AI 中文总结
针对潜空间各向异性导致的三维医学图像合成失真问题，提出潜结构流（LSF）方法，将潜状态分解为结构（低秩全局）与残差（局部），分别建模，仅改生成器，在跨模态合成与肿瘤修复中全面超越基线。
AI 中文摘要
潜变量生成模型通过在压缩空间中生成，使三维医学图像合成在计算上变得可行。然而，我们表明由ℓ2目标引起的常见平坦欧几里得假设是不精确的：潜空间几何是强各向异性的，以至于等量级的误差可以产生截然不同的解码失真。我们进一步发现这种各向异性具有一个清晰的特征：敏感变化集中在一个低秩子空间中。主要的低秩分量捕获整体结构，编码长程、空间协调的变化，同时抵抗局部噪声。相比之下，其正交残差主要捕获局部和图像特定的变化。受这种不对称性的启发，我们引入了潜结构流（LSF）。在每个块中，LSF将潜状态分解为结构和残差，用全局上下文建模结构变化，并局部预测残差变化，同时为输入结构保留直接路径。LSF仅改变生成器，保持冻结的编解码器和逐点训练目标不变。在跨模态合成和肿瘤修复任务中，LSF在全局和肿瘤特定指标上均优于所有比较的基线，证明了显式建模潜空间结构对三维医学图像合成的益处。
英文摘要
Latent generative models make 3D medical image synthesis computationally practical by generating in a compressed space. However, we show that the common flat Euclidean assumption induced by $\ell_2$ objectives is imprecise: latent-space geometry is so strongly anisotropic that equal-magnitude errors can produce drastically different decoded distortions. We further find that this anisotropy has a clear feature: sensitive variation concentrates in a low-rank subspace. The dominant low-rank components capture the overall structure, encoding long-range, spatially coordinated variation while remaining resistant to local noise. Its orthogonal residual, in contrast, mainly captures local and image-specific variation. Motivated by this asymmetry, we introduce Latent Structure Flow (LSF). At each block, LSF decomposes the latent state into structure and residual, models structural changes with global context, and predicts residual variation locally while preserving a direct path for the input structure. LSF changes only the generator, leaving the frozen codec and pointwise training objective unchanged. Across cross-modality synthesis and tumor inpainting tasks, LSF outperforms all compared baselines on both global and tumor-specific metrics, demonstrating the benefit of explicitly modeling latent-space structure for 3D medical image synthesis.
Comments5 pages, 4 figures, 4 tables
