Log In Sign Up

Robust Compressed Sensing MRI with Deep Generative Priors

by   Ajil Jalal, et al.

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 fastMRI dataset, and show that posterior sampling via Langevin dynamics achieves high quality reconstructions. Furthermore, our experiments and theory show that posterior sampling is robust to changes in the ground-truth distribution and measurement process. Our code and models are available at: <>.


page 22

page 23

page 24

page 25

page 26

page 27

page 28

page 32


Instance-Optimal Compressed Sensing via Posterior Sampling

We characterize the measurement complexity of compressed sensing of sign...

Adaptive Compressed Sensing MRI with Unsupervised Learning

In compressed sensing MRI, k-space measurements are under-sampled to ach...

Compressed Sensing MRI Reconstruction Regularized by VAEs with Structured Image Covariance

Learned regularization for MRI reconstruction can provide complex data-d...

Neural Network-based Reconstruction in Compressed Sensing MRI Without Fully-sampled Training Data

Compressed Sensing MRI (CS-MRI) has shown promise in reconstructing unde...

Accelerating Inverse Learning via Intelligent Localization with Exploratory Sampling

In the scope of "AI for Science", solving inverse problems is a longstan...

Quantized Compressed Sensing with Score-Based Generative Models

We consider the general problem of recovering a high-dimensional signal ...

Self-Validation: Early Stopping for Single-Instance Deep Generative Priors

Recent works have shown the surprising effectiveness of deep generative ...