从像素生成到拓扑推断：用于可信跨物理域骨小梁形态学习的结构双重超分辨率
From Pixel Generation to Topological Inference: Structural Dual Super-Resolution for Trustworthy Cross-Physical-Domain Trabecular Morphology Learning
- Institute of High Energy Physics, Chinese Academy of Sciences（中国科学院高能物理研究所）
- Beijing Jishuitan Hospital（北京积水潭医院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对临床CT无法分辨骨小梁且现有超分辨率方法不适定的问题，提出结构双重超分辨率范式，通过结构对偶约束实现从宏观到微观的确定性拓扑推断，在7项指标上与SRμCT一致，并验证跨源泛化。
AI中文摘要：
临床CT和超高分辨率CT（UHRCT）无法分辨单个骨小梁，而同步辐射显微CT（SRμCT）提供3.2微米的高分辨率参考，但不适用于体内成像。两个域的分辨率相差31.25倍，仅粗略配对，且数据量差异巨大。此外，临床UHRCT存在严重的部分容积效应、强噪声以及束硬化/散射伪影，而SRμCT几乎无此问题。现有的超分辨率网络和预训练先验方法表现不佳，因为它们针对像素生成——关注多样细节和SSIM/PSNR——并未显式建模这些物理差异。这表明通过像素生成实现32倍超分辨率本质上是不适定的。我们提出从像素生成到拓扑推断的范式转变：从宏观尺度低分辨率输入确定性地预测不变微结构，并以形态学参数评估。我们通过结构双重超分辨率实现这一范式，通过结构对偶约束耦合正向物理退化（微观到宏观）与逆向结构推断（宏观到微观）。该方法是一个端到端、少样本、紧凑的结构对偶网络（SDN），包括用于正向退化和逆向重建的双向建模网络、金字塔结构一致性判别器以及四个结构对偶约束。在测试集上，SDN在7项指标上实现了与SRμCT大体一致的形态学参数，使临床UHRCT具备显微成像级别的形态学量化能力，SSIM达到0.8。在3.2微米SSRF数据上训练的模型，能很好地泛化到来自独立来源的3.25微米BSRF数据，验证了跨源泛化能力，并确认所设计的网络实现了可信的结构推断而非像素生成。
英文摘要：
Clinical CT and UHRCT cannot resolve individual trabeculae, whereas synchrotron radiation microCT (SRuCT) provides high-resolution references but is not applicable for in vivo imaging. The two domains differ by a 32x resolution gap, are only coarsely paired, and exhibit severe physical differences including partial volume effects, noise, and artifacts. Existing super-resolution networks and pretrained-prior methods (GLEAN/StyleGAN2, Stable SR/LDM) underperform because they target pixel generation---diverse details and SSIM/PSNR---and do not model these physical differences. Pixel generation for a 32x resolution gap is intrinsically ill-posed. We propose a paradigm shift from pixel generation to topological inference: deterministically predicting invariant microstructures from macro-scale low-resolution inputs, evaluated by morphological parameters. The core of our 2D morphology learning lies in training on 2D slices while evaluating on 3D morphological parameters, ensuring that the learned representations capture true three-dimensional trabecular topology rather than 2D pixel statistics. We realize this paradigm via structural dual super-resolution, coupling forward physical degradation (micro-to-macro) with inverse structural inference (macro-to-micro) through structural duality constraints. The method is an end-to-end, few-shot, compact structural dual network (SDN), comprising a bidirectional modeling network, a multi-scale structural consistency discriminator, and four structural duality constraints. On the test set, SDN achieves morphological parameters largely consistent with SRuCT across six metrics, with SSIM reaching 0.8. Trained on 3.2 um SSRF data, the model generalizes well to 3.25 um isotropic BSRF data from an independent source, validating cross-source generalization and confirming trustworthy structural inference rather than pixel generation.
