Skip to main content

Table 2 Data analysis results: prediction logrank statistics and concordance indices.

From: Incorporating higher-order representative features improves prediction in network-based cancer prognosis analysis

  Logrank statistic Concordance index
Data R1 R2 R3 R4 R1 R2 R3 R4
D1 15.30 19.10 34.70 0.18 0.74 0.70 0.77 0.50
D2 0.25 4.56 0.60 0.46 0.61 0.67 0.51 0.58
D3 10.40 0.32 19.00 2.14 0.62 0.53 0.69 0.55
D4 3.89 13.80 11.40 0.01 0.63 0.66 0.64 0.54
D5 7.95 7.50 7.50 12.30 0.72 0.70 0.70 0.76
D6 6.27 2.15 6.46 7.99 0.65 0.61 0.69 0.73
  1. Larger logrank statistics and concordance indices correspond to more predictive power. A logrank statistic greater than 3.84 is significant at the 0.05 level.