Learning a Variational Network for Reconstruction of Accelerated MRI Data

By Kerstin Hammernik, Teresa Klatzer, Erich Kobler, Michael P. Recht, Daniel K. Sodickson, Thomas Pock, Florian Knoll. 2017

1,801 citations recorded by the source corpus.

To allow fast and high‐quality reconstruction of clinical accelerated multi‐coil MR data by learning a variational network that combines the mathematical structure of variational models with deep learning.

Subfields: Computational Mechanics, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data (2019, background): “fine-tuning the trained surrogate on test input [83, 84], learned gradient update [85, 86], meta-learning on a distribution of regression tasks [87], etc.”
Deep Residual Learning for Accelerated MRI Using Magnitude and Phase Networks (2018, methodology): “A multilayer perceptron was proposed for accelerated pMRI [20], and deep CNN architecture using an unfolded iterative CS algorithm was proposed in [21], where trained optimal regularizers were proposed in contrast to the usual approach using handcrafted regularizers [21].”
Assessment of the generalization of learned image reconstruction and the potential for transfer learning (2019, methodology): “We proposed a learning approach based on the framework of variational optimization with the goal of learning the complete reconstruction procedure, which maps from multichannel k-space raw data to image space, and the associated numerical procedure.(5,6) Since then, a substantial increase of develop”
Modern regularization methods for inverse problems (2018, methodology): “These cover the finitedimensional Markov random field models proposed in [325, 346, 143, 124, 334], the optimal model design approaches in [199, 198, 65, 40], the optimal regularization parameter estimation in variational regularization [89, 128, 137, 138, 90, 127], to training optimal operators in ”
Deep-Learning Methods for Parallel Magnetic Resonance Imaging Reconstruction: A Survey of the Current Approaches, Trends, and Issues (2019, methodology): “An illustration of image space based machine learning for parallel MRI along the lines of [43, 44] is shown in Figure 2.”
Deep learning for undersampled MRI reconstruction (2017, result): “It should be mentioned that, at the time of writing this paper, we found that there is another work that proposes almost the same approach as [5], [9], [18].”
Global Guarantees for Enforcing Deep Generative Priors by Empirical Risk (2017, background): “These more sophisticated priors are recently emerging in empirical applications of many fields of imaging, such as medical imaging [70, 30, 24, 55, 17, 1, 49, 68, 53, 43, 32, 42, 16, 69], microscopy [59], inpainting [45, 71], superresolution [60, 33, 28], compressed sensing [38, 37, 51, 50], image m”
Deep Learning Techniques for Inverse Problems in Imaging (2020, background): “deep neural networks can leverage large collections of training data to directly compute regularized reconstructions across a host of computational imaging tasks [16]–[24].”
Learned Primal-Dual Reconstruction (2017, background): “Similar approaches for MRI reconstruction have also been considered [14], [15].”
Quantitative susceptibility mapping using deep neural network: QSMnet (2018, methodology): “Deep neural networks have also been applied for medical image reconstruction such as X-ray, CT, PET, and MRI (Hammernik et al., 2018; Kwon et al., 2016; Mardani et al., 2017; Prasoon et al., 2013; Teramoto et al., 2016; Xu et al., 2017; Yang et al., 2018; Zhu et al., 2017; Zichuan et al., 2017).”
Monarch: Expressive Structured Matrices for Efficient and Accurate Training (2022, background): “, 2015) or unrolled networks (Sandino et al., 2020; Hammernik et al., 2018)) learn a mapping between paired zero-filled and fully-sampled, ground truth images.”
Deep Generative Adversarial Neural Networks for Compressive Sensing (GANCS) MRI (2018, background): “The work in [10] also uses a variational autoencoder for learning the priors, which is then evaluated on knee images.”
On instabilities of deep learning in image reconstruction and the potential costs of AI (2019, methodology): “MRI-VN [12]: This network is for medium to high resolution parallel MRI with 15 coil elements and 15% subsampling.”
Super-Resolution Musculoskeletal MRI Using Deep Learning (2018, background): “53,54 DeepResolve could also be generalized to in-plane resolution enhancements in addition to through-plane resolution enhancements.”