MR image reconstruction using deep density priors

Abstract

Algorithms for Magnetic Resonance (MR) image reconstruction from undersampled measurements exploit prior information to compensate for missing k-space data. Deep learning (DL) provides a powerful framework for extracting such information from existing image datasets, through learning, and then using it for reconstruction. Leveraging this, recent methods employed DL to learn mappings from undersampled to fully sampled images using paired datasets, including undersampled and corresponding fully sampled images, integrating prior knowledge implicitly. In this article, we propose an alternative approach that learns the probability distribution of fully sampled MR images using unsupervised DL, specifically Variational Autoencoders (VAE), and use this as an explicit prior term in reconstruction, completely decoupling the encoding operation from the prior. The resulting reconstruction algorithm enjoys a powerful image prior to compensate for missing k-space data without requiring paired datasets for training nor being prone to associated sensitivities, such as deviations in undersampling patterns used in training and test time or coil settings. We evaluated the proposed method with T1 weighted images from a publicly available dataset, multi-coil complex images acquired from healthy volunteers (N=8) and images with white matter lesions. The proposed algorithm, using the VAE prior, produced visually high quality reconstructions and achieved low RMSE values, outperforming most of the alternative methods on the same dataset. On multi-coil complex data, the algorithm yielded accurate magnitude and phase reconstruction results. In the experiments on images with white matter lesions, the method faithfully reconstructed the lesions. Keywords: Reconstruction, MRI, prior probability, machine learning, deep learning, unsupervised learning, density estimation

Published in IEEE TMI. Main text and supplementary material, 19 pages total

References in corpus (7)

• Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoders
• Recurrent Inference Machines for Solving Inverse Problems
• On learning optimized reaction diffusion processes for effective image restoration
• Tackling Over-pruning in Variational Autoencoders
• Recurrent Generative Adversarial Networks for Proximal Learning and Automated Compressive Image Recovery
• Bayesian Image Quality Transfer with CNNs: Exploring Uncertainty in dMRI Super-Resolution
• Solving Linear Inverse Problems Using GAN Priors: An Algorithm with Provable Guarantees

Cited by in corpus (21)

• On instabilities of deep learning in image reconstruction - Does AI come at a cost?
• Adaptive Diffusion Priors for Accelerated MRI Reconstruction
• Bayesian MRI Reconstruction with Joint Uncertainty Estimation using Diffusion Models
• MRI Reconstruction Using Deep Bayesian Estimation
• VS-Net: Variable splitting network for accelerated parallel MRI reconstruction
• A Tutorial on MRI Reconstruction: From Modern Methods to Clinical Implications
• Robust multi-coil MRI reconstruction via self-supervised denoising
• NPB-REC: A Non-parametric Bayesian Deep-learning Approach for Undersampled MRI Reconstruction with Uncertainty Estimation
• Complementary Time-Frequency Domain Networks for Dynamic Parallel MR Image Reconstruction
• Joint Intensity-Gradient Guided Generative Modeling for Colorization
• Self-Supervised Adversarial Diffusion Models for Fast MRI Reconstruction
• Noise Level Adaptive Diffusion Model for Robust Reconstruction of Accelerated MRI
• Bayesian Learning of Probabilistic Dipole Inversion for Quantitative Susceptibility Mapping
• Deep learning for fast MR imaging: a review for learning reconstruction from incomplete k-space data
• Generative Priors for MRI Reconstruction Trained from Magnitude-Only Images Using Phase Augmentation
• Homotopic Gradients of Generative Density Priors for MR Image Reconstruction
• Learning Priors in High-frequency Domain for Inverse Imaging Reconstruction
• Cine cardiac MRI reconstruction using a convolutional recurrent network with refinement
• Sampling possible reconstructions of undersampled acquisitions in MR imaging
• Joint Motion Correction and Super Resolution for Cardiac Segmentation via Latent Optimisation
• Reconstructing unseen modalities and pathology with an efficient Recurrent Inference Machine