Tensor-based formulation and nuclear norm regularization for multi-energy computed tomography

Abstract

The development of energy selective, photon counting X-ray detectors allows for a wide range of new possibilities in the area of computed tomographic image formation. Under the assumption of perfect energy resolution, here we propose a tensor-based iterative algorithm that simultaneously reconstructs the X-ray attenuation distribution for each energy. We use a multi-linear image model rather than a more standard "stacked vector" representation in order to develop novel tensor-based regularizers. Specifically, we model the multi-spectral unknown as a 3-way tensor where the first two dimensions are space and the third dimension is energy. This approach allows for the design of tensor nuclear norm regularizers, which like its two dimensional counterpart, is a convex function of the multi-spectral unknown. The solution to the resulting convex optimization problem is obtained using an alternating direction method of multipliers (ADMM) approach. Simulation results shows that the generalized tensor nuclear norm can be used as a stand alone regularization technique for the energy selective (spectral) computed tomography (CT) problem and when combined with total variation regularization it enhances the regularization capabilities especially at low energy images where the effects of noise are most prominent.

References in corpus (2)

• A Singular Value Thresholding Algorithm for Matrix Completion
• Cine cone beam CT reconstruction using low-rank matrix factorization: algorithm and a proof-of-princple study

Cited by in corpus (51)

• Square Deal: Lower Bounds and Improved Relaxations for Tensor Recovery
• Guaranteed Tensor Recovery Fused Low-rankness and Smoothness
• Multi-dimensional imaging data recovery via minimizing the partial sum of tubal nuclear norm
• Tensor Analysis and Fusion of Multimodal Brain Images
• Non-local Low-rank Cube-based Tensor Factorization for Spectral CT Reconstruction
• Laplacian Convolutional Representation for Traffic Time Series Imputation
• Dynamic Graph Convolutional Networks Using the Tensor M-Product
• Tensor Robust Principal Component Analysis with A New Tensor Nuclear Norm
• Systematic Review on Learning-based Spectral CT
• Polar $n$-Complex and $n$-Bicomplex Singular Value Decomposition and Principal Component Pursuit
• Fourth-order Tensors with Multidimensional Discrete Transforms
• Provable Tensor-Train Format Tensor Completion by Riemannian Optimization
• Spectral CT Reconstruction via Low-rank Representation and Structure Preserving Regularization
• A joint reconstruction and lambda tomography regularization technique for energy-resolved X-ray imaging
• Tuning-free Plug-and-Play Proximal Algorithm for Inverse Imaging Problems
• Exact tensor completion using t-SVD
• Low-Tubal-Rank Tensor Recovery via Factorized Gradient Descent
• Epigraphical Relaxation for Minimizing Layered Mixed Norms
• On Unifying Multi-View Self-Representations for Clustering by Tensor Multi-Rank Minimization
• Generalized Tensor Function via the Tensor Singular Value Decomposition based on the T-Product
• A Fast Algorithm for Cosine Transform Based Tensor Singular Value Decomposition
• Fast and Accurate Tensor Completion with Total Variation Regularized Tensor Trains
• A tomographic workflow to enable deep learning for X-ray based foreign object detection
• Subspace Estimation from Unbalanced and Incomplete Data Matrices: ${\ell }_{2,\infty }$ Statistical Guarantees
• Uconnect: Synergistic Spectral CT Reconstruction with U-Nets Connecting the Energy bins
• Cross: Efficient Low-rank Tensor Completion
• Truncated nuclear norm regularization for low-rank tensor completion
• TFPnP: Tuning-free Plug-and-Play Proximal Algorithm with Applications to Inverse Imaging Problems
• T product Tensors Part I: Inequalities
• Effective Image Retrieval via Multilinear Multi-index Fusion
• T-Jordan Canonical Form and T-Drazin Inverse based on the T-Product
• A Randomized Tensor Singular Value Decomposition based on the t-product
• Low-Rank Tensor Completion by Truncated Nuclear Norm Regularization
• Tensor completion using enhanced multiple modes low-rank prior and total variation
• MetaInv-Net: Meta Inversion Network for Sparse View CT Image Reconstruction
• Generalized T-product Tensor Bernstein Bounds
• An Orthogonal Equivalence Theorem for Third Order Tensors
• Low-dose spectral CT reconstruction using L0 image gradient and tensor dictionary
• $p$-order Tensor Products with Invertible Linear Transforms
• Randomized Kaczmarz for Tensor Linear Systems
• Arbitrarily large iterative tomographic reconstruction on multiple GPUs using the TIGRE toolbox
• Efficient Algorithms for Convolutional Inverse Problems in Multidimensional Imaging
• Efficient edge-preserving methods for dynamic inverse problems
• A note on perturbation analysis for T-product based tensor singular values
• T-product Tensor Expander Chernoff Bound
• Regularized Kaczmarz Algorithms for Tensor Recovery
• Tensor Kernel Recovery for Spatio-Temporal Hawkes Processes
• T-positive semidefiniteness of third-order symmetric tensors and T-semidefinite programming
• Fast Randomized Algorithms for t-Product Based Tensor Operations and Decompositions with Applications to Imaging Data
• A Splitting-Based Iterative Algorithm for GPU-Accelerated Statistical Dual-Energy X-Ray CT Reconstruction
• Sketch-and-project methods for tensor linear systems