Deep learning based dictionary learning and tomographic image reconstruction

Abstract

This work presents an approach for image reconstruction in clinical low-dose tomography that combines principles from sparse signal processing with ideas from deep learning. First, we describe sparse signal representation in terms of dictionaries from a statistical perspective and interpret dictionary learning as a process of aligning distribution that arises from a generative model with empirical distribution of true signals. As a result we can see that sparse coding with learned dictionaries resembles a specific variational autoencoder, where the decoder is a linear function and the encoder is a sparse coding algorithm. Next, we show that dictionary learning can also benefit from computational advancements introduced in the context of deep learning, such as parallelism and as stochastic optimization. Finally, we show that regularization by dictionaries achieves competitive performance in computed tomography (CT) reconstruction comparing to state-of-the-art model based and data driven approaches.

arXiv comment34 pages, 5 figures

References in corpus (19)

• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Learned Primal-dual Reconstruction
• Image Reconstruction: From Sparsity to Data-adaptive Methods and Machine Learning
• Inertial Proximal Alternating Linearized Minimization (iPALM) for Nonconvex and Nonsmooth Problems
• Computed Tomography Reconstruction Using Deep Image Prior and Learned Reconstruction Methods
• Convolutional Sparse Coding for Compressed Sensing CT Reconstruction
• Multi-Layer Convolutional Sparse Modeling: Pursuit and Dictionary Learning
• PWLS-ULTRA: An Efficient Clustering and Learning-Based Approach for Low-Dose 3D CT Image Reconstruction
• Deep Dictionary Learning: A PARametric NETwork Approach
• Computationally Efficient Deep Neural Network for Computed Tomography Image Reconstruction
• On the Computational Intractability of Exact and Approximate Dictionary Learning
• Convolutional Analysis Operator Learning: Acceleration and Convergence
• Neural Networks-based Regularization for Large-Scale Medical Image Reconstruction
• Deep Residual Autoencoders for Expectation Maximization-inspired Dictionary Learning
• Rethinking the CSC Model for Natural Images
• SPULTRA: Low-Dose CT Image Reconstruction with Joint Statistical and Learned Image Models
• Low Dose CT Image Reconstruction With Learned Sparsifying Transform
• A convex variational model for learning convolutional image atoms from incomplete data
• Deep synthesis regularization of inverse problems