Literature DB >> 35212009

On the shape of convolution kernels in MRI reconstruction: Rectangles versus ellipsoids.

Rodrigo A Lobos1, Justin P Haldar1.   

Abstract

PURPOSE: 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. THEORY: 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.
METHODS: 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).
RESULTS: 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.
CONCLUSION: Ellipsoidal kernel shapes may offer advantages over rectangular kernel shapes in various MRI applications.
© 2022 International Society for Magnetic Resonance in Medicine.

Entities:  

Keywords:  MRI reconstruction; convolution; kernel shape

Mesh:

Year:  2022        PMID: 35212009      PMCID: PMC8957538          DOI: 10.1002/mrm.29189

Source DB:  PubMed          Journal:  Magn Reson Med        ISSN: 0740-3194            Impact factor:   4.668


  18 in total

1.  Nonlinear GRAPPA: a kernel approach to parallel MRI reconstruction.

Authors:  Yuchou Chang; Dong Liang; Leslie Ying
Journal:  Magn Reson Med       Date:  2011-12-12       Impact factor: 4.668

2.  Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays.

Authors:  D K Sodickson; W J Manning
Journal:  Magn Reson Med       Date:  1997-10       Impact factor: 4.668

3.  Linear Predictability in MRI Reconstruction: Leveraging Shift-Invariant Fourier Structure for Faster and Better Imaging.

Authors:  Justin P Haldar; Kawin Setsompop
Journal:  IEEE Signal Process Mag       Date:  2020-01-17       Impact factor: 12.551

4.  Off-the-Grid Recovery of Piecewise Constant Images from Few Fourier Samples.

Authors:  Greg Ongie; Mathews Jacob
Journal:  SIAM J Imaging Sci       Date:  2016-07-21       Impact factor: 2.867

5.  Deep-Learning Methods for Parallel Magnetic Resonance Imaging Reconstruction: A Survey of the Current Approaches, Trends, and Issues.

Authors:  Florian Knoll; Kerstin Hammernik; Chi Zhang; Steen Moeller; Thomas Pock; Daniel K Sodickson; Mehmet Akçakaya
Journal:  IEEE Signal Process Mag       Date:  2020-01-20       Impact factor: 12.551

6.  P-LORAKS: Low-rank modeling of local k-space neighborhoods with parallel imaging data.

Authors:  Justin P Haldar; Jingwei Zhuo
Journal:  Magn Reson Med       Date:  2015-05-07       Impact factor: 4.668

7.  Low-rank modeling of local k-space neighborhoods (LORAKS) for constrained MRI.

Authors:  Justin P Haldar
Journal:  IEEE Trans Med Imaging       Date:  2014-03       Impact factor: 10.048

8.  ESPIRiT--an eigenvalue approach to autocalibrating parallel MRI: where SENSE meets GRAPPA.

Authors:  Martin Uecker; Peng Lai; Mark J Murphy; Patrick Virtue; Michael Elad; John M Pauly; Shreyas S Vasanawala; Michael Lustig
Journal:  Magn Reson Med       Date:  2014-03       Impact factor: 4.668

9.  KerNL: Kernel-Based Nonlinear Approach to Parallel MRI Reconstruction.

Authors:  Jingyuan Lyu; Ukash Nakarmi; Dong Liang; Jinhua Sheng; Leslie Ying
Journal:  IEEE Trans Med Imaging       Date:  2018-08-07       Impact factor: 10.048

10.  Minimal Linear Networks for Magnetic Resonance Image Reconstruction.

Authors:  Gilad Liberman; Benedikt A Poser
Journal:  Sci Rep       Date:  2019-12-20       Impact factor: 4.379

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.