Pediatric Obesity-Associated Liver Disease: What Role Does Artificial Intelligence Play in Metabolic Dysfunction-Associated Steatotic Liver Disease?

Metabolic dysfunction-associated steatotic liver disease (MASLD) has become the most common chronic liver disease in children and adolescents, paralleling the global increase in pediatric obesity. Despite its growing clinical impact, the diagnosis and staging of pediatric MASLD/MASH remain challenging due to the limitations of current non-invasive tools and the complexity of histopathological evaluation. Liver biopsy remains the reference standard for assessing disease activity and fibrosis stage; however, its invasive nature and the substantial inter-observer variability among pathologists highlight the need for more objective and reproducible approaches. In this review, we summarize the current challenges in the diagnosis and management of pediatric MASLD/MASH and discuss the emerging role of artificial intelligence (AI)-driven models in this field. We explore the application of machine learning and deep learning approaches for non-invasive assessment of hepatic steatosis, as well as their potential to improve digital pathology-based evaluation of steatosis, hepatocellular ballooning, inflammation, and fibrosis. Furthermore, we discuss the opportunities and limitations associated with AI implementation in clinical practice, including algorithmic bias, interpretability, data quality, and the need for external validation.