A Foundation Model for Chest X-ray Interpretation with Grounded Reasoning via Online Reinforcement Learning

• URL: http://arxiv.org/abs/2509.03906v1
• Date: Thu, 04 Sep 2025 06:00:04 GMT
• Title: A Foundation Model for Chest X-ray Interpretation with Grounded Reasoning via Online Reinforcement Learning
• Authors: Qika Lin, Yifan Zhu, Bin Pu, Ling Huang, Haoran Luo, Jingying Ma, Zhen Peng, Tianzhe Zhao, Fangzhi Xu, Jian Zhang, Kai He, Zhonghong Ou, Swapnil Mishra, Mengling Feng,
• Abstract summary: DeepMedix-R1 is a holistic medical FM for chest X-ray (CXR) interpretation.<n>It generates both an answer and reasoning steps tied to the image's local regions for each query.
• Score: 41.27625400846057
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Medical foundation models (FMs) have shown tremendous promise amid the rapid advancements in artificial intelligence (AI) technologies. However, current medical FMs typically generate answers in a black-box manner, lacking transparent reasoning processes and locally grounded interpretability, which hinders their practical clinical deployments. To this end, we introduce DeepMedix-R1, a holistic medical FM for chest X-ray (CXR) interpretation. It leverages a sequential training pipeline: initially fine-tuned on curated CXR instruction data to equip with fundamental CXR interpretation capabilities, then exposed to high-quality synthetic reasoning samples to enable cold-start reasoning, and finally refined via online reinforcement learning to enhance both grounded reasoning quality and generation performance. Thus, the model produces both an answer and reasoning steps tied to the image's local regions for each query. Quantitative evaluation demonstrates substantial improvements in report generation (e.g., 14.54% and 31.32% over LLaVA-Rad and MedGemma) and visual question answering (e.g., 57.75% and 23.06% over MedGemma and CheXagent) tasks. To facilitate robust assessment, we propose Report Arena, a benchmarking framework using advanced language models to evaluate answer quality, further highlighting the superiority of DeepMedix-R1. Expert review of generated reasoning steps reveals greater interpretability and clinical plausibility compared to the established Qwen2.5-VL-7B model (0.7416 vs. 0.2584 overall preference). Collectively, our work advances medical FM development toward holistic, transparent, and clinically actionable modeling for CXR interpretation.

