BPG is committed to discovery and dissemination of knowledge
Retrospective Study
Copyright: ©Author(s) 2026.
World J Gastroenterol. Jan 21, 2027; 33(3): 124159
Published online Jan 21, 2027. doi: 10.3748/wjg.124159
Figure 2
Figure 2 Feature variable selection using least absolute shrinkage and selection operator regression in the Training cohort. A: The Y-axis shows binomial deviance, the lower X-axis shows log(λ), and the upper X-axis indicates the number of non-zero coefficients. The left dashed vertical line marks lambda.min (λ = 0.037), the value that minimizes deviance; the right dashed line marks lambda.1se (λ = 7.94 × 10-4), selected by the one-standard-error rule; B: A least absolute shrinkage and selection operator coefficient profile of all feature variables against the log(λ) sequence is shown. The blue dotted vertical lines indicate, respectively, the log(λmin) and log(λ 1se). The resulting eight predictors with non-zero coefficients were identified based on the log(λ 1se) value.


Write to the Help Desk