Optimal selection of regularization parameter in total variation method for reducing noise in magnetic resonance images of the brain

2014 ◽  
Vol 4 (1) ◽  
pp. 80-92 ◽  
Author(s):  
Michael Osadebey ◽  
◽  
Nizar Bouguila ◽  
Douglas Arnold
Keyword(s):  
Magnetic Resonance ◽  
Total Variation ◽  
Regularization Parameter ◽  
Magnetic Resonance Images ◽  
Variation Method ◽  
Optimal Selection ◽  
The Brain ◽  
Selection Of

Discrete Total Variation-Based Non-Local Means Filter for Denoising Magnetic Resonance Images

2020 ◽  
Vol 13 (4) ◽  
pp. 14-31
Author(s):  
Nikita Joshi ◽  
Sarika Jain ◽  
Amit Agarwal
Keyword(s):  
Magnetic Resonance ◽  
Total Variation ◽  
Similarity Index ◽  
Noise Removal ◽  
Magnetic Resonance Images ◽  
Variation Method ◽  
Mr Images ◽  
Denoising Method ◽  
Local Means ◽  
Non Local

Magnetic resonance (MR) images suffer from noise introduced by various sources. Due to this noise, diagnosis remains inaccurate. Thus, removal of noise becomes a very important task when dealing with MR images. In this paper, a denoising method has been discussed that makes use of non-local means filter and discrete total variation method. The proposed approach has been compared with other noise removal techniques like non-local means filter, anisotropic diffusion, total variation, and discrete total variation method, and it proves to be effective in reducing noise. The performance of various denoising methods is compared on basis of metrics such as peak signal-to-noise ratio (PSNR), mean square error (MSE), universal image quality index (UQI), and structure similarity index (SSIM) values. This method has been tested for various noise levels, and it outperformed other existing noise removal techniques, without blurring the image.

Joint total variation‐based reconstruction of multiparametric magnetic resonance images for mapping tissue types

2021 ◽  
Author(s):  
Shraddha Pandey ◽  
A. David Snider ◽  
Wilfrido A. Moreno ◽  
Harshan Ravi ◽  
Ali Bilgin ◽  
...  
Keyword(s):  
Magnetic Resonance ◽  
Total Variation ◽  
Magnetic Resonance Images

Ensemble learning-based classification on local patches from magnetic resonance images to detect iron depositions in the brain

2021 ◽  
Vol 17 (4) ◽  
pp. 260
Author(s):  
Beshiba Wilson ◽  
Ruma Madhu Sreedharan ◽  
Julia Punitha Malar Dhas ◽  
Ram P. Krish
Keyword(s):  
Magnetic Resonance ◽  
Ensemble Learning ◽  
Magnetic Resonance Images ◽  
The Brain

A comparison of neural network and fuzzy clustering techniques in segmenting magnetic resonance images of the brain

1992 ◽  
Vol 3 (5) ◽  
pp. 672-682 ◽  
Author(s):  
L.O. Hall ◽  
A.M. Bensaid ◽  
L.P. Clarke ◽  
R.P. Velthuizen ◽  
M.S. Silbiger ◽  
...  
Keyword(s):  
Neural Network ◽  
Magnetic Resonance ◽  
Fuzzy Clustering ◽  
Magnetic Resonance Images ◽  
Clustering Techniques ◽  
The Brain

scholarly journals A Total Variation-Based Reconstruction Method for Dynamic MRI

2008 ◽  
Vol 9 (1) ◽  
pp. 69-80 ◽  
Author(s):  
Germana Landi ◽  
Elena Loli Piccolomini ◽  
Fabiana Zama
Keyword(s):  
Magnetic Resonance ◽  
Total Variation ◽  
Dynamic Mri ◽  
Magnetic Resonance Images ◽  
Dynamic Imaging ◽  
Test Problems ◽  
High Temporal Resolution ◽  
Reconstruction Method ◽  
Dynamic Magnetic Resonance ◽  
Tv Regularization

In recent years, total variation (TV) regularization has become a popular and powerful tool for image restoration and enhancement. In this work, we apply TV minimization to improve the quality of dynamic magnetic resonance images. Dynamic magnetic resonance imaging is an increasingly popular clinical technique used to monitor spatio-temporal changes in tissue structure. Fast data acquisition is necessary in order to capture the dynamic process. Most commonly, the requirement of high temporal resolution is fulfilled by sacrificing spatial resolution. Therefore, the numerical methods have to address the issue of images reconstruction from limited Fourier data. One of the most successful techniques for dynamic imaging applications is the reduced-encoded imaging by generalized-series reconstruction method of Liang and Lauterbur. However, even if this method utilizesa prioridata for optimal image reconstruction, the produced dynamic images are degraded by truncation artifacts, most notably Gibbs ringing, due to the spatial low resolution of the data. We use a TV regularization strategy in order to reduce these truncation artifacts in the dynamic images. The resulting TV minimization problem is solved by the fixed point iteration method of Vogel and Oman. The results of test problems with simulated and real data are presented to illustrate the effectiveness of the proposed approach in reducing the truncation artifacts of the reconstructed images.

Sign in / Sign up

Export Citation Format

Share Document