PROTOCOL: Predictive Models Based on Artificial Intelligence for Adverse Events in Patients With Chronic Kidney Disease Undergoing Haemodialysis: A Scoping Review

Background: Chronic kidney disease is an increasing global health concern, and haemodialysis is the most common treatment in its advanced stages. However, haemodialysis involves significant risks - such as intradialytic hypotension, infections, and metabolic imbalances - that compromise patient safety. The growing complexity and volume of clinical data in dialysis settings challenge traditional risk-management approaches. In this context, artificial-intelligence-based predictive models have emerged as tools to anticipate adverse events and support timely clinical action. Objective: To identify and map the existing evidence on artificial-intelligence-based predictive models for adverse events in patients with chronic kidney disease undergoing haemodialysis. Methods: This scoping review will be conducted following the Joanna Briggs Institute methodology for scoping reviews and structured with the Population, Concept, and Context framework. A systematic search will be performed in MEDLINE, Embase, CINAHL, Scopus, Web of Science, and LILACS, together with grey literature in Google Scholar, with no date or language restrictions. Two independent reviewers will screen the records, select the studies, and extract the data, with disagreements resolved by a third reviewer. The findings will be presented descriptively - through tables, charts, and a narrative summary - and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews.