从对齐到合成：用于文本到CT生成的对比体 grounding
From Alignment to Synthesis: Contrastive Volumetric Grounding for Text-to-CT Generation
- Unit of Artificial Intelligence and Computer Systems, Department of Engineering, Università Campus Bio-Medico di Roma（人工智能与计算机系统单位，工程系，罗马生物医学大学）
- Department of Diagnostics and Intervention, Biomedical Engineering and Radiation Physics, Umeå University（诊断与干预部门，生物医学工程与放射物理学，乌梅拉大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
该研究针对文本到CT生成的视觉-语言对齐瓶颈，提出带结构化难负样本的3D-CLIP编码器，结合端到端潜在扩散模型，在CT-RATE数据集上实现了优于现有方法的性能。
AI中文摘要：
从放射学报告生成语义可控的3D CT体积数据，不仅需要丰富的文本编码器，还需要基于体积空间的视觉-语言对齐。现有的文本到CT方法依赖仅用语言或2D视觉-语言目标预训练的编码器来条件化生成，提供的条件信号在语言表达上丰富，但在体积层面是盲目的。我们认为这是一个结构性局限：3D视觉-语言对齐的质量，而非文本编码器的丰富度，是体积扩散模型语义可控性的主要瓶颈。为解决该问题，我们提出一种面向生成的3D-CLIP编码器，该编码器采用仅在文本层面运行的结构化难负样本进行训练。此设计在不增加任何额外3D内存成本的情况下提升了对比难度，克服了体积编码器固有的小批量约束。所得编码器对完全端到端的潜在扩散模型进行条件化，该模型直接在3D潜在空间运行，消除了超分辨率流程引入的空间伪影和跨切片不一致性。通过系统的消融实验，我们建立了对齐质量与下游生成可控性之间的明确经验关联。在CT-RATE数据集的18种病理条件上进行评估，我们的方法在图像保真度和事实正确性上均达到了最先进的性能，同时比所有竞争方法需要更少的推理时间和GPU内存。代码位于此https URL。
英文摘要：
Generating semantically controllable 3D CT volumes from radiology reports requires more than a rich text encoder, it requires vision-language alignment grounded in volumetric space. Existing Text-to-CT approaches condition generation on encoders pretrained with language only or 2D vision-language objectives, providing conditioning signals that are linguistically expressive but volumetrically blind. We argue this is a structural limitation: the quality of 3D vision-language alignment, not the richness of the text encoder, is the primary bottleneck for semantic controllability in volumetric diffusion models. To address this, we propose a generation-oriented 3D-CLIP encoder trained with structured hard negatives that operate exclusively at the text level. This design increases contrastive difficulty without any additional 3D memory cost, overcoming the small-batch constraints inherent to volumetric encoders. The resulting encoder conditions a fully end-to-end latent diffusion model that operates directly in 3D latent space, eliminating the spatial artifacts and cross-slice inconsistencies introduced by super-resolution pipelines. Through systematic ablations, we establish a clear empirical link between grounding quality and downstream generative controllability. Evaluated on CT-RATE across 18 pathological conditions, our method achieves state-of-the-art performance on both image fidelity and factual correctness, while requiring less inference time and GPU memory than all competing methods. Code is at https://github.com/danielemolino/Text2CT.
