Duygu TUTAN, Barış ESER, İbrahim DOĞAN, Nilay BEKTAŞ AKPINAR
Turkish Journal of Nephrology - 2026;35(3):234-243
Background: This study aimed to analyze global research trends, thematic focuses, and future opportunities through bibliometric mapping and SWOT (Strengths, Weaknesses, Opportunities, Threats) analysis. Methods: A comprehensive search was conducted in the Web of Science (WoS) database (1989-2025) using the terms "artificial intelligence, " "machine learning, " "deep learning, " "neural network, " and "dialysis. " A total of 199 English-language research and review articles were included. Bibliometric visualization was performed using VOSviewer to identify keyword networks, citation patterns, and international collaborations. A SWOT analysis was further applied to delineate the strengths, weaknesses, opportunities, and threats of the field. Results: The results demonstrated that "machine learning, " "hemodialysis, " and "predictive model" are central topics, with hemodialysis dominating the literature while peritoneal dialysis remains underexplored. Highly cited works focused on blood pressure prediction, anemia management, and mortality modeling. China and the United States of America (USA) were the most productive and influential countries, with the USA acting as a key collaboration bridge. Strategic contributions also came from Spain, Germany, and Taiwan. Opportunities include the integration of explainable AI, personalized medicine approaches such as digital twins, and expanding research on peritoneal dialysis. Conversely, challenges involve the rapid evolution of algorithms and the short-lived nature of pandemic-driven research trends. Conclusion: This study represents the first bibliometric and SWOT analysis specifically dedicated to AI in dialysis, highlighting both its current trajectory and the need for sustained global collaboration and clinical validation to ensure a long-term impact.
