

71.6 (13.4)

0.32

0.22

 5-Year

71.0 (12.9)

70.8 (12.7)

71.5 (13.1)

0.46

0.18

 10-Year

71.0 (13.0)

70.8 (12.6)

72.5 (13.7)

0.21

0.32

  1. PSA prostate-specific antigen
  2. *P values determined using Pearson’s chi-squared test