Iterative training of robust k-space interpolation networks for improved image reconstruction with limited scan specific training samples.
GRAPPA
RAKI
complex-valued machine learning
data augmentation
deep learning
parallel imaging
Journal
Magnetic resonance in medicine
ISSN: 1522-2594
Titre abrégé: Magn Reson Med
Pays: United States
ID NLM: 8505245
Informations de publication
Date de publication:
02 2023
02 2023
Historique:
revised:
12
09
2022
received:
18
07
2022
accepted:
20
09
2022
pubmed:
14
10
2022
medline:
2
12
2022
entrez:
13
10
2022
Statut:
ppublish
Résumé
To evaluate an iterative learning approach for enhanced performance of robust artificial-neural-networks for k-space interpolation (RAKI), when only a limited amount of training data (auto-calibration signals [ACS]) are available for accelerated standard 2D imaging. In a first step, the RAKI model was tailored for the case of limited training data amount. In the iterative learning approach (termed iterative RAKI [iRAKI]), the tailored RAKI model is initially trained using original and augmented ACS obtained from a linear parallel imaging reconstruction. Subsequently, the RAKI convolution filters are refined iteratively using original and augmented ACS extracted from the previous RAKI reconstruction. Evaluation was carried out on 200 retrospectively undersampled in vivo datasets from the fastMRI neuro database with different contrast settings. For limited training data (18 and 22 ACS lines for R = 4 and R = 5, respectively), iRAKI outperforms standard RAKI by reducing residual artifacts and yields better noise suppression when compared to standard parallel imaging, underlined by quantitative reconstruction quality metrics. Additionally, iRAKI shows better performance than both GRAPPA and standard RAKI in case of pre-scan calibration with varying contrast between training- and undersampled data. RAKI benefits from the iterative learning approach, which preserves the noise suppression feature, but requires less original training data for the accurate reconstruction of standard 2D images thereby improving net acceleration.
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
812-827Informations de copyright
© 2022 The Authors. Magnetic Resonance in Medicine published by Wiley Periodicals LLC on behalf of International Society for Magnetic Resonance in Medicine.
Références
Roemer PB, Edelstein WA, Hayes CE, Souza SP, Mueller OM. The NMR phased array. Magn Reson Med. 1990;16:192-225.
Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med. 1999;42:952-962.
Griswold MA, Jakob PM, Heidemann RM, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med. 2002;47:1202-1210.
Akçakaya M, Moeller S, Weingärtner S, Uğurbil K. Scan-specific robust artificial-neural-networks for k-space interpolation (RAKI) reconstruction: database-free deep learning for fast imaging. Magn Reson Med. 2019;81:439-453.
Zhang C, Hosseini SAH, Weingärtner S, Ugurbil K, Moeller S, Akçakaya M. Optimized fast GPU implementation of robust artificial-neural-networks for k-space interpolation (RAKI) reconstruction. PLoS ONE. 2019;14:e0223315.
Zhang C, Moeller S, Demire OB, Ugurbil K, Akcakaya M. Residual RAKI: a hybrid linear and non-linear approach for scan-specific k-space deep learning. Neuroimage. 2022;256:119248.
Taylor L, Nitschke G. Improving deep learning with generic data augmentation. Proceedings of the 2018 IEEE Symposium Series on Computational Intelligence (SSCI); 2018:1542-1547.
Shorten C, Khoshgoftaar TM. A survey on image data augmentation for deep learning. Journal of Big Data. 2019;6:60.
Jakob PM, Griswold MA, Edelman RR, Sodickson DK. AUTO-SMASH: a selfcalibrating technique for SMASH imaging. SiMultaneous Acquisition of Spatial Harmonics MAGMA. 1998;7:42-54.
Heidemann RM, Griswold MA, Haase A, Jakob PM. VD-AUTO-SMASH imaging. Magn Reson Med. 2001;45:1066-1074.
Lustig M, Donoho D, Pauly JM. Sparse MRI: the application of compressed sensing for rapid MR imaging. Magn Reson Med. 2007;58:1182-1195.
Knoll F, Hammernik K, Zhang C, et al. Deep-learning methods for parallel magnetic resonance imaging reconstruction: a survey of the current approaches, trends, and issues. IEEE Signal Processing Magazine. 2020;37:128-140.
Dawood P, Blaimer M, Breuer F, Jakob PM, Oberberger J, et al. Iterative RAKI with complex-valued convolution for improved image reconstruction with limited training samples. Proceedings of the 30th Annual Meeting of ISMRM; 2022 Abstract Program Number 1193.
Virtue P. Complex-Valued Deep Learning with Applications to Magnetic Resonance Image Synthesis. Doctoral Dissertation. University of California at Berkeley; 2019.
Cole E, Cheng J, Pauly J, Vasanawala S. Analysis of deep complex-valued convolutional neural networks for MRI reconstruction and phase-focused applications. Magn Reson Med. 2021;86:1093-1109.
Trabelsi C, Bilaniuk O, Zhang Y, et al. Deep Complex Networks. Proceedings of the 6th International Conference on Learning Representations, (ICLR 2018), Conference Track Proceedings; 2018.
