Noninvasive Evaluation of Liver Fibrosis Reverse Using Artificial Neural Network Model for Chronic Hepatitis B Patients

Figure 1

Flowchart of predicting liver fibrosis reverse factors and analytic method. (a) Summary of enrolled patients and relevant variables of reverse in our chronic HBV-induced fibrosis antiviral treatment cohort. (b) At the end of treatment, fifty-five patients reversed according to the Ishak scoring system. (c) Assessing predictors in the ANN model. Seven statistically different variables were pointed as the input layer, and the outcome was liver fibrosis reverse. (d) Evaluating diagnosis efficacy by AUC, sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio.