Literature DB >> 35284287

Improved magnetic resonance myelin water imaging using multi-channel denoising convolutional neural networks (MCDnCNN).

Guojun Xu1,2, Yongquan He1, Qiurong Yu1, Hongjian He3,4, Zhiyong Zhao4, Mingxia Fan1, Jianqi Li1, Dongrong Xu2.   

Abstract

Background: Myelin water imaging (MWI) is powerful and important for studying and diagnosing neurological and psychiatric diseases. In particular, myelin water fraction (MWF) is derived from MWI data for quantifying myelination. However, MWF estimation is typically sensitive to noise. Improving the accuracy of MWF estimation based on WMI data acquired using a magnetic resonance (MR) multiple gradient recalled echo (mGRE) imaging sequence is desired.
Methods: The proposed method employs a recently introduced the multi-channel denoising convolutional neural networks (MCDnCNN). Five different MCDnCNN models, denoted as Delevel1, Delevel2, Delevel3, Delevel4 and DelevelMix corresponding to five noise levels (Level1, Level2, Level3, Level4 and LevelMix), were trained using the data of the first echo of the mGRE brain images acquired from 15 healthy human subjects. Using simulated noisy data that employed a hollow cylinder model, we first evaluated the improvement in estimating MWF based on data denoised by the five different MCDnCNNs, by comparing the MWF maps calculated from the denoised data with ground truth. Next, we again evaluated the improvement using real-world in vivo datasets of 11 human participants acquired using the mGRE sequence. The datasets were first denoised by five different MCDnCNNs (Delevel1, 2, 3, 4 and DelevelMix), and subsequently their MWF maps were calculated and compared with the MWF maps directly calculated from the raw mGRE images without being denoised.
Results: Experiments using the simulation data denoised by the appropriate MCDnCNN models showed that the standard deviation (SD) of the absolute error (AE) of the derived MWF results was significantly reduced (maximal reduction =15.5%, Level3 simulated noisy data, orientation angle =0, all the five MCDnCNN models). In the test using in vivo data, estimating MWF based on data particularly denoised by the appropriate MCDnCNN models was found to be the best, compared to otherwise not using the appropriate models. The results demonstrated that the appropriate MCDnCNN models may permit high-quality MWF mapping, i.e., substantial reduction of random variation in estimating MWF-maps while preserving accuracy and structural details. Conclusions: Appropriate MCDnCNN models as proposed may improve both the accuracy and precision in estimating MWF maps, thereby making it a more clinically feasible alternative. 2022 Quantitative Imaging in Medicine and Surgery. All rights reserved.

Entities:  

Keywords:  Myelin water fraction (MWF); magnetic resonance imaging (MRI); multi-channel denoising convolutional neural network (MCDnCNN) models; multiple gradient recalled echo (mGRE)

Year:  2022        PMID: 35284287      PMCID: PMC8899954          DOI: 10.21037/qims-21-404

Source DB:  PubMed          Journal:  Quant Imaging Med Surg        ISSN: 2223-4306


  47 in total

1.  Nonexponential T₂ decay in white matter.

Authors:  Peter van Gelderen; Jacco A de Zwart; Jongho Lee; Pascal Sati; Daniel S Reich; Jeff H Duyn
Journal:  Magn Reson Med       Date:  2011-05-31       Impact factor: 4.668

2.  Probing signal phase in direct visualization of short transverse relaxation time component (ViSTa).

Authors:  Daeun Kim; Hyo Min Lee; Se-Hong Oh; Jongho Lee
Journal:  Magn Reson Med       Date:  2014-08-22       Impact factor: 4.668

Review 3.  Diffusion kurtosis imaging: an emerging technique for evaluating the microstructural environment of the brain.

Authors:  Andrew J Steven; Jiachen Zhuo; Elias R Melhem
Journal:  AJR Am J Roentgenol       Date:  2014-01       Impact factor: 3.959

4.  Fiber orientation-dependent white matter contrast in gradient echo MRI.

Authors:  Samuel Wharton; Richard Bowtell
Journal:  Proc Natl Acad Sci U S A       Date:  2012-10-22       Impact factor: 11.205

5.  Field inhomogeneity correction for gradient echo myelin water fraction imaging.

Authors:  Eva Alonso-Ortiz; Ives R Levesque; Raphaël Paquin; G Bruce Pike
Journal:  Magn Reson Med       Date:  2016-07-15       Impact factor: 4.668

6.  In vivo visualization of myelin water in brain by magnetic resonance.

Authors:  A MacKay; K Whittall; J Adler; D Li; D Paty; D Graeb
Journal:  Magn Reson Med       Date:  1994-06       Impact factor: 4.668

Review 7.  MRI in multiple sclerosis: current status and future prospects.

Authors:  Rohit Bakshi; Alan J Thompson; Maria A Rocca; Daniel Pelletier; Vincent Dousset; Frederik Barkhof; Matilde Inglese; Charles R G Guttmann; Mark A Horsfield; Massimo Filippi
Journal:  Lancet Neurol       Date:  2008-07       Impact factor: 44.182

8.  Micro-compartment specific T2* relaxation in the brain.

Authors:  Pascal Sati; Peter van Gelderen; Afonso C Silva; Daniel S Reich; Hellmut Merkle; Jacco A de Zwart; Jeff H Duyn
Journal:  Neuroimage       Date:  2013-03-22       Impact factor: 6.556

9.  Biophysical mechanisms of myelin-induced water frequency shifts.

Authors:  Dmitriy A Yablonskiy; Alexander L Sukstanskii
Journal:  Magn Reson Med       Date:  2014-04-03       Impact factor: 4.668

10.  Improved Visualization of Cartilage Canals Using Quantitative Susceptibility Mapping.

Authors:  Mikko J Nissi; Ferenc Tóth; Luning Wang; Cathy S Carlson; Jutta M Ellermann
Journal:  PLoS One       Date:  2015-07-13       Impact factor: 3.240

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.