生成报告还是重复模板？测量和缓解三维CT报告生成中的模板崩溃
Generating Reports or Repeating Templates? Measuring and Mitigating Template Collapse in 3D CT Report Generation
- Technical University of Munich (TUM)（慕尼黑技术大学）
- TUM Hospital（TUM医院）
- Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对三维CT报告生成中模型输出多样性低、病理检测能力差的模板崩溃问题，提出解耦框架CLarGen，通过分离临床检测与语言合成，显著提升临床准确性并保持报告流畅性。
AI中文摘要：
现代三维医学视觉语言模型（VLM）能够生成流畅的放射学风格文本，但表现出极低的病理检测率和输出多样性，崩溃为低估罕见但关键发现的通用模板。我们将这种失败模式识别为模板崩溃。这种失败源于三维医学成像的独特限制，例如数据有限、标签严重不平衡以及体积编码器的弱信号。在这些限制下，文本生成目标鼓励捷径学习和流畅但基础薄弱的报告。我们通过临床保真度、输出多样性、正常模板偏差和罕见发现存活率系统性地诊断模板崩溃。为了缓解它，我们提出CLarGen，一个解耦框架，将说什么（临床检测）与怎么说（语言合成）分开。CLarGen使用（i）用于多标签病理检测的潜在查询变换器，（ii）用于临床匹配示例的病理引导检索，以及（iii）用于从检测到的发现和检索到的上下文中合成最终报告的医学语言模型。在最新的三维CT报告生成基线中，CLarGen缓解了模板崩溃，并在保持流畅报告的同时显著提高了临床准确性（macro-F1 0.487 vs. 0.189；CRG 0.472 vs. 0.368）。我们的结果表明，明确、可测量的临床基础对于抗模板崩溃的三维CT报告生成至关重要。代码将在接收后发布。
英文摘要：
Modern 3D medical vision-language models (VLMs) can generate fluent radiology-style text while exhibit critically low pathology detection and output diversity, collapsing to generic templates that under-report rare yet critical findings. We identify this failure mode as Template Collapse. This failure stems from the unique constraints of 3D medical imaging, e.g., limited data, severe label imbalance, and weak signals from volumetric encoders. Under these constraints, text-generation objectives encourage shortcut learning and fluent but weakly grounded reports. We systematically diagnose the Template Collapse through clinical fidelity, output diversity, normal-template bias, and rare-finding survival. To mitigate it, we propose CLarGen, a decoupled framework that separates what to say (clinical detection) from how to say it (language synthesis). CLarGen uses (i) a Latent Query Transformer for multi-label pathology detection, (ii) pathology-guided retrieval for clinically matched exemplars, and (iii) a medical language model to synthesize the final report from detected findings and retrieved context. Across state-of-the-art 3D CT report generation baselines, CLarGen mitigates Template Collapse and substantially improves clinical accuracy (macro-F1 0.487 vs. 0.189; CRG 0.472 vs. 0.368) while maintaining fluent reporting. Our results suggest that explicit, measurable clinical grounding is essential for template-collapse-resistant 3D CT report generation. Code is available at https://github.com/ai-med/CLarGen.
