A comprehensive narrative review of artificial intelligence use in the diagnosis and management of metabolic dysfunction-associated steatotic liver disease

Background and Objective: Metabolic dysfunction-associated steatotic liver disease (MASLD) is the leading cause of chronic liver disease, affecting approximately 30 percent of the population worldwide. Despite this high prevalence, the disease remains underdiagnosed, partially due to the low sensitivity of non-invasive tests (NITs) and reliance on invasive liver biopsies. This review aims to summarize the current literature regarding the role of artificial intelligence (AI) in optimizing the diagnosis and management of MASLD. Methods: We conducted a review of literature using the PubMed/MEDLINE database for English-language articles published from January 2005 through December 2025. The search focused on AI applications in MASLD, including machine learning (ML), deep learning (DL), and natural language processing (NLP). Key Content and Findings: AI tools can improve the diagnosis of MASLD from already existing data-laboratory results, radiology reports, magnetic resonance imaging (MRI) scans, and histopathology slides-by utilizing methods such as NLP. Beyond diagnosis, AI can predict critical outcomes, such as hepatic decompensation and mortality. Additionally, it plays an important role in digital therapeutics and mobile health interventions that can subsequently improve the clinical trajectory. Conclusions: AI holds the potential to transform MASLD care by improving diagnostic accuracy and personalizing management. However, widespread implementation will require addressing challenges related to data safety, standardization and validation in the general population.