Skip to main content
Fig. 1 | BMC Surgery

Fig. 1

From: A novel nomogram to predict the risk of anastomotic leakage in patients after oesophagectomy

Fig. 1

Predictor selection using a LASSO logistic regression model. a Shrinkage estimation parameter: fivefold cross validation (CV) is the most common repeated sampling method. According to the one standard error (SE) rule, glmnet usually recommends lambda.1se instead of lambda.min to identify the model with the best performance but with the fewest number of independent variables. b Dotted vertical lines were drawn at the optimal values by using the minimum criteria (lambda.min) and 1 SE of the minimum criteria (lambda.1se)

Back to article page