Deep Bayesian Inversion

Abstract

Characterizing statistical properties of solutions of inverse problems is essential for decision making. Bayesian inversion offers a tractable framework for this purpose, but current approaches are computationally unfeasible for most realistic imaging applications in the clinic. We introduce two novel deep learning based methods for solving large-scale inverse problems using Bayesian inversion: a sampling based method using a WGAN with a novel mini-discriminator and a direct approach that trains a neural network using a novel loss function. The performance of both methods is demonstrated on image reconstruction in ultra low dose 3D helical CT. We compute the posterior mean and standard deviation of the 3D images followed by a hypothesis test to assess whether a "dark spot" in the liver of a cancer stricken patient is present. Both methods are computationally efficient and our evaluation shows very promising performance that clearly supports the claim that Bayesian inversion is usable for 3D imaging in time critical applications.

References in corpus (11)

• Conditional Generative Adversarial Nets
• Deep D-bar: Real time Electrical Impedance Tomography Imaging with Deep Neural Networks
• Image Transformer
• Neural Optimizer Search with Reinforcement Learning
• Deep Generative Adversarial Networks for Compressed Sensing Automates MRI
• Banach Wasserstein GAN
• Conditional Neural Processes
• Discretization-invariant Bayesian inversion and Besov space priors
• Task adapted reconstruction for inverse problems
• Neural Processes
• Learning to solve inverse problems using Wasserstein loss

Cited by in corpus (24)

• TomoGAN: Low-Dose Synchrotron X-Ray Tomography with Generative Adversarial Networks
• High Perceptual Quality Image Denoising with a Posterior Sampling CGAN
• Solving Bayesian Inverse Problems via Variational Autoencoders
• Stabilizing Invertible Neural Networks Using Mixture Models
• Uncertainty Quantification with Generative Models
• Big in Japan: Regularizing networks for solving inverse problems
• Bayesian Inference with Generative Adversarial Network Priors
• Learned imaging with constraints and uncertainty quantification
• Optimal Experimental Design for Uncertain Systems Based on Coupled Differential Equations
• Bayesian Image Reconstruction using Deep Generative Models
• Interval Neural Networks: Uncertainty Scores
• Deep Learning Techniques for Inverse Problems in Imaging
• Variational Inference for Computational Imaging Inverse Problems
• Quantifying Sources of Uncertainty in Deep Learning-Based Image Reconstruction
• Conditional Normalizing Flows for Low-Dose Computed Tomography Image Reconstruction
• Quantifying Model Uncertainty in Inverse Problems via Bayesian Deep Gradient Descent
• Embodied Self-supervised Learning by Coordinated Sampling and Training
• Application of Deep Learning-based Interpolation Methods to Nearshore Bathymetry
• GAN-based Priors for Quantifying Uncertainty
• Digital phase-only holography using deep conditional generative models
• Tomographic Auto-Encoder: Unsupervised Bayesian Recovery of Corrupted Data
• Sampling possible reconstructions of undersampled acquisitions in MR imaging
• About exchanging expectation and supremum for conditional Wasserstein GANs
• Learning to Predict Error for MRI Reconstruction