Fig. 2

Predictors selection based on the least absolute shrinkage and selection operator (LASSO) regression. A LASSO Select Model ([lambda]) of the adjustment parameter by the minimum standard, and tenfold cross-validation. Use log(λ) to plot the area under the receiver operating characteristic curve. Draw a vertical dashed line at the optimal value with one standard error of the minimum standard and 1 standard error of the minimum standard (1-SE standard). B The tuning parameter (lambda) selection in LASSO regression uses tenfold cross-validation. The binomial deviation is plotted on the logarithm (lambda). Use the 1-SE standard to draw a dashed line at the optimal value