Poster
A Machine Learning Model Based on Liver Histology Predicts the Hepatic Venous Pressure Gradient in Patients With Compensated Cirrhosis Due to Nonalcoholic Steatohepatitis
AASLD 2020
Study Background
♦ The hepatic venous pressure gradient (HVPG) reliably measures portal pressure, including the presence of clinically significant portal hypertension (CSPH; HVPG ≥10 mm Hg)
♦ CSPH is associated with an increased risk of hepatic decompensation and mortality
♦ Measurement and interpretation of HVPG require expertise, with up to 30% of HVPG readings deemed inaccurate
♦ We previously utilized a machine learning (ML) research platform (PathAI, Inc., Boston, MA) to quantify fibrosis and other histologic features of nonalcoholic steatohepatitis (NASH), and characterize the heterogeneity of fibrosis in patients with cirrhosis
♦ CSPH is associated with an increased risk of hepatic decompensation and mortality
♦ Measurement and interpretation of HVPG require expertise, with up to 30% of HVPG readings deemed inaccurate
♦ We previously utilized a machine learning (ML) research platform (PathAI, Inc., Boston, MA) to quantify fibrosis and other histologic features of nonalcoholic steatohepatitis (NASH), and characterize the heterogeneity of fibrosis in patients with cirrhosis
Authors
Bosch et al.