English

Learning from Complementary Ultrasound Representations for Liver Disease Classification

Computer Vision and Pattern Recognition 2026-07-13 v1 Machine Learning

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.

Cite

@article{arxiv.2607.12062,
  title  = {Learning from Complementary Ultrasound Representations for Liver Disease Classification},
  author = {Sabahattin Mert Daloglu and Gokce Bekar and Ceren Coskun and Senanur Sahin and Harvey Castro and Soner Hacihaliloglu and Halley P. Letter and Ilker Hacihaliloglu},
  journal= {arXiv preprint arXiv:2607.12062},
  year   = {2026}
}

Comments

Submitted to the MICCAI 2026 ASMUS Workshop (under review)