Fig. 3

Feature selection was performed using the Minimum Absolute shrink and selection operator (LASSO) binary logistic regression model. (A) The adjustment parameter (λ) in the LASSO model is selected by minimum criteria using 10 × cross-validation. The relationship between the area under the subject operating characteristic (AUC) curve and log (λ) is plotted. Draw a vertical dotted line at the best value using 1 standard error of the minimum standard and the minimum standard (1-SE standard). According to tenfold cross-validation, the standard error of the minimum distance λ is 0.058. (B) Lasso coefficient profile for each texture feature. Generate coefficient profiles from logarithmic (λ) sequences. A vertical line is drawn at the selected value using × 10 cross validation, where the best λ yields 6 filter variables.