Unsupervised Deep Basis Pursuit: Learning inverse problems without ground-truth data

Abstract

Basis pursuit is a compressed sensing optimization in which the l1-norm is minimized subject to model error constraints. Here we use a deep neural network prior instead of l1-regularization. Using known noise statistics, we jointly learn the prior and reconstruct images without access to ground-truth data. During training, we use alternating minimization across an unrolled iterative network and jointly solve for the neural network weights and training set image reconstructions. At inference, we fix the weights and pass the measurements through the network. We compare reconstruction performance between unsupervised and supervised (i.e. with ground-truth) methods. We hypothesize this technique could be used to learn reconstruction when ground-truth data are unavailable, such as in high-resolution dynamic MRI.

https://deep-inverse.org/

References in corpus (3)

• On instabilities of deep learning in image reconstruction - Does AI come at a cost?
• Unrolled Optimization with Deep Priors
• Multi-Layer Convolutional Sparse Modeling: Pursuit and Dictionary Learning

Cited by in corpus (8)

• Self-Supervised Learning of Physics-Guided Reconstruction Neural Networks without Fully-Sampled Reference Data
• Dense Recurrent Neural Networks for Accelerated MRI: History-Cognizant Unrolling of Optimization Algorithms
• Multi-Mask Self-Supervised Learning for Physics-Guided Neural Networks in Highly Accelerated MRI
• Unsupervised MRI Reconstruction with Generative Adversarial Networks
• Analysis of Deep Complex-Valued Convolutional Neural Networks for MRI Reconstruction
• Scan-specific Self-supervised Bayesian Deep Non-linear Inversion for Undersampled MRI Reconstruction
• Memory-efficient Learning for High-Dimensional MRI Reconstruction
• Reconstructing unseen modalities and pathology with an efficient Recurrent Inference Machine