Multi-Agent AI Systems in Healthcare: A Systematic Review Enhancing Clinical Decision-Making
Izuchukwu Prince Nweke *
Department of Epidemiology and Evidence based Medicine, First Moscow State Medical University, I.M Sechenov, Russia.
Cletus Okechukwu Ogadah
Department of Epidemiology and Evidence based Medicine, First Moscow State Medical University, I.M Sechenov, Russia.
Konstantin Koshechkin
Center for Digital Medicine, First Moscow State University, I.M Sechenov, Russia.
Popoola Michael Oluwasegun
Department of Business Administration in Public Health Care, First Moscow State Medical University, I.M Sechenov, Russia.
*Author to whom correspondence should be addressed.
Abstract
Background: The increasing complexity of modern healthcare has driven interest in artificial intelligence (AI) systems capable of supporting decision-making, optimizing workflows, and enhancing patient outcomes. Among the most advanced technologies are multi-agent AI systems networks of autonomous agents designed to collaborate across tasks such as diagnosis, treatment, monitoring, and logistics. These systems offer a decentralized, intelligent approach to managing healthcare processes at scale.
Objective: This systematic review explores the design, application, and impact of multi-agent AI systems in healthcare enhancing clinical decision making, highlighting their benefits, implementation challenges, and future potential.
Methods: The review was conducted following the PRISMA 2020 guidelines. A comprehensive search of the PubMed database was performed for literature published between January 2017 and March 2025. A total of 6,059 articles were retrieved; after screening and eligibility assessment, 18 studies were included in the final synthesis. Data were extracted on study design, healthcare domain, agent architecture, clinical relevance, and implementation outcomes. A narrative synthesis approach was used to organize findings into key thematic areas.
Results: The included studies demonstrated broad applications of multi-agent AI systems across clinical and administrative domains. Key themes identified were clinical decision support, personalized treatment plans, optimizing healthcare operations, and public health interventions. Multi-agent systems were found to improve diagnostic accuracy, treatment planning, real-time monitoring, and interdepartmental coordination. However, issues such as data bias, lack of interoperability, and ethical concerns regarding accountability and transparency were frequently reported.
Conclusion: Multi-agent AI systems offer a promising framework for transforming healthcare delivery through enhanced automation, intelligence, and collaboration. While their impact is evident in improving efficiency and personalized care, widespread adoption requires overcoming technical, ethical, and infrastructural barriers. Continued research and real-world validation are essential to ensure these systems deliver safe, equitable, and transparent healthcare solutions.
Keywords: Multi-agent systems, artificial intelligence, healthcare technology, clinical decision support