跳到主要导航 跳到搜索 跳到主要内容

FiR-Rad: Fine-Grained Reinforcement With Structured Reasoning for Chest X-Ray Report Generation

  • Xin Mei
  • , Libin Yang
  • , Dehong Gao
  • , Xiaoyan Cai
  • , Junwei Han
  • , Tianming Liu
  • Northwestern Polytechnical University Xian
  • Chongqing University of Posts and Telecommunications
  • University of Georgia

科研成果: 期刊稿件 › 文章 › 同行评审

1 引用 (Scopus)

摘要

Automated chest X-ray report generation requires not only clinical accuracy but also transparent and interpretable diagnostic reasoning. In this work, we propose FiR-Rad, a two-stage framework that combines explicit structured reasoning with targeted fine-grained optimization. In the first stage, a supervised chain-of-thought approach guides the model to sequentially analyze and describe a comprehensive range of clinically significant thoracic abnormalities, ensuring clinically meaningful coverage. In the second stage, we introduce a segment-level reinforcement learning strategy based on Group Relative Policy Optimization (GRPO), which assigns precise rewards to each disease-specific reasoning step by evaluating the accuracy of corresponding findings in the synthesized report. This design provides direct feedback for intermediate reasoning and encourages consistency between detailed abnormality analysis and final diagnostic conclusions. Experimental results on the MIMIC-CXR and IU-Xray datasets demonstrate that our framework achieves state-of-the-art performance across clinical and linguistic metrics, with strong zero-shot generalization on IU-Xray. The proposed method significantly enhances interpretability and clinical accuracy, effectively addressing key limitations in automated radiology report generation.

源语言英语
页(从-至)2922-2935
页数14
期刊IEEE Transactions on Medical Imaging
卷45
期6
DOI
出版状态已出版 - 6月 2026
已对外发布是

学术指纹

探究 'FiR-Rad: Fine-Grained Reinforcement With Structured Reasoning for Chest X-Ray Report Generation' 的科研主题。它们共同构成独一无二的学术指纹。

引用此