Hegazy MAE. Artificial intelligence in metabolic dysfunction-associated steatotic liver disease: Machine learning for non-invasive diagnosis and risk stratification. World J Hepatol 2025; 17(11): 111354 [DOI: 10.4254/wjh.v17.i11.111354]
Corresponding Author of This Article
Mona Abd-Elmonem Hegazy, MD, Department of Internal Medicine, Division of Hepatology and Gastroenterology, Kasr Aliny Hospital, Faculty of Medicine, Cairo University, Kasr Alainy Street, Garden City, Cairo 12556, Egypt. monahegazy@cu.edu.eg
Research Domain of This Article
Gastroenterology & Hepatology
Article-Type of This Article
Minireviews
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
World J Hepatol. Nov 27, 2025; 17(11): 111354 Published online Nov 27, 2025. doi: 10.4254/wjh.v17.i11.111354
Artificial intelligence in metabolic dysfunction-associated steatotic liver disease: Machine learning for non-invasive diagnosis and risk stratification
Mona Abd-Elmonem Hegazy
Mona Abd-Elmonem Hegazy, Department of Internal Medicine, Division of Hepatology and Gastroenterology, Kasr Aliny Hospital, Faculty of Medicine, Cairo University, Cairo 12556, Egypt
Author contributions: Hegazy MA design and wrote the whole mini review.
Conflict-of-interest statement: The author declares no potential conflict of interest with respect to the research, authorship, and/or publication of this article.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Mona Abd-Elmonem Hegazy, MD, Department of Internal Medicine, Division of Hepatology and Gastroenterology, Kasr Aliny Hospital, Faculty of Medicine, Cairo University, Kasr Alainy Street, Garden City, Cairo 12556, Egypt. monahegazy@cu.edu.eg
Received: June 30, 2025 Revised: July 26, 2025 Accepted: October 27, 2025 Published online: November 27, 2025 Processing time: 152 Days and 18.2 Hours
Abstract
Metabolic dysfunction-associated steatotic liver disease (MASLD), formerly known as non-alcoholic fatty liver disease, represents a growing global health burden, contributing significantly to liver-related morbidity and mortality. Early detection and timely intervention are essential to prevent disease progression. Conventional diagnostic methods, which rely on specialized imaging and invasive liver biopsies, underscore the need for non-invasive, cost-effective alternatives. Artificial intelligence—particularly machine learning and deep learning—has emerged as a transformative tool in MASLD diagnostics, offering improved accuracy in risk prediction, imaging interpretation, and disease stratification. This review synthesizes recent advancements in AI-based MASLD diagnostics, highlighting key models, performance metrics, and clinical applications, while addressing ongoing challenges such as data standardization, interpretability, and clinical validation.
Core Tip: Artificial intelligence (AI), machine learning and deep learning, holds transformative potential for the non-invasive diagnosis and risk stratification of metabolic dysfunction-associated steatotic liver disease (MASLD). These AI models utilize readily available clinical data, biomarkers, and imaging modalities (ultrasound, computed tomography, magnetic resonance imaging) to detect steatosis, predict disease risk, and stage fibrosis with greater accuracy than conventional scoring systems such as fibrosis-4. Despite their promising performance, several challenges hinder widespread clinical adoption, including the need for data standardization, rigorous prospective validation, model interpretability, and seamless integration into existing healthcare workflows. Overcoming these barriers is essential to fully harness AI potential in improving MASLD diagnosis and management.
