RARE: Image Reconstruction using Deep Priors Learned without Ground Truth

Abstract

Regularization by denoising (RED) is an image reconstruction framework that uses an image denoiser as a prior. Recent work has shown the state-of-the-art performance of RED with learned denoisers corresponding to pre-trained convolutional neural nets (CNNs). In this work, we propose to broaden the current denoiser-centric view of RED by considering priors corresponding to networks trained for more general artifact-removal. The key benefit of the proposed family of algorithms, called regularization by artifact-removal (RARE), is that it can leverage priors learned on datasets containing only undersampled measurements. This makes RARE applicable to problems where it is practically impossible to have fully-sampled groundtruth data for training. We validate RARE on both simulated and experimentally collected data by reconstructing a free-breathing whole-body 3D MRIs into ten respiratory phases from heavily undersampled k-space measurements. Our results corroborate the potential of learning regularizers for iterative inversion directly on undersampled and noisy measurements.

In press for IEEE Journal of Special Topics in Signal Processing

References in corpus (29)

• Adam: A Method for Stochastic Optimization
• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)
• MoDL: Model Based Deep Learning Architecture for Inverse Problems
• Noise2Noise: Learning Image Restoration without Clean Data
• A deep convolutional neural network using directional wavelets for low-dose X-ray CT reconstruction
• A Review of Convolutional Neural Networks for Inverse Problems in Imaging
• Deep-STORM: super-resolution single-molecule microscopy by deep learning
• Denoising Prior Driven Deep Neural Network for Image Restoration
• Deep speckle correlation: a deep learning approach towards scalable imaging through scattering media
• A Deep Learning Approach to Structured Signal Recovery
• Plug and play methods for magnetic resonance imaging (long version)
• Learning Proximal Operators: Using Denoising Networks for Regularizing Inverse Imaging Problems
• An Online Plug-and-Play Algorithm for Regularized Image Reconstruction
• Efficient and accurate inversion of multiple scattering with deep learning
• Learning Deep CNN Denoiser Prior for Image Restoration
• Deep Learning Diffuse Optical Tomography
• Noise2Self: Blind Denoising by Self-Supervision
• Plug-and-Play Methods Provably Converge with Properly Trained Denoisers
• Dense Recurrent Neural Networks for Accelerated MRI: History-Cognizant Unrolling of Optimization Algorithms
• Learning optimal nonlinearities for iterative thresholding algorithms
• High-Resolution Limited-Angle Phase Tomography of Dense Layered Objects Using Deep Neural Networks
• Regularized Fourier Ptychography using an Online Plug-and-Play Algorithm
• DeepRED: Deep Image Prior Powered by RED
• Learning-based Image Reconstruction via Parallel Proximal Algorithm
• A New Recurrent Plug-and-Play Prior Based on the Multiple Self-Similarity Network
• Image Restoration and Reconstruction using Variable Splitting and Class-adapted Image Priors
• Boosting CNN beyond Label in Inverse Problems

Cited by in corpus (33)

• Self-Supervised Learning of Physics-Guided Reconstruction Neural Networks without Fully-Sampled Reference Data
• Adaptive Diffusion Priors for Accelerated MRI Reconstruction
• Plug-and-Play Methods for Integrating Physical and Learned Models in Computational Imaging
• Noise2Inverse: Self-supervised deep convolutional denoising for tomography
• Image Compressed Sensing Using Non-local Neural Network
• Unsupervised Deep Learning Methods for Biological Image Reconstruction and Enhancement
• Multi-Mask Self-Supervised Learning for Physics-Guided Neural Networks in Highly Accelerated MRI
• PUERT: Probabilistic Under-sampling and Explicable Reconstruction Network for CS-MRI
• Adaptive 3D descattering with a dynamic synthesis network
• SGD-Net: Efficient Model-Based Deep Learning with Theoretical Guarantees
• Unsupervised MRI Reconstruction with Generative Adversarial Networks
• Self-Supervised Physics-Guided Deep Learning Reconstruction For High-Resolution 3D LGE CMR
• CoIL: Coordinate-based Internal Learning for Imaging Inverse Problems
• Learned convex regularizers for inverse problems
• Accelerated deep self-supervised ptycho-laminography for three-dimensional nanoscale imaging of integrated circuits
• Data and Physics driven Deep Learning Models for Fast MRI Reconstruction: Fundamentals and Methodologies
• An untrained deep learning method for reconstructing dynamic magnetic resonance images from accelerated model-based data
• Scan-specific Self-supervised Bayesian Deep Non-linear Inversion for Undersampled MRI Reconstruction
• Learning Task-Specific Strategies for Accelerated MRI
• A Tutorial on MRI Reconstruction: From Modern Methods to Clinical Implications
• Self-supervised Neural Networks for Spectral Snapshot Compressive Imaging
• Validation and Generalizability of Self-Supervised Image Reconstruction Methods for Undersampled MRI
• Self-Supervised Adversarial Diffusion Models for Fast MRI Reconstruction
• Recovery Analysis for Plug-and-Play Priors using the Restricted Eigenvalue Condition
• Deep Image Reconstruction using Unregistered Measurements without Groundtruth
• Enhancing quality and speed in database-free neural network reconstructions of undersampled MRI with SCAMPI
• Joint Reconstruction and Calibration using Regularization by Denoising
• Adaptive convolutional neural networks for k-space data interpolation in fast magnetic resonance imaging
• Unsupervised Learning From Incomplete Measurements for Inverse Problems
• Improved Supervised Training of Physics-Guided Deep Learning Image Reconstruction with Multi-Masking
• Equivariant Imaging: Learning Beyond the Range Space
• Interpretation of Plug-and-Play (PnP) algorithms from a different angle
• Fully Unsupervised Dynamic MRI Reconstruction via Diffeo-Temporal Equivariance