Fig. 1

The LASSO algorithm and tenfold cross-validation were used to extract the optimal subgroup of radiomic features. a The left vertical line indicates the value of the characteristic parameters corresponding to the minimum of λ in cross-validation, and the vertical line on the right represents the parameter value of the more simplified model λ within a standard error. When the value of λ increased to -2.752, it corresponded to the optimal number of radiomic features. b The penalty diagram of the characteristic coefficient of radiomics