Intelligent system for fibrosis stage prediction in chronic hepatitis C patients, combining ultrasonic transient elastography, usual ultrasonographic parameters and laboratory tests

Abstract

Background and aims: Currently, the golden standard in evaluating fibrosis stage in chronic hepatitis C (CHC) patients is liver biopsy, which is a costly and invasive procedure. On the other hand, transient elastography (FibroScan®) performs well in identifying severe fibrosis or cirrhosis, but is less accurate in identifying lower degrees of fibrosis. Our aim was to build a predictive model for staging liver fibrosis in CHC patients by means of artificial intelligence (AI), adding to liver stiffness measurement (LSM) several non-invasive approaches (basic ultrasonografic parameters and routine laboratory tests).