Deep Learning for Liver Disease Analysis Using MRI: A Scoping Review of Current Methodologies and Future Directions

With the advancement in Artificial Intelligence in recent years, the application of deep learning techniques within the medical domain has become an increasingly researched area. Among these applications, the intervention of deep learning in liver disease diagnosis using MRI has seen an increased interest in recent years. Despite the growing volume of studies, there remains a need to systematically map the existing literature to guide the new researchers. The present study undertakes a scoping review of research on liver disease related deep learning applications using MRI published in and between 2015 and 2025 to determine the extent these researches have undertaken, the different types of liver diseases they have researched, publication trends, specific diagnosis steps that deep learning was applied to, methodologies explored, deep learning architecture trends, issues encountered, existing gaps, and future direction of utilizing deep learning in liver disease classification using MRI. By synthesizing issues encountered and identifying existing gaps, this study will provide a roadmap for future research directions in developing non-invasive, AI-enhanced diagnostic tools.