DispatchMAS: Fusing taxonomy and artificial intelligence agents for emergency medical services
🤖 AI Summary
Emergency medical dispatch (EMD) faces real-world challenges including emotionally distressed callers, ambiguous information, and high cognitive load on dispatchers. To address these, we propose the first high-fidelity EMD simulation system integrating clinical taxonomy—derived from MIMIC-III and comprising a six-stage call protocol plus disease/symptom ontology—with a multi-agent collaborative framework (AutoGen). The system features two specialized agents—Caller and Dispatcher—and is grounded in a fact-enhanced clinical knowledge base to ensure clinically plausible interactions. It supports dispatcher training, protocol evaluation, and real-time decision support. In 100 simulated cases, dispatch effectiveness reached 94% and guidance effectiveness 91%; expert reviewers rated outputs highly for clinical plausibility, neutrality, readability, and politeness. Our key contribution lies in deeply embedding structured clinical taxonomy into the multi-agent collaboration workflow, markedly enhancing simulation fidelity, safety, and clinical credibility.
Application Category
📝 Abstract
Objective: Emergency medical dispatch (EMD) is a high-stakes process challenged by caller distress, ambiguity, and cognitive load. Large Language Models (LLMs) and Multi-Agent Systems (MAS) offer opportunities to augment dispatchers. This study aimed to develop and evaluate a taxonomy-grounded, LLM-powered multi-agent system for simulating realistic EMD scenarios. Methods: We constructed a clinical taxonomy (32 chief complaints, 6 caller identities from MIMIC-III) and a six-phase call protocol. Using this framework, we developed an AutoGen-based MAS with Caller and Dispatcher Agents. The system grounds interactions in a fact commons to ensure clinical plausibility and mitigate misinformation. We used a hybrid evaluation framework: four physicians assessed 100 simulated cases for "Guidance Efficacy" and "Dispatch Effectiveness," supplemented by automated linguistic analysis (sentiment, readability, politeness). Results: Human evaluation, with substantial inter-rater agreement (Gwe's AC1 > 0.70), confirmed the system's high performance. It demonstrated excellent Dispatch Effectiveness (e.g., 94 % contacting the correct potential other agents) and Guidance Efficacy (advice provided in 91 % of cases), both rated highly by physicians. Algorithmic metrics corroborated these findings, indicating a predominantly neutral affective profile (73.7 % neutral sentiment; 90.4 % neutral emotion), high readability (Flesch 80.9), and a consistently polite style (60.0 % polite; 0 % impolite). Conclusion: Our taxonomy-grounded MAS simulates diverse, clinically plausible dispatch scenarios with high fidelity. Findings support its use for dispatcher training, protocol evaluation, and as a foundation for real-time decision support. This work outlines a pathway for safely integrating advanced AI agents into emergency response workflows.
Problem
Research questions and friction points this paper is trying to address.
Developing AI agents to simulate emergency medical dispatch scenarios
Creating taxonomy-grounded system for clinical plausibility in EMS
Evaluating AI performance for dispatcher training and decision support
Innovation
Methods, ideas, or system contributions that make the work stand out.
Fuses clinical taxonomy with AI multi-agent systems
Uses grounded interactions to ensure clinical plausibility
Integrates hybrid evaluation combining human and algorithmic metrics
🔎 Similar Papers
No similar papers found.
