改进跨站点全心分割
Improving Cross-Site Whole-Heart Segmentation
- Purdue University（普渡大学）
- Carmel High School（卡梅尔高中）
- Case Western Reserve University（凯斯西储大学）
- The Pennsylvania State University（宾夕法尼亚州立大学）
- Georgia Institute of Technology（佐治亚理工学院）
- University at Albany, State University of New York（纽约州立大学奥尔巴尼分校）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对多中心多模态下全心分割的分布偏移问题，提出结合TotalSegmentator初始化nnU-Netv2与站点特征外观增强的方法，提升了CT和MRI的跨站点分割性能。
AI中文摘要：
从CT和MRI图像中进行全心分割是定量心脏图像分析的基础，但在多中心、多模态分布偏移的情况下仍具有挑战性。在CARE全心分割任务中，模型必须从有限的标记站点泛化到未见过的采集分布，其中间距、强度、重建纹理和解剖结构的变化会降低分布外的性能。我们提出了一种模态路由的3D心脏分割流程，该流程结合了由TotalSegmentator初始化的nnU-Netv2模型，以及具有站点特征的、保留标签的外观增强。我们首先使用可测量的图像属性对可用站点进行表征，并利用该分析来确定候选数据空间泛化路径。最终采用的方案应用了Bias Field + Bezier外观增强，结合了平滑的空间强度扰动与非线性强度重映射，随后进行轻量级的逐类别最大连通分量清理。在主要的保留站点验证划分上，该配置将CT的平均Dice从0.8350提升至0.9135，MRI的平均Dice从0.7695提升至0.7830，同时降低了HD95。这些结果表明，基于站点的外观增强是在有限数据下提高跨站点全心分割鲁棒性的实用策略。我们的代码可在该https URL获取。
英文摘要：
Whole-heart segmentation from CT and MRI is essential for quantitative cardiac image analysis, but remains challenging under multi-center and multi-modality distribution shift. In the CARE whole-heart segmentation task, models must generalize from limited labeled sites to unseen acquisition distributions, where variation in spacing, intensity, reconstruction texture, and anatomy can degrade out-of-distribution performance. We propose a modality-routed 3D cardiac segmentation pipeline that combines TotalSegmentator-initialized nnU-Netv2 models with site-characterized, label-preserving appearance augmentation. We first characterize the available sites using measurable image properties and use this analysis to motivate candidate data-space generalization routes. The final retained recipe applies Bias Field + Bezier appearance augmentation, combining smooth spatial intensity perturbation with nonlinear intensity remapping, followed by lightweight class-wise largest-connected-component cleanup. On the primary held-out-site validation splits, the final configuration improves CT mean Dice from 0.8350 to 0.9135 and MRI mean Dice from 0.7695 to 0.7830, while also reducing HD95. These results suggest that site-motivated appearance augmentation is a practical strategy for improving cross-site robustness in limited-data whole-heart segmentation. Our code can be found in https://github.com/Purdue-M2/Improving-Cross-Site-Whole-Heart-Segmentation
