丙型病毒性肝炎肝硬化失代偿期患者发生细菌感染的列线图预测模型构建及评价
Construction and evaluation of a nomogram prediction model for bacterial infection in patients with decompensated hepatitis C cirrhosis
Note： A. Path diagram of regression coefficient. The upper abscissas was the number of variables with non-zero coefficients in the model at this time, the lower abscissas was the logarithm of the penalty coefficient (λ), and the ordinate was the value of the coefficient. B. Cross-verification curve of LASSO regression. The upper and lower abscissas were the same as Fig A, and the ordinate was likelihood bias. The dotted line on the left of Fig B indicates the number of variables corresponding to the minimum λ (when the model has the highest fitting effect), and the number of variables was 14. The dotted line on the right indicates one standard error of the least λ (when the model has better fitting effect, fewer and simpler variables are included), and the number of variables was 7.
