Skip to main content

Table 3 Results of center2 data using the proposed algorithm compared with Cox’s model using AIC, BIC, Lasso, and SCAD

From: Feature selection through validation and un-censoring of endovascular repair survival data for predicting the risk of re-intervention

Algorithm used Number of features p-value (Logrank) CI(SE) Uno’s AUC Sensitivity
Proposed Algorithm 7 0.022 0.63 (0.0739) 0.612 0.73
AIC Algorithm 14 0.034 0.61 (0.0725) 0.60 0.35
BIC Algorithm 14 0.029 0.63 (0.0685) 0.605 0.23
Lasso Algorithm 7 0.0068 0.615 (0.0864) 0.627 0.5
SCAD Algorithm 3 0.2759 0.592 (0.0658) 0.58 0.6923