Fig. 2

Radiomics feature selection using LASSO binary logistic regression and the performance of the radiomics signature. A Tuning parameter (λ) selection in the LASSO model used tenfold cross-validation via minimum criteria. The binomial deviance was plotted versus log (λ). The dotted vertical lines were drawn at the optimal λ values based on the minimum criteria and 1 standard error of the minimum criteria. The optimal λ value of 0.0454 with log (λ) =  − 3.093 was selected. B LASSO coefficient profiles of the 961 radiomics features. The dotted vertical line was drawn at the λ value of 0.0454, where optimal λ resulted in 16 nonzero coefficients. Plots C present the ROC curves of the radiomics signature in the 66 patients, respectively. D Waterfall plot for distribution of radiomics score and response to steroid of individual patients. Green bars show scores for patients who experienced progression, while blue bars show scores for those who exhibited radiological improvement