REDAEP
Paper: REDAEP: Robust and Enhanced Denoising Autoencoding Prior for Sparse-View CT Reconstruction
Authors: Fengqin Zhang, Minghui Zhang, Binjie Qin, Yi Zhang, Zichen Xu, Qiegen Liu, and Dong Liang
IEEE Transactions on Radiation and Plasma Medical Sciences, https://ieeexplore.ieee.org/document/9076295
Motivation
In X-ray computed tomography, radiation doses are harmful but can be significantly reduced by intuitively decreasing the number of projections. However, less projection views usually lead to low resolution images. To address this issue, we propose a robust and enhanced mechanism on the basis of denoising autoencoding prior, or REDAEP for sparse-view CT reconstruction. REDAEP can substantially improve the reconstruction quality with two novel contri-butions. First, by employing the variable augmentation technique, REDAEP learns higher-dimensional network with three-channel image and proceeds to the single-channel image reconstruction. Second, REDAEP replaces the L2 regression loss function with a more robust Lp (0<p<2) regression, in order to preserve more texture details. Empirical results demonstrate that REDAEP can achieve better performance, i.e., quantitative measures and subjective visual quality, as compared to state-of-the-arts.
Algorithm
The overall REDAEP algorithm is as follows:
The flowchart illustration of REDEAP for single-channel CT reconstruction
Requirements and Dependencies
SysMat_ImgSize_256_Views_64_Bins_512.mat (https://pan.baidu.com/s/1Mjybi8Me6lRwAE0s9Gu1qw Password:auj6)
Other Related Projects
-
Iterative Reconstruction for Low-Dose CT using Deep Gradient Priors of Generative Model
[Paper] [Code] [PPT] -
Highly Undersampled Magnetic Resonance Imaging Reconstruction using Autoencoding Priors
[Paper] [Code] [Slide] -
Multi-Channel and Multi-Model-Based Autoencoding Prior for Grayscale Image Restoration
[Paper] [Code] [Slide] -
High-dimensional Embedding Network Derived Prior for Compressive Sensing MRI Reconstruction
[Paper] [Code] [PPT] -
Homotopic Gradients of Generative Density Priors for MR Image Reconstruction
[Paper] [Code] [PPT] -
Denoising Auto-encoding Priors in Undecimated Wavelet Domain for MR Image Reconstruction
[Paper] [Paper] [Code] -
Deep Frequency-Recurrent Priors for Inverse Imaging Reconstruction
[Paper] [Code] -
Learning Multi-Denoising Autoencoding Priors for Image Super-Resolution
[Paper] [Code]