Abstract: Hemodialysis is one of the crucial treatments for end stage renal disease. However, there are still issues to be solved, such as higher mortality, various dialysis-related complications and increased demands for chronic disease management. These situations highlight urgent necessities to enhance risk prediction,  precise diagnosis of complication, and comprehensive management of the patients. The rapid development of artificial intelligence and its gradual integration into hemodialysis have provided great potentials to address the existing issues. This review aims to outline the current application status of artificial intelligence in prediction and diagnosis of hemodialysis complications and to elaborate its development directions in patient management based on existed achievements and trends.

Key words: Artificial intelligence, Hemodialysis, Intradialytic hypotension, Arteriovenous fistula, Large language model