Artificial Intelligence in Managing Liver Cirrhosis and Variceal Bleeding: A Review

Abstract

Context: Liver cirrhosis (LC) represents a major driver of mortality on a global scale, with upper gastrointestinal (GI) bleeding (UGIB) considerably increasing its related mortality risk. The objective of this review is to investigate the applicability of artificial intelligence (AI) and machine learning (ML) in managing LC and its complications, particularly esophageal variceal bleeding (EVB). Evidence Acquisition: This study was performed by searching electronic databases and search engines from 2014 to December 2024, thereby including articles that examined the effects of AI on patients with LC bleeding. Results: This review synthesizes findings from multiple studies to highlight the limitations of current scoring systems and summarizes the latest progress of AI and ML in detecting esophageal/gastric varices (EV/GV), diagnosing liver fibrosis (LF) and LC, and predicting the prognosis and complications in patients with LC. Conclusions: Overall, AI and ML offer more precise and personalized decision support for managing LC. Future research should focus on optimizing models and conducting multi-center validations to ensure their clinical reliability and generalizability.