Fig. 3

Kaplan–Meier survival analysis for ten and four-feature models in testing dataset. a In-hospital survival is analyzed by the ten-feature classification, where features are selected based on an alpha setting of 0.20 for LASSO regression. b In-hospital survival is analyzed by the four-feature classification, where features are selected based on an alpha setting of 0.25 for LASSO regression. Significance is determined by the log-rank test (p < 0.05)