On the shape of convolution kernels in MRI reconstruction: Rectangles versus ellipsoids.
On the shape of convolution kernels in MRI reconstruction: Rectangles versus ellipsoids.
复制标题
关于 MRI 重建中卷积核的形状：矩形与椭圆体。
DOI：
10.1002/mrm.29189
复制
发表时间：
2022-06
影响因子：
3.3
通讯作者：
Haldar JP
中科院分区：
文献类型：
--
作者：
Lobos RA;Haldar JP
许多MRI重建方法（包括GRAPPA、SPIRiT、ESPIRiT、LORAKS和卷积神经网络（CNN）方法）涉及移位不变卷积模型。通常默认情况下选择矩形卷积核形状，尽管椭圆形核形状具有潜在的吸引人的理论特性。在这项工作中，我们系统地研究了不同的内核形状的选择在几个上下文中的差异。众所周知，k空间的矩形区域与各向异性的空间分辨率相关联，而椭圆形区域可以与更各向同性的分辨率相关联。此外，对于固定的空间分辨率，椭圆形核与比矩形核少得多的参数相关联。这些特性表明椭圆形内核可能比矩形内核具有某些优势。我们使用了真实的回溯欠采样k空间数据，在七种方法（GRAPPA，SPIRiT，ESPIRiT，SAKE，LORAKS，AC-LORAKS和基于CNN的重建）的背景下，实证研究了矩形和椭圆形内核的特性。经验结果表明，两种内核形状可以产生具有相似误差度量的重建图像，尽管椭圆形形状通常可以通过减少计算时间和内存使用和/或更少的模型参数来实现这一点。在各种MRI应用中，椭圆形核形状可以提供优于矩形核形状的优点。
Many MRI reconstruction methods (including GRAPPA, SPIRiT, ESPIRiT, LORAKS, and convolutional neural network (CNN) methods) involve shift-invariant convolution models. Rectangular convolution kernel shapes are often chosen by default, although ellipsoidal kernel shapes have potentially appealing theoretical characteristics. In this work, we systematically investigate the differences between different kernel shape choices in several contexts. It is well-understood that a rectangular region of k-space is associated with anisotropic spatial resolution, while ellipsoidal regions can be associated with more isotropic resolution. Further, for a fixed spatial resolution, ellipsoidal kernels are associated with substantially fewer parameters than rectangular kernels. These characteristics suggest that ellipsoidal kernels may have certain advantages over rectangular kernels. We used real retrospectively-undersampled k-space data to empirically study the characteristics of rectangular and ellipsoidal kernels in the context of seven methods (GRAPPA, SPIRiT, ESPIRiT, SAKE, LORAKS, AC-LORAKS, and CNN-based reconstructions). Empirical results suggest that both kernel shapes can produce reconstructed images with similar error metrics, although the ellipsoidal shape can often achieve this with reduced computation time and memory usage and/or fewer model parameters. Ellipsoidal kernel shapes may offer advantages over rectangular kernel shapes in various MRI applications.
登录
查看更多内容
影响因子：
3.3
作者：
Uecker, Martin;Lai, Peng;Murphy, Mark J.;Virtue, Patrick;Elad, Michael;Pauly, John M.;Vasanawala, Shreyas S.;Lustig, Michael
通讯作者：
Lustig, Michael
影响因子：
3.3
作者：
Sodickson, DK;Manning, WJ
通讯作者：
Manning, WJ
影响因子：
2.1
作者：
Ongie G;Jacob M
通讯作者：
Jacob M
影响因子：
3.3
作者：
Haldar JP;Zhuo J
通讯作者：
Zhuo J
影响因子：
10.6
作者：
Lyu J;Nakarmi U;Liang D;Sheng J;Ying L
通讯作者：
Ying L