Literature DB >> 34601751

Scan-specific artifact reduction in k-space (SPARK) neural networks synergize with physics-based reconstruction to accelerate MRI.

Yamin Arefeen1, Onur Beker2, Jaejin Cho3, Heng Yu4, Elfar Adalsteinsson1,5,6, Berkin Bilgic3,5,7.   

Abstract

PURPOSE: To develop a scan-specific model that estimates and corrects k-space errors made when reconstructing accelerated MRI data.
METHODS: Scan-specific artifact reduction in k-space (SPARK) trains a convolutional-neural-network to estimate and correct k-space errors made by an input reconstruction technique by back-propagating from the mean-squared-error loss between an auto-calibration signal (ACS) and the input technique's reconstructed ACS. First, SPARK is applied to generalized autocalibrating partially parallel acquisitions (GRAPPA) and demonstrates improved robustness over other scan-specific models, such as robust artificial-neural-networks for k-space interpolation (RAKI) and residual-RAKI. Subsequent experiments demonstrate that SPARK synergizes with residual-RAKI to improve reconstruction performance. SPARK also improves reconstruction quality when applied to advanced acquisition and reconstruction techniques like 2D virtual coil (VC-) GRAPPA, 2D LORAKS, 3D GRAPPA without an integrated ACS region, and 2D/3D wave-encoded imaging.
RESULTS: SPARK yields SSIM improvement and 1.5 - 2× root mean squared error (RMSE) reduction when applied to GRAPPA and improves robustness to ACS size for various acceleration rates in comparison to other scan-specific techniques. When applied to advanced reconstruction techniques such as residual-RAKI, 2D VC-GRAPPA and LORAKS, SPARK achieves up to 20% RMSE improvement. SPARK with 3D GRAPPA also improves RMSE performance by ~2×, SSIM performance, and perceived image quality without a fully sampled ACS region. Finally, SPARK synergizes with non-Cartesian, 2D and 3D wave-encoding imaging by reducing RMSE between 20% and 25% and providing qualitative improvements.
CONCLUSION: SPARK synergizes with physics-based acquisition and reconstruction techniques to improve accelerated MRI by training scan-specific models to estimate and correct reconstruction errors in k-space.
© 2021 International Society for Magnetic Resonance in Medicine.

Entities:  

Keywords:  accelerated imaging; convolutional neural networks; image reconstruction; machine learning; parallel imaging

Mesh:

Year:  2021        PMID: 34601751      PMCID: PMC8627503          DOI: 10.1002/mrm.29036

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


  49 in total

1.  SENSE: sensitivity encoding for fast MRI.

Authors:  K P Pruessmann; M Weiger; M B Scheidegger; P Boesiger
Journal:  Magn Reson Med       Date:  1999-11       Impact factor: 4.668

2.  Improved pediatric MR imaging with compressed sensing.

Authors:  Shreyas S Vasanawala; Marcus T Alley; Brian A Hargreaves; Richard A Barth; John M Pauly; Michael Lustig
Journal:  Radiology       Date:  2010-06-07       Impact factor: 11.105

3.  Compressed sensing reconstruction for magnetic resonance parameter mapping.

Authors:  Mariya Doneva; Peter Börnert; Holger Eggers; Christian Stehning; Julien Sénégas; Alfred Mertins
Journal:  Magn Reson Med       Date:  2010-10       Impact factor: 4.668

4.  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

5.  Wave-CAIPI for highly accelerated 3D imaging.

Authors:  Berkin Bilgic; Borjan A Gagoski; Stephen F Cauley; Audrey P Fan; Jonathan R Polimeni; P Ellen Grant; Lawrence L Wald; Kawin Setsompop
Journal:  Magn Reson Med       Date:  2014-07-01       Impact factor: 4.668

Review 6.  Deep learning.

Authors:  Yann LeCun; Yoshua Bengio; Geoffrey Hinton
Journal:  Nature       Date:  2015-05-28       Impact factor: 49.962

7.  Undersampled MR image reconstruction using an enhanced recursive residual network.

Authors:  Lijun Bao; Fuze Ye; Congbo Cai; Jian Wu; Kun Zeng; Peter C M van Zijl; Zhong Chen
Journal:  J Magn Reson       Date:  2019-07-09       Impact factor: 2.229

8.  A deep error correction network for compressed sensing MRI.

Authors:  Liyan Sun; Yawen Wu; Zhiwen Fan; Xinghao Ding; Yue Huang; John Paisley
Journal:  BMC Biomed Eng       Date:  2020-02-27

9.  Optimized fast GPU implementation of robust artificial-neural-networks for k-space interpolation (RAKI) reconstruction.

Authors:  Chi Zhang; Seyed Amir Hossein Hosseini; Sebastian Weingärtner; Kâmil Uǧurbil; Steen Moeller; Mehmet Akçakaya
Journal:  PLoS One       Date:  2019-10-23       Impact factor: 3.240

10.  Accelerated coronary MRI with sRAKI: A database-free self-consistent neural network k-space reconstruction for arbitrary undersampling.

Authors:  Seyed Amir Hossein Hosseini; Chi Zhang; Sebastian Weingärtner; Steen Moeller; Matthias Stuber; Kamil Ugurbil; Mehmet Akçakaya
Journal:  PLoS One       Date:  2020-02-21       Impact factor: 3.240

View more
  2 in total

1.  Residual RAKI: A hybrid linear and non-linear approach for scan-specific k-space deep learning.

Authors:  Chi Zhang; Steen Moeller; Omer Burak Demirel; Kâmil Uğurbil; Mehmet Akçakaya
Journal:  Neuroimage       Date:  2022-04-27       Impact factor: 7.400

2.  A review and experimental evaluation of deep learning methods for MRI reconstruction.

Authors:  Arghya Pal; Yogesh Rathi
Journal:  J Mach Learn Biomed Imaging       Date:  2022-03-11
  2 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.