Computed tomography reconstruction using deep image prior and learned reconstruction methods
From MaRDI portal
Abstract: In this work, we investigate the application of deep learning methods for computed tomography in the context of having a low-data regime. As motivation, we review some of the existing approaches and obtain quantitative results after training them with different amounts of data. We find that the learned primal-dual has an outstanding performance in terms of reconstruction quality and data efficiency. However, in general, end-to-end learned methods have two issues: a) lack of classical guarantees in inverse problems and b) lack of generalization when not trained with enough data. To overcome these issues, we bring in the deep image prior approach in combination with classical regularization. The proposed methods improve the state-of-the-art results in the low data-regime.
Recommendations
- Deep Learning--Based Dictionary Learning and Tomographic Image Reconstruction
- A dataset-free deep learning method for low-dose CT image reconstruction
- Image decomposition algorithm for dual-energy computed tomography via fully convolutional network
- Tomographic image reconstruction using training images
- Biomedical image reconstruction: from the foundations to deep neural networks
- A tensor-based dictionary learning approach to tomographic image reconstruction
- Simultaneous tomographic reconstruction and segmentation with class priors
Cites work
- A convergence rates result for Tikhonov regularization in Banach spaces with non-smooth operators
- Deep Convolutional Neural Network for Inverse Problems in Imaging
- Deep null space learning for inverse problems: convergence analysis and rates
- scientific article; zbMATH DE number 4056355 (Why is no real title available?)
- scientific article; zbMATH DE number 41289 (Why is no real title available?)
- scientific article; zbMATH DE number 2057413 (Why is no real title available?)
- scientific article; zbMATH DE number 936298 (Why is no real title available?)
- Ideal spatial adaptation by wavelet shrinkage
- Learning the invisible: a hybrid deep learning-shearlet framework for limited angle computed tomography
- NETT: solving inverse problems with deep neural networks
- Regularization by architecture: a deep prior approach for inverse problems
- Solving ill-posed inverse problems using iterative deep neural networks
- Solving inverse problems using data-driven models
- The mathematics of computerized tomography
Cited in
(37)- CT image reconstruction on a low dimensional manifold
- Neural network for data preprocessing in computed tomography
- Learning to scan: a deep reinforcement learning approach for personalized scanning in CT imaging
- Constrained and unconstrained deep image prior optimization models with automatic regularization
- A review on deep learning in medical image reconstruction
- Unpaired low-dose CT denoising network based on cycle-consistent generative adversarial network with prior image information
- Deep learning for photoacoustic tomography from sparse data
- Regularization theory of the analytic deep prior approach
- Deep Learning--Based Dictionary Learning and Tomographic Image Reconstruction
- A Generative Variational Model for Inverse Problems in Imaging
- Wasserstein-based projections with applications to inverse problems
- Probabilistic approach to limited-data computed tomography reconstruction
- Biomedical image reconstruction: from the foundations to deep neural networks
- Learning the invisible: a hybrid deep learning-shearlet framework for limited angle computed tomography
- A dataset-free deep learning method for low-dose CT image reconstruction
- Unsupervised knowledge-transfer for learned image reconstruction
- Foreword to special issue of Inverse Problems on modern challenges in imaging
- Imaging based on Compton scattering: model uncertainty and data-driven reconstruction methods
- In focus -- hybrid deep learning approaches to the HDC2021 challenge
- Restoring severely out-of-focus blurred text images with deep image prior
- WPPNets and WPPFlows: The Power of Wasserstein Patch Priors for Superresolution
- Deep unrolling networks with recurrent momentum acceleration for nonlinear inverse problems
- Bayesian imaging inverse problem with SA-roundtrip prior via HMC-pCN sampler
- Convergent data-driven regularizations for CT reconstruction
- Low-resolution prior equilibrium network for CT reconstruction
- Equivariant neural networks for indirect measurements
- Deep learning methods for limited data problems in X-ray tomography
- Provably convergent learned inexact descent algorithm for low-dose CT reconstruction
- Nonlinear ill-posed problem in low-dose dental cone-beam computed tomography
- Normalizing flow regularization for photoacoustic tomography
- Parameter identification by deep learning of a material model for granular media
- Superiorized iteration algorithm for CT image simultaneous reconstruction and segmentation
- Learning from small data sets: patch-based regularizers in inverse problems for image reconstruction
- ECALL: estimation-calibrated learning for joint deconvolution and kernel inference from unpaired data
- Fast inexact bilevel optimization for analytical deep image priors
- A guide to stochastic optimisation for large-scale inverse problems
- Early stopping of untrained convolutional neural networks
Describes a project that uses
Uses Software
This page was built for publication: Computed tomography reconstruction using deep image prior and learned reconstruction methods
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5123693)
