用图结构精准定位CT中的病灶，提升放射报告与影像的对应精度。
GLeVE: Graph-Guided Lesion Grounding with Proposal Verification in 3D CT
- 构建病灶关系图，通过语义图推理生成可区分的查询特征。
- 结合解剖先验验证候选区域，实现文本与病灶的一一对应。
- 基于八叉树渐进细化边界，适合临床精确病灶定位场景。
将放射科报告描述与3D CT影像进行对齐，对可验证的临床解读至关重要，但受限于自由文本叙述与体数据解剖结构之间的语义-空间鸿沟。现有报告辅助与视觉-语言对齐方法多依赖短语级匹配或密集像素监督，导致病灶级对应不足，定位精度有限。我们提出GLeVE，一种基于图引导的病灶定位框架，包含解剖先验验证与八叉树自回归精修。该方法将每个病灶描述视为原子语义单元，通过关系感知图推理编码器官归属、属性及病灶间关系，生成具有区分性的病灶级查询。结合区域级验证的解剖感知候选生成，强制实现一对一文本-病灶对齐；层级八叉树精修则逐步优化边界分割。在AbdomenAtlas 3.0数据集上的实验表明，GLeVE在分割准确率与病灶级定位性能上均显著优于经典多模态基础模型与报告监督基线。
原文摘要 · Abstract (English)
Grounding radiology report descriptions to 3D CT volumes is essential for verifiable clinical interpretation, yet remains challenging due to the semantic-spatial gap between free-text narratives and volumetric anatomy. Existing report-assisted and vision-language grounding methods typically rely on phrase-level alignment or dense pixel supervision, resulting in limited lesion-wise correspondence and suboptimal localization accuracy. We propose GLeVE, a graph-guided lesion grounding framework with anatomical prior verification and octree-based autoregressive refinement. GLeVE treats each lesion description as an atomic semantic unit and encodes organ attribution, attributes, and inter-lesion relations through relation-aware graph reasoning to produce discriminative lesion-wise queries. Anatomy-aware proposal generation with region-level verification enforces one-to-one text-lesion alignment, while hierarchical octree refinement progressively improves boundary delineation. Experiments on AbdomenAtlas 3.0 demonstrate consistent gains over classical multimodal foundation models and report-supervised baselines in both segmentation accuracy and lesion-level localization.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。