Advanced Strategies for Dietary Recommendations in Liver Disease: A Comprehensive Literature Review

Abstract

The global prevalence of non-alcoholic fatty liver disease (NAFLD) driven by unhealthy dietary habits and genetic factors is closely associated with obesity, diabetes, and high cholesterol and can progress to life-threatening cirrhosis. This presents a significant health challenge, especially as no approved medications or targeted treatments currently exist for NAFLD, emphasizing the critical role of diet and exercise in managing the disease. This paper presents a comprehensive review combining both literature analysis and expert consultation to explore advanced methodologies for developing a food recommendation system tailored to liver disease patients. Approaches such as Machine Learning (ML), Deep Learning (DL), and Ontology-based AI were systematically evaluated, with the ontology-based approach identified as the most effective. Insights from the literature were confirmed through expert consultation, highhigh-lighting parameters like BMI, blood sugar levels, disease stage, and food preferences are closely related to liver disease and crucial for providing personalized dietary recommendations. The review also highlights limitations in existing systems such as inadequate expert knowledge integration and insufficient attention to individual dietary needs. Future work aims to develop a comprehensive ontology-based food recommendation system, leveraging insights from both the literature and expert consultation to improve patient outcomes and quality of life.