Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes

Diffusion MRI offers great potential in studying the human brain microstructure and connectivity. However, diffusion images are marred by technical problems, such as image distortions and spurious signal loss. Correcting for these problems is non-trivial and relies on having a mechanism that predict...

Full description

Bibliographic Details
Main Authors:	Andersson, Jesper L.R., Sotiropoulos, Stamatios N.
Format:	Article
Language:	English
Published:	Elsevier 2015
Online Access:	http://eprints.nottingham.ac.uk/50942/ http://eprints.nottingham.ac.uk/50942/ http://eprints.nottingham.ac.uk/50942/ http://eprints.nottingham.ac.uk/50942/1/1-s2.0-S1053811915006874-main.pdf

id	nottingham-50942
recordtype	eprints
spelling	nottingham-509422018-04-05T13:37:24Z http://eprints.nottingham.ac.uk/50942/ Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes Andersson, Jesper L.R. Sotiropoulos, Stamatios N. Diffusion MRI offers great potential in studying the human brain microstructure and connectivity. However, diffusion images are marred by technical problems, such as image distortions and spurious signal loss. Correcting for these problems is non-trivial and relies on having a mechanism that predicts what to expect. In this paper we describe a novel way to represent and make predictions about diffusion MRI data. It is based on a Gaussian process on one or several spheres similar to the Geostatistical method of “Kriging”. We present a choice of covariance function that allows us to accurately predict the signal even from voxels with complex fibre patterns. For multi-shell data (multiple non-zero b-values) the covariance function extends across the shells which means that data from one shell is used when making predictions for another shell. Elsevier 2015-11-15 Article PeerReviewed application/pdf en cc_by http://eprints.nottingham.ac.uk/50942/1/1-s2.0-S1053811915006874-main.pdf Andersson, Jesper L.R. and Sotiropoulos, Stamatios N. (2015) Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes. NeuroImage, 122 . pp. 166-176. ISSN 1053-8119 https://www.sciencedirect.com/science/article/pii/S1053811915006874 doi:10.1016/j.neuroimage.2015.07.067 doi:10.1016/j.neuroimage.2015.07.067
repository_type	Digital Repository
institution_category	Local University
institution	University of Nottingham Malaysia Campus
building	Nottingham Research Data Repository
collection	Online Access
language	English
description	Diffusion MRI offers great potential in studying the human brain microstructure and connectivity. However, diffusion images are marred by technical problems, such as image distortions and spurious signal loss. Correcting for these problems is non-trivial and relies on having a mechanism that predicts what to expect. In this paper we describe a novel way to represent and make predictions about diffusion MRI data. It is based on a Gaussian process on one or several spheres similar to the Geostatistical method of “Kriging”. We present a choice of covariance function that allows us to accurately predict the signal even from voxels with complex fibre patterns. For multi-shell data (multiple non-zero b-values) the covariance function extends across the shells which means that data from one shell is used when making predictions for another shell.
format	Article
author	Andersson, Jesper L.R. Sotiropoulos, Stamatios N.
spellingShingle	Andersson, Jesper L.R. Sotiropoulos, Stamatios N. Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes
author_facet	Andersson, Jesper L.R. Sotiropoulos, Stamatios N.
author_sort	Andersson, Jesper L.R.
title	Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes
title_short	Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes
title_full	Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes
title_fullStr	Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes
title_full_unstemmed	Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes
title_sort	non-parametric representation and prediction of single- and multi-shell diffusion-weighted mri data using gaussian processes
publisher	Elsevier
publishDate	2015
url	http://eprints.nottingham.ac.uk/50942/ http://eprints.nottingham.ac.uk/50942/ http://eprints.nottingham.ac.uk/50942/ http://eprints.nottingham.ac.uk/50942/1/1-s2.0-S1053811915006874-main.pdf
first_indexed	2018-09-06T14:17:10Z
last_indexed	2018-09-06T14:17:10Z
_version_	1610867991492165632

Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes

Similar Items