用生成模型提升内部CT重建精度，保持数据一致性。
FORCE-Interior: Measurement-Consistent Adaptation of a Poisson-Flow Generative Prior for Interior CT
- 不训练直接适配预训练生成模型，结合截断掩码优化迭代过程。
- 在严重截断条件下，PSNR、SSIM、LPIPS均优于现有方法。
- 适合需要高精度内部CT重建的临床场景，尤其关注数据一致性。
内部断层成像从截断投影中重建感兴趣区域（ROI），这是一个病态问题，存在解不唯一和截断引起的偏差。现有深度学习方法对ROI几何和噪声敏感，而表达性强的生成先验可能产生与测量不一致的内容。我们提出FORCE-Interior，一种无需训练即可将预训练泊松流生成先验适配到内部CT的方法。通过全视野的OS-SART热启动避免将所有测量衰减强制集中于ROI，截断掩码感知的OS-SART更新在采样过程中保持数据一致性。实验表明，当ROI尺寸更小时，FORCE-Interior在两个最严重截断情况下达到最优的PSNR、SSIM和LPIPS，最大ROI下表现也具竞争力，同时保持低投影域残差。这些结果支持可复用生成先验的测量一致性适配，但还需进一步临床与患者级验证。
原文摘要 · Abstract (English)
Interior tomography reconstructs a region of interest (ROI) from truncated projections, an ill-posed problem with non-unique solutions and truncation-induced bias. Existing deep-learning methods can be sensitive to changes in ROI geometry and noise, while expressive generative priors may produce measurement-inconsistent content without measurement constraints. We propose FORCE-Interior, a training-free adaptation of a pretrained Poisson-flow generative prior to interior CT. A full-field-of-view (FOV) OS-SART warm start avoids forcing all measured attenuation into the ROI, and truncation-mask-aware OS-SART updates enforce data consistency throughout sampling. In our experiment, FORCE-Interior achieves the best PSNR, SSIM, and LPIPS at the two more severely truncated synthetic ROI sizes and competitive performance at the largest ROI, while maintaining low projection-domain residuals. These findings support the measurement-consistent adaptation of a reusable generative CT prior, while further clinical and patient-level validation remains necessary.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。