RicianNet

RicianNet for MRI denoising

#reader should unzip the file named matconvnet-1.0-beta24.rar
#your computer should have a successful matcaffe environment

The Code is created based on the method described in the following papers:
[1] Progressively distribution-based Rician noise removal for magnetic resonance imaging, ISMRM 2018, Oral.
Authors: Q. Liu, S. Li, J. Lv, D. Liang
[2] MRI Denoising using Progressively Distribution-based Neural Network, Magnetic Resonance Imaging, 2020.
Authors: S. Li, J. Zhou, D. Liang, Q. Liu
https://doi.org/10.1016/j.mri.2020.04.006

Date : 09/2018
Version : 1.0

The code and the algorithm are for non-comercial use only.
Copyright 2018, Department of Electronic Information Engineering, Nanchang University.

The flowchart of RicianNet for MRI denoising

The Conv and ReLU layers are denoted as "C"and"R",respectively.The ResNet and ResNet are denoted as "Res" and "ResB",respectively.

Network architecture of RicianNet

Visual quality comparison on a T1-weighted Brain1 image corrupted with Rician noise level of 5%.

From left to right: ground truth image, Rician noisy image and images denoised by NLM, UNLM, BM3D-VST and RicianNet.
The corresponding residual images are listed at the second row.

Other Related Projects

• Variable augmented neural network for decolorization and multi-exposure fusion [Paper] [Code] [Slide]

• IFR-Net: Iterative Feature Refinement Network for Compressed Sensing MRI [Paper] [Code]

• Iterative scheme-inspired network for impulse noise removal [Paper] [Code]

• A Comparative Study of CNN-based Super-resolution Methods in MRI Reconstruction and Its Beyond [Paper] [Code]

• Complex-valued MRI data from SIAT--test31 [Data]

• Complex-valued MRI data from SIAT--SIAT_MRIdata200 [Data]