Skip to main content
. 2019 Nov 28;9:1330. doi: 10.3389/fonc.2019.01330

Figure 4.

Figure 4

Features selection using the LASSO algorithm. (A) Selection of the tuning parameter (Lambda) in the LASSO model using 5-fold cross-validation. Binomial deviances from the LASSO regression cross-validation model were plotted as a function of log(Lambda). The dotted vertical line at the right was drawn at the optimal value based on the minimum criteria and the 1-standard error rule (the 1-SE criteria). An optimal Lambda value of 0.102 with log(Lambda) = −2.280 and 5 non-zero coefficients were selected. (B) LASSO coefficient profiles of the 10 texture features. A vertical line was drawn at the optimal value selected using the 5-fold cross-validation process in (A). The 5 features with non-zero coefficients were included to construct the radiomics signature.