Learning from Complementary Ultrasound Representations for Liver Disease Classification

2026-07-13

AI Summary

This study addresses the clinical challenge of differentiating non-alcoholic steatohepatitis (NASH) from simple steatosis (NAFLD), which is difficult using conventional B-mode ultrasound due to subtle tissue differences. The authors propose a novel classification framework that systematically integrates multiple complementary ultrasound representations—B-mode, physics-guided imaging, and local phase images—acquired from the same ultrasound scan. By leveraging a self-supervised masked autoencoder (MAE) combined with graph convolutional networks (GCNs), the method substantially enhances diagnostic discrimination. Evaluated across multiple independent cohorts, the approach achieves up to a 32.4% improvement in accuracy and a 91.2% gain in F1 score, demonstrating the generalizability and clinical promise of multimodal ultrasound representations for precise NASH/NAFLD subtyping.

Abstract

Differentiating non-alcoholic steatohepatitis (NASH) from non-alcoholic fatty liver disease (NAFLD) using ultrasound remains challenging due to subtle tissue alterations and the limited information available in conventional B-mode imaging. In this work, we investigate whether complementary ultrasound representations derived from the same acquisition can improve NASH versus NAFLD classification. Specifically, we combine conventional B-mode ultrasound with physics-guided and local phase-based image representations and evaluate their effectiveness using self-supervised masked autoencoders (MAEs) and graph convolutional networks (GCNs). Experiments were conducted on a multi-site Mayo Clinic cohort consisting of 2,547 liver ultrasound scans from 125 patients. Compared with conventional B-mode ultrasound alone, complementary ultrasound representations consistently improved classification performance, yielding gains of up to 32.4% in accuracy and 91.2% in F1-score. Furthermore, performance improvements were consistently observed across age groups, sex, race, ethnicity,and acquisition sites.

Problem

• NASH
• NAFLD
• ultrasound
• liver disease classification
• B-mode imaging

Innovation

• complementary ultrasound representations
• masked autoencoders
• graph convolutional networks
• NASH classification
• physics-guided imaging