Fig. 1

Plot illustrating the feature selection process using the LASSO regression analysis. (a) Variability in coefficient characteristics of the parameters identified as risk factors using LASSO regression analysis. The horizontal axis represents the log(lambda) values, where lambda is the regularization parameter. As lambda increases, the degree of shrinkage increases, resulting in more coefficients being reduced to zero. The vertical axis shows the coefficients of the features included in the model. Each of the colored lines corresponds to a different feature. (b) Selection process of optimal parameters in the LASSO regression model through cross-validation methods. The point where the dashed vertical line intersects the curves indicates the optimal lambda value, selected through 10-fold cross-validation.