Department of Bioengineering and Department of Radiation Oncology, University of California, Los Angeles, CA, USA.
Med Phys. 2021 Jul;48(7):3815-3826. doi: 10.1002/mp.14935. Epub 2021 Jun 2.
Multiresolution hierarchical strategy is typically used in conventional optimization-based image registration to capture varying magnitudes of deformations while avoiding undesirable local minima. A rough concept of the scale is captured in deep networks by the reception field of kernels, and it has been realized to be both desirable and challenging to capture convolutions of different scales simultaneously in registration networks. In this study, we propose a registration network that is conscious of and self-adaptive to deformation of various scales to improve registration performance.
Dilated inception modules (DIMs) are proposed to incorporate receptive fields of different sizes in a computationally efficient way. Scale adaptive modules (SAMs) are proposed to guide and adjust shallow features using convolutional kernels with spatially adaptive dilation rate learned from deep features. DIMs and SAMs are integrated into a registration network which takes a U-net structure. The network is trained in an unsupervised setting and completes registration with a single evaluation run.
Experiment with two-dimensional (2D) cardiac MRIs showed that the adaptive dilation rate in SAM corresponded well to the deformation scale. Evaluated with left ventricle segmentation, our method achieved a Dice coefficient of (0.93 ± 0.02), significantly better than SimpleElastix and networks without DIM or SAM. The average surface distance was less than 2 mm, comparable to SimpleElastix without statistical significance. Experiment with synthetic data demonstrated the effectiveness of DIMs and SAMs, which led to a significant reduction in target registration error (TRE) based on dense deformation field. The three-dimensional (3D) version of the network achieved a 2.52 mm mean TRE on anatomical landmarks in DIR-Lab thoracic 4DCTs, lower than SimpleElastix and networks without DIM or SAM with statistical significance. The average registration times were 0.002 s for 2D images with size 256 × 256 and 0.42 s for 3D images with size 256 × 256 × 96.
The introduction and integration of DIMs and SAMs addressed the heterogeneous scale problem in an efficient and self-adaptive way. The proposed method provides an alternative to the inefficient multiresolution registration setups.
在基于优化的传统图像配准中，多分辨率分层策略通常用于捕获不同程度的变形，同时避免出现不理想的局部最小值。核的感受野在深度网络中捕捉到了尺度的大致概念，并且已经意识到在配准网络中同时捕获不同尺度的卷积既理想又具有挑战性。在这项研究中，我们提出了一种对各种尺度的变形具有感知和自适应能力的配准网络，以提高配准性能。
提出了扩张的 inception 模块（DIM），以便以计算有效的方式合并不同大小的感受野。提出了尺度自适应模块（SAM），以指导和调整浅层特征，方法是使用从深层特征学习到的具有空间自适应扩张率的卷积核。DIM 和 SAM 被集成到一个以 U-net 结构为基础的配准网络中。该网络在无监督环境中进行训练，并通过单次评估运行完成配准。
二维（2D）心脏 MRI 的实验表明，SAM 中的自适应扩张率与变形尺度很好地对应。通过左心室分割进行评估，我们的方法的 Dice 系数为（0.93±0.02），明显优于 SimpleElastix 和没有 DIM 或 SAM 的网络。平均表面距离小于 2mm，与 SimpleElastix 相比没有统计学意义。合成数据的实验证明了 DIM 和 SAM 的有效性，它们导致基于密集变形场的目标注册误差（TRE）显著降低。网络的 3D 版本在 DIR-Lab 胸部 4DCT 的解剖学标志上实现了 2.52mm 的平均 TRE，低于 SimpleElastix 和没有 DIM 或 SAM 的网络，具有统计学意义。对于大小为 256×256 的 2D 图像，平均注册时间为 0.002s，对于大小为 256×256×96 的 3D 图像，平均注册时间为 0.42s。
DIM 和 SAM 的引入和集成以高效和自适应的方式解决了异质尺度问题。所提出的方法为低效的多分辨率配准设置提供了一种替代方案。
