Image Reconstruction by Splitting Deep Learning Regularization from Iterative Inversion

@inproceedings{Liu2018ImageRB,
  title={Image Reconstruction by Splitting Deep Learning Regularization from Iterative Inversion},
  author={Jiulong Liu and Tao Kuang and Xiaoqun Zhang},
  booktitle={MICCAI},
  year={2018}
}
Image reconstruction from downsampled and corrupted measurements, such as fast MRI and low dose CT, is mathematically ill-posed inverse problem. In this work, we propose a general and easy-to-use reconstruction method based on deep learning techniques. In order to address the intractable inversion of general inverse problems, we propose to train a network to refine intermediate images from classical reconstruction procedure to the ground truth, i.e. the intermediate images that satisfy the data… 
Learned Full-Sampling Reconstruction From Incomplete Data
TLDR
Inspired by the success of deep learning methods in imaging, the convolutional neural networks are utilized to embed and learn both the interrelationship between raw data and reconstructed images and prior information such as regularization, which is implemented in an end-to-end training process.
Self-supervised Bayesian Deep Learning for Image Recovery with Applications to Compressive Sensing
TLDR
The experiments showed that the proposed dataset-free deep learning method not only significantly outperforms traditional non-learning methods, but also is very competitive to the state-of-the-art supervised deep learning methods, especially when the measurements are few and noisy.
Structurally Adaptive Multi-Derivative Regularization for Image Recovery from Sparse Fourier Samples
The importance of regularization has been well established in image reconstruction—which is the computational inversion of imaging forward model–with applications including deconvolution for
Learnable Douglas-Rachford iteration and its applications in DOT imaging
TLDR
A deep neural network (DNN) based image reconstruction method, the so-called DR-Net, that leverages the interpretability of existing regularization methods and adaptive modeling capacity of DNN to solve image reconstruction problem in diffusion optical tomography (DOT).
Learned Full-Sampling Reconstruction
TLDR
A novel iterative reconstruction model to fit the target images and the corresponding high resolution measurements in Radon domain by the full-sampling system matrices is proposed and outperforms some latest learning based reconstruction methods for the sparse-view CT problems.
k-t NEXT: Dynamic MR Image Reconstruction Exploiting Spatio-temporal Correlations
TLDR
A novel deep learning based approach for dynamic MR image reconstruction, termed k-t NEXT (k-t NEtwork with X-f Transform), inspired by traditional methods to reconstruct the true signals from aliased signals in x-f domain to exploit the spatio-temporal redundancies.
Can learning from natural image denoising be used for seismic data interpolation?
TLDR
The primary results of synthetic and field data show promising interpolation performances of the adopted CNN-POCS method in terms of the signal-to-noise ratio, dealiasing, and weak-feature reconstruction, in comparison with the traditional [Formula] prediction filtering, curvelet transform, and block-matching 3D filtering methods.
Image Restoration by Combined Order Regularization With Optimal Spatial Adaptation
TLDR
This work proposes a novel non-convex regularization functional which adaptively combines Hessian-Schatten (HS) norm and first order TV (TV1) functionals with spatially varying weight and compares the proposed method with existing regularization methods, and a recently proposed Deep GAN method.
Tomographic Inverse Problems: Theory and Applications
TLDR
This workshop encompassed classical areas such as X-ray computed tomography (CT) as well as new modalities and applications such as dynamic imaging, Compton scattering tomography, hybrid imaging, optical tomography or multi-energy CT and addressed inter alia the use of methods from machine learning.
...
...

References

SHOWING 1-10 OF 18 REFERENCES
Solving ill-posed inverse problems using iterative deep neural networks
TLDR
The method builds on ideas from classical regularization theory and recent advances in deep learning to perform learning while making use of prior information about the inverse problem encoded in the forward operator, noise model and a regularizing functional to results in a gradient-like iterative scheme.
Learned Primal-Dual Reconstruction
We propose the Learned Primal-Dual algorithm for tomographic reconstruction. The algorithm accounts for a (possibly non-linear) forward operator in a deep neural network by unrolling a proximal
A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstruction
TLDR
A framework for reconstructing dynamic sequences of 2-D cardiac magnetic resonance images from undersampled data using a deep cascade of convolutional neural networks (CNNs) to accelerate the data acquisition process is proposed and it is demonstrated that CNNs can learn spatio-temporal correlations efficiently by combining convolution and data sharing approaches.
Deep Generative Adversarial Networks for Compressed Sensing Automates MRI
TLDR
A novel CS framework that permeates benefits from generative adversarial networks (GAN) to train a (low-dimensional) manifold of diagnostic-quality MR images from historical patients, which offers reconstruction under a few milliseconds, two orders of magnitude faster than state-of-the-art CS-MRI schemes.
Deep ADMM-Net for Compressive Sensing MRI
TLDR
Experiments on MRI image reconstruction under different sampling ratios in k-space demonstrate that the proposed novel ADMM-Net algorithm significantly improves the baseline ADMM algorithm and achieves high reconstruction accuracies with fast computational speed.
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
TLDR
SRGAN, a generative adversarial network (GAN) for image super-resolution (SR), is presented, to its knowledge, the first framework capable of inferring photo-realistic natural images for 4x upscaling factors and a perceptual loss function which consists of an adversarial loss and a content loss.
Image Denoising and Inpainting with Deep Neural Networks
TLDR
A novel approach to low-level vision problems that combines sparse coding and deep networks pre-trained with denoising auto-encoder (DA) is presented and can automatically remove complex patterns like superimposed text from an image, rather than simple patterns like pixels missing at random.
InverseNet: Solving Inverse Problems with Splitting Networks
TLDR
A new method that uses deep learning techniques to solve the inverse problems, referred to as InverseNet, is flexible in the sense that most of the existing end-to-end network structure can be leveraged in the first network and the existing denoising network structureCan be used in the second one.
Non-Local Auto-Encoder With Collaborative Stabilization for Image Restoration
TLDR
The (stacked) non-local auto-encoder is developed, which exploits self-similar information in natural images for stability by constraining the difference between the hidden representations of non- local similar image blocks during training.
Sparse MRI: The application of compressed sensing for rapid MR imaging
TLDR
Practical incoherent undersampling schemes are developed and analyzed by means of their aliasing interference and demonstrate improved spatial resolution and accelerated acquisition for multislice fast spin‐echo brain imaging and 3D contrast enhanced angiography.
...
...