通过判别引导增强3D CT报告生成中的细粒度空间定位
Enhancing Fine-Grained Spatial Grounding in 3D CT Report Generation via Discriminative Guidance
AI总结：
本文提出DCP-PD框架，通过提炼自由文本报告中的细粒度线索，指导报告生成并减少对快捷方式的依赖，提升了CT-RATE和Rad-ChestCT的性能。
摘要：
视觉-语言模型（VLMs）用于放射学报告生成（RRG）可以生成长篇胸部CT报告，从体积扫描中展示强大的潜力以提高放射学工作流程效率和一致性。然而，现有方法面临两个关键限制：（i）训练监督通常较粗，将整个CT体积与完整的自由文本报告对齐，而没有明确对细粒度属性或病理位置进行对齐；（ii）评估通常是整体的（词汇重叠、实体匹配或LLM-as-a-judge分数），并且不适用于空间定位的诊断。我们提出判别线索提示与提示丢弃（DCP-PD），一种即插即用的框架，该框架从自由文本报告中提炼细粒度线索，并利用这些线索指导报告生成，同时通过提示丢弃减少对快捷方式的依赖。DCP-PD在CT-RATE上实现了最先进的性能，将宏F1从=0.501提高到0.603（20%相对），并且显著提高了Rad-ChestCT的分布外性能，从F1=0.266提高到0.503（89%相对）。最后，我们引入了一种分层、位置感知的问题集协议（存在→侧向→叶），直接评估病理位置定位，显示即使对于在当前基准上得分高的模型，细粒度空间定位仍然具有挑战性。
英文摘要：
Vision--language models (VLMs) for radiology report generation (RRG) can produce long-form chest CT reports from volumetric scans and show strong potential to improve radiology workflow efficiency and consistency. However, existing methods face two key limitations: (i) training supervision is often coarse, aligning a whole CT volume with a full free-text report without explicit alignment for fine-grained attributes or pathology locations; and (ii) evaluation is typically holistic (lexical overlap, entity matching, or LLM-as-a-judge scores) and not diagnostic for spatial grounding. We propose \emph{Discriminative Cue-Prompting with Prompt Dropout (DCP-PD)}, a plug-and-play framework that distills fine-grained cues from free-text reports and uses them to guide report generation while mitigating shortcut reliance via prompt dropout. DCP-PD achieves state-of-the-art performance on CT-RATE, improving macro F1 from $=0.501$ to $0.603$ (20% relative), and substantially boosts out-of-distribution performance on Rad-ChestCT from F1 $=0.266$ to $0.503$ (89% relative). Finally, we introduce a hierarchical, location-aware question-set protocol (presence $\rightarrow$ laterality $\rightarrow$ lobe) to directly assess pathology-location grounding, showing that fine-grained spatial localization remains challenging even for models that score highly on current benchmarks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
