MeDxAgent: Multi-Agent Consultation for Interactive Medical Diagnosis
For researchers building interactive diagnostic AI systems, this work provides a benchmark and a multi-agent approach that improves accuracy through sequential questioning.
The paper introduces MeDxBench, a benchmark of 4,421 clinical cases for interactive diagnosis, and MeDxAgent, a multi-agent system that achieves a 10.3% accuracy gain over the baseline, closing 52.3% of the gap to a full-information oracle.
Large language models (LLMs) are increasingly used for health-related decision support. Yet most evaluations treat diagnosis as a single-shot task with complete information provided upfront, often as a multiple-choice selection. This diverges from clinical practice, where diagnosis is interactive and open-ended, involving sequential hypothesis refinement through targeted questioning. We address this gap. We build MeDxBench, a large-scale benchmark of 4,421 clinical cases across 20 specialties. We further propose MeDxAgent, a multi-agent consultation system for interactive diagnosis, and systematically study its prompt-, flow- and agent-level design choices. MeDxAgent achieves a 10.3% accuracy gain over the baseline on MeDxBench, closing 52.3% of the gap to a full-information oracle. We find that specific design choices: collecting demographics first, passing summarized dialogue for diagnosis, and feeding candidate diagnoses for targeted questioning, improve accuracy, mirroring how physicians reason, though their effect emerges fully only in combination. Code and dataset will be released upon publication.