MedRAX: Medical Reasoning Agent for Chest X-ray

2025-02-04

AI Summary

Current chest X-ray (CXR) analysis models operate in isolation and lack the capability to address complex, multi-step clinical queries. To bridge this gap, we propose ChestAgent—the first zero-shot medical reasoning agent specifically designed for CXR interpretation. Without additional training, ChestAgent dynamically orchestrates multimodal large language models (MLLMs) and domain-specific CXR vision models via an end-to-end composable agent architecture and prompt-driven dynamic tool invocation, enabling cross-modal clinical reasoning. Our key contributions are threefold: (1) the novel ChestAgent framework; (2) ChestAgentBench—the first benchmark tailored for CXR agent evaluation, comprising 2,500 diverse, clinically grounded queries across multiple categories; and (3) state-of-the-art performance on ChestAgentBench, significantly outperforming both leading open-source and proprietary models in automated CXR interpretation. All code and data are publicly released.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Computer Vision: Multi-modal Vision
• Multiagent Systems: Modeling other Agents

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Chest X-rays (CXRs) play an integral role in driving critical decisions in disease management and patient care. While recent innovations have led to specialized models for various CXR interpretation tasks, these solutions often operate in isolation, limiting their practical utility in clinical practice. We present MedRAX, the first versatile AI agent that seamlessly integrates state-of-the-art CXR analysis tools and multimodal large language models into a unified framework. MedRAX dynamically leverages these models to address complex medical queries without requiring additional training. To rigorously evaluate its capabilities, we introduce ChestAgentBench, a comprehensive benchmark containing 2,500 complex medical queries across 7 diverse categories. Our experiments demonstrate that MedRAX achieves state-of-the-art performance compared to both open-source and proprietary models, representing a significant step toward the practical deployment of automated CXR interpretation systems. Data and code have been publicly available at https://github.com/bowang-lab/MedRAX

Problem

• Unified AI for CXR analysis
• Dynamic model integration without training
• Benchmark for complex medical queries

Innovation

• Integrates CXR analysis tools
• Uses multimodal language models
• No additional training required