Nair V, Hinton GE. Rectified linear units improve restricted Boltzmann machines. Proceedings of the 27th International Conference on International Conference on Machine Learning ICML'10; 2010:807-814.
LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521:436-444.
Kingma DP, Ba J. Adam: a method for stochastic optimization. Proceedings of 3rd International Conference on Learning Representations. ICLR; 2015.
Paszke A, Gross S, Massa F, et al. PyTorch: an imperative style, high-performance deep learning library. Proceedings of the 33rd Conference on Neural Information Processing Systems; 2019.
Knoll F, Murrell T, Sriram A, et al. Advancing machine learning for MR image reconstruction with an open competition: overview of the 2019 fastMRI challenge. Magn Reson Med. 2020;84:3054-3070.
Bauer S, Markl M, Honal M, Jung BA. The effect of reconstruction and acquisition parameters for GRAPPA-based parallel imaging on the image quality. Magn Reson Med. 2011;66:402-409.
Sandino CM, Lai P, Vasanawala SS, Cheng JY. Accelerating cardiac cine MRI using a deep learning-based ESPIRiT reconstruction. Magn Reson Med. 2021;85:152-167.
Nencka AS, Arpinar VE, Bhave S, et al. Split-slice training and hyperparameter tuning of RAKI networks for simultaneous multi-slice reconstruction. Magn Reson Med. 2021;85:3272-3280.
Zhao T, Hu X. Iterative GRAPPA (iGRAPPA) for improved parallel imaging reconstruction. Magn Reson Med. 2008;59:903-907.
Wang Z, Bovik AC, Sheikh HR, Simoncelli EP. Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process. 2004;13:600-612.
Uecker M, Lai P, Murphy MJ, et al. ESPIRiT-an eigenvalue approach to autocalibrating parallel MRI: where SENSE meets GRAPPA. Magn Reson Med. 2014;71:990-1001.
Blaimer M, Gutberlet M, Kellman P, Breuer F, Koestler H, Griswold M. Virtual coil concept for improved parallel MRI employing conjugate symmetric signals. Magn Reson Med. 2009;61:93-102.
Hammernik K, Klatzer T, Kobler E, Pock T, Knoll F, et al. Learning a variational network for reconstruction of accelerated MRI data. Magn Reson Med. 2018;79:3055-3071.
Uecker M, Ong F, Tamir JI, Bahri D, Virtue P, et al. Berkeley advanced reconstruction toolbox. Proceedings of the 23th Annual Meeting of ISMRM. Vol 23; 2015:2486.
Lustig M, Pauly JM. SPIRiT: iterative self-consistent parallel imaging reconstruction from arbitrary k-space. Magn Reson Med. 2010;64:457-471.
Ding Y, Xue H, Ahmad R, Chang T-c, Ting ST, Simonetti OP. Paradoxical effect of the signal-to-noise ratio of GRAPPA calibration lines: a quantitative study. Magn Reson Med. 2015;74:231-239.
Arefeen Y, Beker O, Cho J, Yu H, Adalsteinsson E, Bilgic B. Scan-specific artifact reduction in k-space (SPARK) neural networks synergize with physics-based reconstruction to accelerate MRI. Magn Reson Med. 2022;87:764-780.
Haldar JP. Low-rank modeling of local k-space neighborhoods (LORAKS) for constrained MRI. IEEE Trans Med Imaging. 2014;33:668-681.
Kim TH, Garg P, Haldar JP. LORAKI: Reconstruction of undersampled k-space data using scan-specific autocalibrated recurrent neural networks. Proceedings of the 27th Annual Meeting of ISMRM; 2019 Abstract Program Number 4647.
Dawood P, Blaimer M, Jakob PM, Oberberger J. Influence of training data on RAKI reconstruction quality in standard 2D imaging. Proceedings of the 29th Annual Meeting of ISMRM, (Virtual Meeting); 2021 Abstract Program Number 1961.
Weiger M, Pruessman KP, Boesiger P. 2D SENSE for faster 3D MRI. MAGMA. 2002;14:10-19.
Breuer FA, Blaimer M, Mueller MF, et al. Controlled aliasing in volumetric parallel imaging (2D CAIPIRINHA). Magn Reson Med. 2006;55:549-556.
Bilgic B, Gagoski BA, Cauley SF, Fan CA, Polimeni JR, et al. Wave-CAIPI for highly accelerated 3D imaging. Magn Reson Med. 2015;73:2152-2162.
Bilgic B, Kim TH, Liao C, et al. Improving parallel imaging by jointly reconstructing multi-contrast data. Magn Reson Med. 2018;80:619-632.
Breuer F, Kannengiesser S, Blaimer M, Seiberlich N, Jakob P, Griswold M. General formulation for quantitative G-factor calculation in GRAPPA reconstructions. Magn Reson Med. 2009;62:739-746.
Hammernik K, Schlemper J, Qin C, Duan J, Summers RM, Rueckert D. Systematic evaluation of iterative deep neural networks for fast parallel MRI reconstruction with sensitivity-weighted coil combination. Magn Reson Med. 2021;86:1859-1872.