ARIES: A Scalable Multi-Agent Orchestration Framework for Real-Time Epidemiological Surveillance and Outbreak Monitoring
Aniket Wattamwar, Sampson Akwafuo
TL;DR
The paper addresses knowledge gaps in global health surveillance by introducing ARIES, a task-specific, hierarchical agentic ecosystem that autonomously queries authoritative sources (WHO DONs, CDC WONDER, PubMed) and synthesizes real-time surveillance insights. It leverages a Manager Agent to decompose queries and coordinate specialized sub-agents within the CrewAI framework, enabling cross-source verification and transparent reasoning through a four-stage Investigative Loop. The approach demonstrates that specialized agentic swarms can outperform generic LLMs in both depth of reasoning and provenance of sources, with configuration-dependent tradeoffs in output length and source coverage. The work highlights significant practical impact for rapid outbreak response and sets a roadmap for integrating protocol-level connectivity, self-correction with human oversight, and broader data streams to maintain up-to-date epidemiological intelligence.
Abstract
Global health surveillance is currently facing a challenge of Knowledge Gaps. While general-purpose AI has proliferated, it remains fundamentally unsuited for the high-stakes epidemiological domain due to chronic hallucinations and an inability to navigate specialized data silos. This paper introduces ARIES (Agentic Retrieval Intelligence for Epidemiological Surveillance), a specialized, autonomous multi-agent framework designed to move beyond static, disease-specific dashboards toward a dynamic intelligence ecosystem. Built on a hierarchical command structure, ARIES utilizes GPTs to orchestrate a scalable swarm of sub-agents capable of autonomously querying World Health Organization (WHO), Center for Disease Control and Prevention (CDC), and peer-reviewed research papers. By automating the extraction and logical synthesis of surveillance data, ARIES provides a specialized reasoning that identifies emergent threats and signal divergence in near real-time. This modular architecture proves that a task-specific agentic swarm can outperform generic models, offering a robust, extensible for next-generation outbreak response and global health intelligence.