MedReason-R1: Learning to Reason for CT Diagnosis with Reinforcement Learning and Local Zoom

• URL: http://arxiv.org/abs/2510.19626v1
• Date: Wed, 22 Oct 2025 14:21:59 GMT
• Title: MedReason-R1: Learning to Reason for CT Diagnosis with Reinforcement Learning and Local Zoom
• Abstract summary: General-purpose Vision-Language Models (VLMs) demonstrate strong capabilities in generating detailed descriptions for natural images.<n>However, their performance in the medical domain remains suboptimal, even for relatively straightforward tasks.<n>We propose MedReason-R1, a medical VLM with explicit reasoning process for disease diagnosis.
• Score: 5.198944594852047
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: General-purpose large Vision-Language Models (VLMs) demonstrate strong capabilities in generating detailed descriptions for natural images. However, their performance in the medical domain remains suboptimal, even for relatively straightforward tasks, primarily due to the lack of large-scale, high-quality, specialized medical imaging datasets and the neglect of the diagnostic process that progresses from coarse to fine-grained. To address the first issue, we construct the CT-RATE-VQA dataset, which has 84K QA pairs. For the second issue, we propose MedReason-R1, a medical VLM with explicit reasoning process for disease diagnosis. MedReason-R1 incorporates a novel strategy that embeds zoom-in disease region-of-interest areas into the image, highlighting the crucial role of both global localization and disease-specific details in enhancing the model's diagnostic performance. Furthermore, we introduce the GRPO reinforcement learning framework to MedReason-R1, which enables effective reasoning without relying on costly manual annotations. Compared to recent general-purpose and medical VLMs, MedReason-R1 achieves state-of-the-art performance in CT disease diagnosis while retaining generalization. The code, checkpoints, and dataset are available at: https://github.com/Leevan001/MedReason-R1

