发表机构
University of California, San Francisco; Cedars-Sinai Medical Center; University of California, Los Angeles（加利福尼亚大学旧金山分校; 西达赛奈医疗中心; 加利福尼亚大学洛杉矶分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究对比多种连续图像配准方法，发现配准精度取决于变形先验与目标运动模式的匹配度，多分辨率B样条方法MR-D-BSCP在脑部和肺部配准任务中表现最优。
AI 中文摘要
可变形图像配准模型通过其参数化和优化过程隐式编码变形先验。本研究对连续配准方法开展验证研究，以探究这些隐式先验如何影响不同配准任务的性能。经典B样条变换通过其控制点结构施加局部性、平滑性和尺度约束，而近期基于隐式神经表示（INR）的方法则通过神经参数化和优化施加不同的先验。我们对比了四种方法：INR-Dense（IDIR），采用基于SIREN的INR直接建模密集位移场；INR-BSCP（SINR），采用INR预测B样条控制点；D-BSCP，直接优化单尺度B样条控制点；MR-D-BSCP，添加多分辨率由粗到精方案。在跨个体脑部MR配准（OASIS）和跨个体呼气至吸气肺部CT配准（DIR-LAB 4DCT）上的实验显示，不同变形模式下各方法表现不同。在变形中等但局部复杂的OASIS数据集上，D-BSCP性能与INR-BSCP相当或略优，表明B样条参数化是INR-BSCP有效性的重要来源。在呼吸运动更大且更具连贯性的DIR-LAB 4DCT数据集上，单尺度B样条方法（D-BSCP和INR-BSCP）适用性较差，而INR-Dense和MR-D-BSCP更有效。在两个任务中，MR-D-BSCP在测试的连续参数化方法中性能最佳。这些发现表明配准精度高度依赖于使诱导的变形先验与目标运动模式相匹配，支持先验-变形匹配作为医学图像配准的实用设计原则。我们的代码将在该httpsURL发布。
英文摘要
Deformable image registration models implicitly encode deformation priors through their parametrization and optimization. In this work, we conduct a validation study on continuous registration methods to examine how these implicit priors affect performance across different registration tasks. Classic B-Spline transformations impose locality, smoothness, and scale through their control-point structure, whereas recent INR-based methods impose different priors through neural parameterization and optimization. We compare INR-Dense (IDIR), which directly models a dense displacement field using a SIREN-based INR; INR-BSCP (SINR), which predicts B-Spline control points with an INR; D-BSCP, which directly optimizes single-scale B-Spline control points; and MR-D-BSCP, which adds a multiresolution coarse-to-fine scheme. Experiments on inter-subject brain MR registration (OASIS) and intra-subject exhale-to-inhale lung CT registration (DIR-LAB 4DCT) reveal different behavior across deformation regimes. On OASIS, where deformations are moderate but locally complex, D-BSCP matches or slightly outperforms INR-BSCP, suggesting that the B-Spline parameterization accounts for much of INR-BSCP's effectiveness. On DIR-LAB 4DCT, where respiratory motion is larger and more coherent, single-scale B-Spline methods (D-BSCP and INR-BSCP) are less suitable, while INR-Dense and MR-D-BSCP are more effective. Across both tasks, MR-D-BSCP achieves the best performance among the tested continuous parameterizations. These findings highlight that registration accuracy depends strongly on matching the induced deformation prior to the target motion pattern, and support prior-deformation matching as a practical design principle for medical image registration. Our code will be available at https://github.com/HengjieLiu/RightPriorDIR.