Related papers

• A DeepSeek-Powered AI System for Automated Chest Radiograph Interpretation in Clinical Practice [83.11942224668127]
  Janus-Pro-CXR (1B) is a chest X-ray interpretation system based on DeepSeek Janus-Pro model.<n>Our system outperforms state-of-the-art X-ray report generation models in automated report generation.
  arXiv  Detail & Related papers  (2025-12-23T13:26:13Z)
• An Explainable Hybrid AI Framework for Enhanced Tuberculosis and Symptom Detection [55.35661671061754]
  Tuberculosis remains a critical global health issue, particularly in resource-limited and remote areas.<n>We propose a framework which enhances disease and symptom detection on chest X-rays by integrating two supervised heads and a self-supervised head.<n>Our model achieves an accuracy of 98.85% for distinguishing between COVID-19, tuberculosis, and normal cases, and a macro-F1 score of 90.09% for multilabel symptom detection.
  arXiv  Detail & Related papers  (2025-10-21T17:18:55Z)
• X-Ray-CoT: Interpretable Chest X-ray Diagnosis with Vision-Language Models via Chain-of-Thought Reasoning [0.0]
  We propose X-Ray-CoT (Chest X-Ray Chain-of-Thought), a novel framework for intelligent chest X-ray diagnosis and interpretable report generation.<n>X-Ray-CoT simulates human radiologists' "chain-of-thought" by first extracting multi-modal features and visual concepts.<n>It achieves competitive quantitative performance, with a Balanced Accuracy of 80.52% and F1 score of 78.65% for disease diagnosis.
  arXiv  Detail & Related papers  (2025-08-17T18:00:41Z)
• CX-Mind: A Pioneering Multimodal Large Language Model for Interleaved Reasoning in Chest X-ray via Curriculum-Guided Reinforcement Learning [28.737391224748798]
  We propose CX-Mind, the first generative model to achieve interleaved "think-answer" reasoning for chest X-ray (CXR) tasks.<n> CX-Mind is driven by curriculum reinforcement learning and verifiable process rewards (RL-VPR)<n>Experiments demonstrate that CX-Mind significantly outperforms existing medical and generaldomain MLLMs in visual understanding, text generation, and alignment.
  arXiv  Detail & Related papers  (2025-07-31T05:07:18Z)
• GEMeX-RMCoT: An Enhanced Med-VQA Dataset for Region-Aware Multimodal Chain-of-Thought Reasoning [60.03671205298294]
  Medical visual question answering aims to support clinical decision-making by enabling models to answer natural language questions based on medical images.<n>Current methods still suffer from limited answer reliability and poor interpretability.<n>This work first proposes a Region-Aware Multimodal Chain-of-Thought dataset, in which the process of producing an answer is preceded by a sequence of intermediate reasoning steps.
  arXiv  Detail & Related papers  (2025-06-22T08:09:58Z)
• RadFabric: Agentic AI System with Reasoning Capability for Radiology [61.25593938175618]
  RadFabric is a multi agent, multimodal reasoning framework that unifies visual and textual analysis for comprehensive CXR interpretation.<n>System employs specialized CXR agents for pathology detection, an Anatomical Interpretation Agent to map visual findings to precise anatomical structures, and a Reasoning Agent powered by large multimodal reasoning models to synthesize visual, anatomical, and clinical data into transparent and evidence based diagnoses.
  arXiv  Detail & Related papers  (2025-06-17T03:10:33Z)
• Look & Mark: Leveraging Radiologist Eye Fixations and Bounding boxes in Multimodal Large Language Models for Chest X-ray Report Generation [2.821158017021184]
  Look & Mark (L&M) is a novel grounding fixation strategy that integrates radiologist eye fixations (Look) and bounding box annotations (Mark)<n>General-purpose models also benefit from L&M combined with in-context learning, with LLaVA-OV achieving an 87.3% clinical average performance (C.AVG)-the highest among all models.
  arXiv  Detail & Related papers  (2025-05-28T10:54:40Z)
• ChestX-Reasoner: Advancing Radiology Foundation Models with Reasoning through Step-by-Step Verification [57.22053411719822]
  ChestX-Reasoner is a radiology diagnosis MLLM designed to leverage process supervision mined directly from clinical reports.<n>Our two-stage training framework combines supervised fine-tuning and reinforcement learning guided by process rewards to better align model reasoning with clinical standards.
  arXiv  Detail & Related papers  (2025-04-29T16:48:23Z)
• MedVLM-R1: Incentivizing Medical Reasoning Capability of Vision-Language Models (VLMs) via Reinforcement Learning [29.84956540178252]
  Reasoning is a critical frontier for advancing medical image analysis.<n>We introduce MedVLM-R1, a medical VLM that explicitly generates natural language reasoning.<n>MedVLM-R1 boosts accuracy from 55.11% to 78.22% across MRI, CT, and X-ray benchmarks.
  arXiv  Detail & Related papers  (2025-02-26T23:57:34Z)
• A Vision-Language Foundation Model to Enhance Efficiency of Chest X-ray Interpretation [22.8169684575764]
  Over 1.4 billion chest X-rays (CXRs) are performed annually due to their cost-effectiveness as an initial diagnostic test.<n>This scale of radiological studies provides a significant opportunity to streamline CXR interpretation and documentation.<n>We constructed a large-scale dataset (CheXinstruct), which we utilized to train a vision-language foundation model (CheXagent)
  arXiv  Detail & Related papers  (2024-01-22T18:51:07Z)
• ChatRadio-Valuer: A Chat Large Language Model for Generalizable Radiology Report Generation Based on Multi-institution and Multi-system Data [115.0747462486285]
  ChatRadio-Valuer is a tailored model for automatic radiology report generation that learns generalizable representations. The clinical dataset utilized in this study encompasses a remarkable total of textbf332,673 observations. ChatRadio-Valuer consistently outperforms state-of-the-art models, especially ChatGPT (GPT-3.5-Turbo) and GPT-4 et al.
  arXiv  Detail & Related papers  (2023-10-08T17:23:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.