发表机构
Medical University of Graz（格拉茨医科大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出一种基于2.5D无条件流先验的零样本脑部MRI修复框架，通过自回归处理切片三元组解决2D方法的不连续性问题，并在BraTS 2026验证集上取得高SSIM和PSNR。
AI 中文摘要
脑部MRI体数据的生成式修复对于在病理区域合成健康组织至关重要，这能提高自动化下游脑部分析应用（如图像配准、脑提取和分割）的准确性和可靠性。然而，标准的3D方法计算开销巨大，而高效的2D逐切片方法则遭受严重的切片间不连续性。此外，传统模型依赖条件训练，需要对掩膜输入进行特定任务的学习。我们提出了一种零样本脑部MRI修复框架，利用2.5D无条件流先验来捕获沿上下轴的空间上下文，而无需完整3D卷积的开销。在训练期间，我们的流匹配模型学习相邻轴向切片三元组的联合分布，对健康脑解剖流形进行建模，同时明确地将病理区域排除在损失函数之外。在推理时，模型沿深度轴自回归地处理输入三元组。我们采用Restora-Flow求解器，利用输入掩膜约束无条件先验，实现准确的零样本修复。评估表明，我们的2.5D策略解决了2D基线的结构不连续性问题，在轴向、矢状面和冠状面上合成合理的健康组织，同时保持体积一致性。作为最后一步，我们生成并平均多个随机重建的集成，以形成最终预测。在官方BraTS 2026修复挑战验证集上基准测试的定量结果证明了我们提出方法的有效性，其SSIM为0.816±0.112，MSE为0.007±0.005，PSNR为22.923±4.343。代码可在以下https URL获取。
英文摘要
Generative inpainting of brain MRI volumes is essential for synthesizing healthy tissue in pathological regions, improving the accuracy and reliability of automated downstream brain analysis applications such as image registration, brain extraction, and segmentation. However, standard 3D approaches are computationally prohibitive, while efficient 2D slice-wise methods suffer from severe inter-slice discontinuities. Furthermore, traditional models rely on conditional training, requiring task-specific learning of masked inputs. We propose a zero-shot brain MRI inpainting framework utilizing 2.5D unconditional flow priors to capture spatial context along the superior-inferior axis without the overhead of full 3D convolutions. During training, our flow matching model learns the joint distribution of adjacent axial slice triplets, modeling the manifold of healthy brain anatomy while explicitly excluding pathological regions from the loss function. At inference, the model processes the input triplets autoregressively along the depth axis. We employ the Restora-Flow solver to constrain the unconditional prior using the input mask, achieving accurate zero-shot inpainting. Evaluations show our 2.5D strategy resolves the structural discontinuities of 2D baselines, synthesizing plausible healthy tissue while maintaining volumetric consistency across the axial, sagittal, and coronal planes. As a final step, we generate and average an ensemble of multiple stochastic reconstructions to form the final prediction. Quantitative results benchmarked on the official BraTS 2026 Inpainting Challenge validation set demonstrate the effectiveness of our proposed approach, yielding an SSIM of 0.816 $\pm$ 0.112, MSE of 0.007 $\pm$ 0.005, and PSNR of 22.923 $\pm$ 4.343. Code is available at https://github.com/imigraz/brats2026-inpainting.
