Figure 1.

The selection of predictive variables by LASSO regression analysis with 5-fold cross-validation. Notes: (a) Log (Lambda) value of the 27 features in the LASSO model. A coefficient profile plot was produced against the log (lambda) sequence. (b) The optimal parameter (lambda) selection in the LASSO model adopted 5-fold cross-validation based on minimum criteria. The minimum criteria and 1-SE criteria were used to draw a vertical dashed line with the optimal values. LASSO = least absolute shrinkage and selection operator, SE = standard error.