Fig. 4

Radiomic feature selection using the least absolute shrinkage and selection operator (Lasso) cox regression model. a Tuning parameter (λ) selection in the Lasso logistic model was chosen using tenfold cross-validation based on the minimum criteria. The optimal values of the LASSO tuning parameter (λ) are indicated by the vertical dashed lines, and a value λ of 0.028 with log (λ) = 1.545 was chosen. b LASSO coefficient profiles of the 679 radiomics features. A coefficient profile plot was generated versus the selected log (λ) value using tenfold cross-validation, and 26 radiomics features were selected with non-zero coefficients