• Corpus ID: 236881431

Robust Compressed Sensing MRI with Deep Generative Priors

@article{Jalal2021RobustCS,
  title={Robust Compressed Sensing MRI with Deep Generative Priors},
  author={Ajil Jalal and Marius Arvinte and Giannis Daras and Eric Price and Alexandros G. Dimakis and Jonathan I. Tamir},
  journal={ArXiv},
  year={2021},
  volume={abs/2108.01368}
}
The CSGM framework (Bora-Jalal-Price-Dimakis’17) has shown that deep generative priors can be powerful tools for solving inverse problems. However, to date this framework has been empirically successful only on certain datasets (for example, human faces and MNIST digits), and it is known to perform poorly on out-of-distribution samples. In this paper, we present the first successful application of the CSGM framework on clinical MRI data. We train a generative prior on brain scans from the… 

Adaptive Diffusion Priors for Accelerated MRI Reconstruction

—Deep MRI reconstruction is commonly per- formed with conditional models that map undersampled data as input onto fully-sampled data as output. Condi- tional models perform de-aliasing under

Conditional Injective Flows for Bayesian Imaging

TLDR
C-Trumpets are proposed—conditional injective flows specifically designed for imaging problems, which greatly diminish these challenges of ill-posedness, nonlinearity, model mismatch, and noise and enable fast approximation of point estimates like MMSE or MAP as well as physically-meaningful uncertainty quantiflcation.

One-shot Generative Prior Learned from Hankel-k-space for Parallel Imaging Reconstruction

TLDR
Experimental results confirmed that the internal statistics of patches within a single k-space data carry enough information for learning a powerful generative model and provide state-of-the-art reconstruction.

MRI Reconstruction via Data Driven Markov Chain with Joint Uncertainty Estimation

TLDR
A framework that enablescient sampling from learned probability distributions for MRI reconstruction using the Markov chain Monte Carlo method based on a reverse diffusion process to be able to utilize advanced generative models is introduced.

WKGM: Weight-K-space Generative Model for Parallel Imaging Reconstruction

TLDR
Experimental results demonstrate that WKGM can attain state-of-the-art reconstruction results with the well-learned k-space generative prior, and can be synergistically combined with various traditional k- space PI models, generating learn-ing-based priors to produce high-fidelity reconstructions.

MR Image Denoising and Super-Resolution Using Regularized Reverse Diffusion

TLDR
This work proposes a new denoising method based on score-based reverse diffusion sampling, which overcomes all the aforementioned drawbacks and establishes state-of-the-art performance, while having desirable properties which prior MMSE denoisers did not have.

A coherence parameter characterizing generative compressed sensing with Fourier measurements

TLDR
This work proves the first known restricted isometry guarantee for generative compressed sensing with subsampled isometries, and provides recovery bounds with nearly order-optimal sample complexity, addressing an open problem of Scarlett et al. (2022).

Differentiable Gaussianization Layers for Inverse Problems Regularized by Deep Generative Models

TLDR
This work proposes to reparameterize and Gaussianize the latent vector using novel differentiable data-dependent layers wherein custom operators are defined by solving optimization problems to constrain an inversion to feasible in-distribution solutions.

Test-Time Training Can Close the Natural Distribution Shift Performance Gap in Deep Learning Based Compressed Sensing

TLDR
This work proposes a domain adaptation method for deep learning based compressive sensing that relies on self-supervision during training paired with test-time training at inference that essentially closes the distribution shift performance gap for state-of-the-art architectures for accelerated MRI.

Deep Model-Based Architectures for Inverse Problems under Mismatched Priors

There is a growing interest in deep model-based architectures (DMBAs) for solving imaging inverse problems by combining physical measurement models and learned image priors specified using

References

SHOWING 1-10 OF 92 REFERENCES

Uncertainty Quantification in Deep MRI Reconstruction

TLDR
This study aims to quantify the uncertainty in image recovery with DL models by leveraging variational autoencoders (VAEs) to develop a probabilistic reconstruction scheme that maps out (low-quality) short scans with aliasing artifacts to the diagnostic-quality ones.

Measuring Robustness in Deep Learning Based Compressive Sensing

TLDR
The robustness of different approaches for image reconstruction including trained and un-trained neural networks as well as traditional sparsity-based methods is measured and it is found that both trained andun-trained methods are vulnerable to adversarial perturbations.

Invertible generative models for inverse problems: mitigating representation error and dataset bias

TLDR
It is demonstrated that invertible neural networks, which have zero representation error by design, can be effective natural signal priors at inverse problems such as denoising, compressive sensing, and inpainting.

Inverse GANs for accelerated MRI reconstruction

TLDR
An iterative reconstruction approach that incorporates the implicit prior provided by a generative adversarial network (GAN), which learns the probability distribution of uncorrupted MRI data in an off-line step, that is evaluated on multi-coil knee MRI data and compared to a classical as well as a deep learning-based approach.

Inference With Deep Generative Priors in High Dimensions

TLDR
This paper shows that the performance of ML-VAMP can be exactly predicted in a certain high-dimensional random limit, and provides a computationally efficient method for multi-layer inference with an exact performance characterization and testable conditions for optimality in the large-system limit.

Deep Generative Adversarial Neural Networks for Compressive Sensing MRI

TLDR
A novel CS framework that uses generative adversarial networks (GAN) to model the (low-dimensional) manifold of high-quality MR images that retrieves higher quality images with improved fine texture details compared with conventional Wavelet-based and dictionary- learning-based CS schemes as well as with deep-learning-based schemes using pixel-wise training.

Transform Learning for Magnetic Resonance Image Reconstruction: From Model-Based Learning to Building Neural Networks

TLDR
A unified framework for incorporating various TL-based models is presented and the connections between TL and convolutional or filter-bank models and corresponding multilayer extensions, with connections to deep learning are discussed.

Task-Aware Compressed Sensing with Generative Adversarial Networks

TLDR
This paper uses Generative Adversarial Networks (GANs) to impose structure in compressed sensing problems, replacing the usual sparsity constraint, and proposes to train the GANs in a task-aware fashion, specifically for reconstruction tasks.

Robust compressed sensing using generative models

TLDR
This paper proposes an algorithm inspired by the Median-of-Means (MOM) that guarantees recovery for heavy-tailed data, even in the presence of outliers, and results show the novel MOM-based algorithm enjoys the same sample complexity guarantees as ERM under sub-Gaussian assumptions.

A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstruction

TLDR
A framework for reconstructing dynamic sequences of 2-D cardiac magnetic resonance images from undersampled data using a deep cascade of convolutional neural networks (CNNs) to accelerate the data acquisition process is proposed and it is demonstrated that CNNs can learn spatio-temporal correlations efficiently by combining convolution and data sharing approaches.
...