EviAgent: Evidence-Driven Agent for Radiology Report Generation

TL;DR

EviAgent is a transparent, evidence-driven system for radiology report generation that improves trustworthiness and accuracy by explicitly incorporating visual evidence and domain knowledge, outperforming existing models.

Contribution

It introduces a modular, explainable approach that combines visual experts and retrieval modules to enhance report generation in radiology.

Findings

Outperforms existing models on multiple datasets

Provides explicit visual evidence supporting diagnoses

Enhances trustworthiness and clinical relevance

Abstract

Automated radiology report generation holds immense potential to alleviate the heavy workload of radiologists. Despite the formidable vision-language capabilities of recent Multimodal Large Language Models (MLLMs), their clinical deployment is severely constrained by inherent limitations: their "black-box" decision-making renders the generated reports untraceable due to the lack of explicit visual evidence to support the diagnosis, and they struggle to access external domain knowledge. To address these challenges, we propose the Evidence-driven Radiology Report Generation Agent (EviAgent). Unlike opaque end-to-end paradigms, EviAgent coordinates a transparent reasoning trajectory by breaking down the complex generation process into granular operational units. We integrate multi-dimensional visual experts and retrieval mechanisms as external support modules, endowing the system with explicit visual evidence and high-quality clinical priors. Extensive experiments on MIMIC-CXR, CheXpert Plus, and IU-Xray datasets demonstrate that EviAgent outperforms both large-scale generalist models and specialized medical models, providing a robust and trustworthy solution for automated radiology report generation.