Related papers

• An Explainable Vision-Language Model Framework with Adaptive PID-Tversky Loss for Lumbar Spinal Stenosis Diagnosis [1.2871968485402088]
  Existing vision-language models fail to address the extreme class imbalance prevalent in clinical segmentation datasets.<n>We present an end-to-end Explainable Vision-Language Model framework designed to overcome these limitations.<n>We show considerable performance: a diagnostic classification accuracy of 90.69%, a macro-averaged Dice score of 0.9512 for segmentation, and a CIDEr score of 92.80%.
  arXiv  Detail & Related papers  (2026-04-02T20:18:35Z)
• MMedExpert-R1: Strengthening Multimodal Medical Reasoning via Domain-Specific Adaptation and Clinical Guideline Reinforcement [63.82954136824963]
  Medical Vision-Language Models excel at perception tasks with complex clinical reasoning required in real-world scenarios.<n>We propose a novel reasoning MedVLM that addresses these challenges through domain-specific adaptation and guideline reinforcement.
  arXiv  Detail & Related papers  (2026-01-16T02:32:07Z)
• XAI-Driven Diagnosis of Generalization Failure in State-Space Cerebrovascular Segmentation Models: A Case Study on Domain Shift Between RSNA and TopCoW Datasets [0.5735035463793009]
  We present a rigorous, two-phase approach to diagnose the generalization failure of state-of-the-art State-Space Models (SSMs)<n>We quantified the model's focus by measuring the overlap between its attention maps and the Ground Truth segmentations.<n>Our analysis proves the model failed to generalize because its attention mechanism abandoned true anatomical features in the Target domain.
  arXiv  Detail & Related papers  (2025-12-16T00:34:32Z)
• Lesion-Aware Post-Training of Latent Diffusion Models for Synthesizing Diffusion MRI from CT Perfusion [12.777937532267105]
  Latent diffusion models (LDMs) leverage efficient learning in compressed latent space.<n>This efficiency comes with a trade-off, potentially compromising crucial pixel-level detail.<n>We propose a novel post-training framework for LDMs in medical image-to-image translation by incorporating lesion-aware medical pixel space objectives.
  arXiv  Detail & Related papers  (2025-10-10T06:51:27Z)
• A Chain of Diagnosis Framework for Accurate and Explainable Radiology Report Generation [4.61181046331792]
  We propose a framework named chain of diagnosis (CoD), which maintains a chain of diagnostic process for clinically accurate and explainable RRG.<n>To enhance explainability, a diagnosis grounding module is designed to match QA diagnoses and generated sentences, where the diagnoses act as a reference.<n>Our efforts lead to 1) an omni-labeled RRG dataset with QA pairs and lesion boxes; 2) a evaluation tool for assessing the accuracy of reports in describing lesion location and severity; 3) extensive experiments to demonstrate the effectiveness of CoD.
  arXiv  Detail & Related papers  (2025-08-13T07:32:28Z)
• AMRG: Extend Vision Language Models for Automatic Mammography Report Generation [4.366802575084445]
  Mammography report generation is a critical yet underexplored task in medical AI.<n>We introduce AMRG, the first end-to-end framework for generating narrative mammography reports.<n>We train and evaluate AMRG on DMID, a publicly available dataset of paired high-resolution mammograms and diagnostic reports.
  arXiv  Detail & Related papers  (2025-08-12T06:37:41Z)
• Multimodal Causal-Driven Representation Learning for Generalizable Medical Image Segmentation [56.52520416420957]
  We propose Multimodal Causal-Driven Representation Learning (MCDRL) to tackle domain generalization in medical image segmentation.<n>MCDRL consistently outperforms competing methods, yielding superior segmentation accuracy and exhibiting robust generalizability.
  arXiv  Detail & Related papers  (2025-08-07T03:41:41Z)
• Med-R1: Reinforcement Learning for Generalizable Medical Reasoning in Vision-Language Models [6.176432104264649]
  Vision-language models (VLMs) have achieved impressive progress in natural image reasoning, yet their potential in medical imaging remains underexplored.<n>We propose Med-R1, a reinforcement learning (RL)-enhanced vision-language model designed to improve generalization and reliability in medical reasoning.<n>We evaluate Med-R1 across eight distinct medical imaging modalities.
  arXiv  Detail & Related papers  (2025-03-18T06:12:38Z)
• MedRAG: Enhancing Retrieval-augmented Generation with Knowledge Graph-Elicited Reasoning for Healthcare Copilot [47.77948063906033]
  Retrieval-augmented generation (RAG) is a well-suited technique for retrieving privacy-sensitive Electronic Health Records.<n>This paper proposes MedRAG, a RAG model enhanced by knowledge graph (KG)-elicited reasoning for the medical domain.<n>Tests show MedRAG provides more specific diagnostic insights and outperforms state-of-the-art models in reducing misdiagnosis rates.
  arXiv  Detail & Related papers  (2025-02-06T12:27:35Z)
• MiniGPT-Med: Large Language Model as a General Interface for Radiology Diagnosis [28.421857904824627]
  MiniGPT-Med is a vision-language model derived from large-scale language models and tailored for medical applications. It is capable of performing tasks such as medical report generation, visual question answering (VQA), and disease identification within medical imagery. It achieves state-of-the-art performance on medical report generation, higher than the previous best model by 19% accuracy.
  arXiv  Detail & Related papers  (2024-07-04T18:21:10Z)
• Cross-Modal Causal Intervention for Medical Report Generation [107.76649943399168]
  Radiology Report Generation (RRG) is essential for computer-aided diagnosis and medication guidance.<n> generating accurate lesion descriptions remains challenging due to spurious correlations from visual-linguistic biases.<n>We propose a two-stage framework named CrossModal Causal Representation Learning (CMCRL)<n> Experiments on IU-Xray and MIMIC-CXR show that our CMCRL pipeline significantly outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-03-16T07:23:55Z)
• AlignTransformer: Hierarchical Alignment of Visual Regions and Disease Tags for Medical Report Generation [50.21065317817769]
  We propose an AlignTransformer framework, which includes the Align Hierarchical Attention (AHA) and the Multi-Grained Transformer (MGT) modules. Experiments on the public IU-Xray and MIMIC-CXR datasets show that the AlignTransformer can achieve results competitive with state-of-the-art methods on the two datasets.
  arXiv  Detail & Related papers  (2022-03-18T13:43:53Z)
• Auxiliary Signal-Guided Knowledge Encoder-Decoder for Medical Report Generation [107.3538598876467]
  We propose an Auxiliary Signal-Guided Knowledge-Decoder (ASGK) to mimic radiologists' working patterns. ASGK integrates internal visual feature fusion and external medical linguistic information to guide medical knowledge transfer and learning.
  arXiv  Detail & Related papers  (2020-06-06T01:00:15Z)

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.