ReportMedSAM：通过放射学报告指导分割

中文摘要 AI 辅助

自由形式的放射学报告包含丰富的临床描述，但由于自然语言的固有变异性，将其转换为可靠的分割仍然具有挑战性。现有管道通常依赖预定义的器官短语或基于规则的脆弱推理时间提取，限制了对新解剖结构的可扩展性且对语言变化敏感。为此，我们提出ReportMedSAM，一个报告驱动的框架，用可学习的概念库取代离散提取。通过利用冻结的医学视觉语言编码器（BiomedCLIP），通过对比学习将器官级概念嵌入与大规模临床语料库对齐，建立相互正交的语义锚点。我们的方法明确减轻器官级语义崩溃，并确保对不同临床同义词的高鲁棒性。在推理过程中，临床报告被嵌入并与概念库匹配以动态激活特定任务的专家混合（MoE）模块。这种解耦设计允许在不重新训练现有组件的情况下添加新概念和专家，提供参数隔离的扩展机制，同时保持先前学习的专家不变。在AbdomenAtlas 3.0数据集上评估，ReportMedSAM有效解释自由形式报告，实现有竞争力的分割精度，并展示对新临床任务的无缝、无干扰扩展。

英文摘要

Free-form radiology reports contain rich clinical descriptions, yet converting them for reliable segmentation remains challenging due to the inherent variability of natural language. Existing pipelines often rely on predefined organ phrases or brittle rule-based inference-time extraction, which limits their scalability to novel anatomical structures and makes them sensitive to linguistic variations. To address this, we propose ReportMedSAM, a report-driven framework that replaces discrete extraction with a learnable concept bank. By leveraging a frozen medical vision-language encoder (BiomedCLIP), we align organ-level concept embeddings with large-scale clinical corpora through contrastive learning, establishing mutually orthogonal semantic anchors. Our approach explicitly mitigates organ-level semantic collapse and ensures high robustness against diverse clinical synonyms (e.g., "renal" vs. "kidney" ). During inference, a clinical report is embedded and matched against this concept bank to dynamically activate task-specific Mixture-of-Experts (MoE) modules. This decoupled design allows new concepts and experts to be added without retraining existing components, providing a parameter-isolated extension mechanism while keeping previously learned experts unchanged. Evaluated on the AbdomenAtlas 3.0 dataset, ReportMedSAM effectively interprets free-form reports, achieves competitive segmentation accuracy, and demonstrates seamless, non-interfering extension to novel clinical tasks.

发表机构

• School of Engineering, College of Engineering and Physical Sciences, University of Birmingham（工程学院，工程与物理科学学院，伯明翰大学）
• Department of Cancer and Genomic Sciences, College of Medicine and Health, University of Birmingham（癌症与基因组科学系，医学与健康学院，伯明翰大学）
• Department of Computer Science, Faculty of Science and Engineering, University of Manchester（计算机科学系，科学与工程学院，曼彻斯特大学）
• The Division of Informatics, Imaging and Data Science, Faculty of Biology, Medicine, and Health, University of Manchester（信息学、成像与数据科学部，生物、医学与健康学院，曼彻斯特大学）

机构由 AI 辅助整理，请以论文原文为准。