Fig. 2
From: A CT-based radiomics nomogram for distinguishing between benign and malignant bone tumours
Texture features selected by the LASSO regression model. a Tuning parameter (Ī») selection in the LASSO model. The top value represents the corresponding characteristic number. In this study, the optimal Ī» value corresponding to the perpendicular line was selected to obtain 10 features with non-zero coefficients. b Variation of LASSO coefficients for different features as modulation parameter (Ī» value) changes. c Contributions of the 10 selected features with nonzero coefficients to the radiomics signature, with their respective coefficient values