RADAR: Enhancing Radiology Report Generation with Supplementary Knowledge Injection

TL;DR

Radar enhances radiology report generation by intelligently combining the internal knowledge of large language models with externally retrieved domain-specific information, leading to more accurate and informative reports.

Contribution

The paper introduces Radar, a novel framework that systematically integrates internal LLM knowledge with external data for improved radiology report generation.

Findings

Outperforms state-of-the-art LLMs on MIMIC-CXR, CheXpert-Plus, IU X-ray datasets.

Generates reports with higher language quality and clinical accuracy.

Effectively leverages both internal and external knowledge sources.

Abstract

Large language models (LLMs) have demonstrated remarkable capabilities in various domains, including radiology report generation. Previous approaches have attempted to utilize multimodal LLMs for this task, enhancing their performance through the integration of domain-specific knowledge retrieval. However, these approaches often overlook the knowledge already embedded within the LLMs, leading to redundant information integration. To address this limitation, we propose Radar, a framework for enhancing radiology report generation with supplementary knowledge injection. Radar improves report generation by systematically leveraging both the internal knowledge of an LLM and externally retrieved information. Specifically, it first extracts the model's acquired knowledge that aligns with expert image-based classification outputs. It then retrieves relevant supplementary knowledge to further enrich this information. Finally, by aggregating both sources, Radar generates more accurate and informative radiology reports. Extensive experiments on MIMIC-CXR, CheXpert-Plus, and IU X-ray demonstrate that our model outperforms state-of-the-art LLMs in both language quality and clinical accuracy.