Risk Prediction Models for Invasive Mechanical Ventilation in Patients with Autoimmune Encephalitis: A Retrospective Cohort Study

Figure 2

Significant clinical risk factors identified by the least absolute shrinkage and selection operator (LASSO) logistic regression model. (a) Minimum criterion (λ. min) and the one standard error (SE) of the minimum criterion (λ. 1se). The dashed vertical line indicates the optimal value. (b) The LASSO synergistic analysis of 25 variables. The curve coefficients were plotted according to a logarithmic series. The best predictor of the model was determined by tenfold cross-validation of the minimum criterion. The optimal λ was defined as the value that produces nonzero coefficients for 10 features.

(a)

(b)