Figure 3.

Feature selection using the LASSO binary regression model to identify independent factors of cardiac injury. (A) Tuning parameter (λ) selection in the LASSO model used tenfold cross-validation via minimum criteria. The binomial deviance was plotted versus log (λ). Dotted vertical lines were drawn at the optimal values by using minimum criteria and the 1 standard error (1-SE criteria). (B) LASSO coefficient profiles of the 15 features. A coefficient profile plot was produced against the log(λ) sequence. Dotted vertical line was drawn at the optimal λ at minimum criteria and 1 standard error (1-SE criteria). The model at 1-SE criteria was selected as the final model with 4 nonzero coefficients including age, leukocytes, D-dimer, and serum ferritin. LASSO, least absolute shrinkage and selection operator.