发表机构
Stanford University; Texas A&M University; University of Minnesota; University of Wisconsin-Madison; Johns Hopkins University; UTHealth Houston; Stony Brook University（斯坦福大学; 德克萨斯农工大学; 明尼苏达大学; 威斯康星大学麦迪逊分校; 约翰霍普金斯大学; 休斯顿健康科学中心; 纽约州立大学石溪分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
AnaDiffusion是一种解剖学组成式潜在扩散框架，通过分解3D脑部MRI生成为解剖区域并组装细化，在ADNI测试集上实现低FID与高Cohen's d值，支持无需密集分割图的可控部分编辑。
AI 中文摘要
3D脑部MRI生成在医学成像、模拟及可控解剖分析领域已取得显著进展。然而，现有生成模型通常整体合成3D体积，往往忽略区域解剖结构，限制了局部可控性。为解决这些局限，我们提出AnaDiffusion，这是一种解剖学组成式潜在扩散框架，将生成过程分解为不同的、具有解剖学意义的区域，随后进行部分到整体的组装与全局细化。我们的方法首先训练部分扩散模型以捕捉局部结构先验，随后将各部分组装成的解剖学复合体注入全脑潜在表示中并继续去噪。该机制使模型能在保留注入解剖结构的同时解析全局上下文。因此，AnaDiffusion既生成明确的部分资产，又生成全局连贯的体积，从而支持可控的部分编辑，且无需在推理时使用特定受试者的密集分割图，同时保持一致的部分到整体脑部结构。在受试者不重叠的ADNI测试划分中，AnaDiffusion在全脑、左右半球、小脑-脑干复合体及接缝区域均实现最低的FID；在评估方法中，其还实现最佳的小脑绝对Cohen's d值，以及次佳的脑室和脑干绝对Cohen's d值。在局部编辑实验中，配对MS-SSIM显示出高目标迁移率和脱靶保留率，支持以最小的非预期解剖学改变实现可控的部分替换。
英文摘要
3D brain MRI generation has made significant advances in medical imaging, simulation, and controllable anatomical analysis. However, existing generative models typically synthesize 3D volumes monolithically, often overlooking regional anatomical structures and limiting local controllability. To address these limitations, we introduce AnaDiffusion, an anatomically compositional latent diffusion framework that factorizes the generation process into distinct, anatomically meaningful regions, followed by part-to-whole assembly and global refinement. Our approach first trains part diffusion models to capture local structural priors. We then inject an assembled anatomical composite of the parts into the whole-brain latent representation and continue denoising. This mechanism enables the model to resolve global context while preserving the injected anatomy. As a result, AnaDiffusion produces both explicit part assets and a globally coherent volume, thereby enabling controllable part editing without requiring subject-specific dense segmentation maps at inference time while maintaining consistent part-to-whole brain structure. On the subject-disjoint ADNI test split, AnaDiffusion achieves the lowest FID across the whole brain, left and right hemispheres, cerebellar-brainstem complex, and seam regions. It also achieves the best cerebellar and second-best ventricular and brainstem absolute Cohen's d values among the evaluated methods. In localized editing experiments, paired MS-SSIM demonstrates high target transfer and off-target preservation, supporting controllable part replacement with minimal unintended anatomical alterations.
