Sparse bayesian inference of white matter fiber orientations from compressed multi-resolution diffusion MRI

Pramod K Pisharady, Julio M. Duarte-Carvajalino, Stamatios N. Sotiropoulos, Guillermo Sapiro, Christophe Lenglet

Research output: Chapter in Book/Report/Conference proceedingChapter

3 Citations (Scopus)

Abstract

The RubiX [1] algorithm combines high SNR characteristics of low resolution data with high spacial specificity of high resolution data, to extract microstructural tissue parameters from diffusion MRI. In this paper we focus on estimating crossing fiber orientations and introduce sparsity to the RubiX algorithm, making it suitable for reconstruction from compressed (under-sampled) data. We propose a sparse Bayesian algorithm for estimation of fiber orientations and volume fractions from compressed diffusion MRI. The data at high resolution is modeled using a parametric spherical deconvolution approach and represented using a dictionary created with the exponential decay components along different possible directions. Volume fractions of fibers along these orientations define the dictionary weights. The data at low resolution is modeled using a spatial partial volume representation. The proposed dictionary representation and sparsity priors consider the dependence between fiber orientations and the spatial redundancy in data representation. Our method exploits the sparsity of fiber orientations, therefore facilitating inference from under-sampled data. Experimental results show improved accuracy and decreased uncertainty in fiber orientation estimates. For under-sampled data, the proposed method is also shown to produce more robust estimates of fiber orientations.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer- Verlag
Pages117-124
Number of pages8
DOIs
StatePublished - Jan 1 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9349
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Fingerprint

Fiber Orientation
Bayesian inference
Fiber reinforced materials
Multiresolution
Magnetic resonance imaging
Glossaries
Sparsity
Volume fraction
Volume Fraction
High Resolution
Deconvolution
Robust Estimate
Redundancy
Exponential Decay
Tissue
Specificity
Fibers
Fiber
Uncertainty
Partial

Keywords

  • Brain imaging
  • Compressive sensing
  • Diffusion MRI
  • Fiber orientation
  • Linear un-mixing
  • Sparse bayesian inference

Cite this

Pisharady, P. K., Duarte-Carvajalino, J. M., Sotiropoulos, S. N., Sapiro, G., & Lenglet, C. (2015). Sparse bayesian inference of white matter fiber orientations from compressed multi-resolution diffusion MRI. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 117-124). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9349). Springer- Verlag. https://doi.org/10.1007/978-3-319-24553-9_15

Sparse bayesian inference of white matter fiber orientations from compressed multi-resolution diffusion MRI. / Pisharady, Pramod K; Duarte-Carvajalino, Julio M.; Sotiropoulos, Stamatios N.; Sapiro, Guillermo; Lenglet, Christophe.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer- Verlag, 2015. p. 117-124 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9349).

Research output: Chapter in Book/Report/Conference proceedingChapter

Pisharady, PK, Duarte-Carvajalino, JM, Sotiropoulos, SN, Sapiro, G & Lenglet, C 2015, Sparse bayesian inference of white matter fiber orientations from compressed multi-resolution diffusion MRI. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9349, Springer- Verlag, pp. 117-124. https://doi.org/10.1007/978-3-319-24553-9_15
Pisharady PK, Duarte-Carvajalino JM, Sotiropoulos SN, Sapiro G, Lenglet C. Sparse bayesian inference of white matter fiber orientations from compressed multi-resolution diffusion MRI. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer- Verlag. 2015. p. 117-124. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-24553-9_15
Pisharady, Pramod K ; Duarte-Carvajalino, Julio M. ; Sotiropoulos, Stamatios N. ; Sapiro, Guillermo ; Lenglet, Christophe. / Sparse bayesian inference of white matter fiber orientations from compressed multi-resolution diffusion MRI. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer- Verlag, 2015. pp. 117-124 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inbook{28c6c0da12b1460780427b7dad443e93,
title = "Sparse bayesian inference of white matter fiber orientations from compressed multi-resolution diffusion MRI",
abstract = "The RubiX [1] algorithm combines high SNR characteristics of low resolution data with high spacial specificity of high resolution data, to extract microstructural tissue parameters from diffusion MRI. In this paper we focus on estimating crossing fiber orientations and introduce sparsity to the RubiX algorithm, making it suitable for reconstruction from compressed (under-sampled) data. We propose a sparse Bayesian algorithm for estimation of fiber orientations and volume fractions from compressed diffusion MRI. The data at high resolution is modeled using a parametric spherical deconvolution approach and represented using a dictionary created with the exponential decay components along different possible directions. Volume fractions of fibers along these orientations define the dictionary weights. The data at low resolution is modeled using a spatial partial volume representation. The proposed dictionary representation and sparsity priors consider the dependence between fiber orientations and the spatial redundancy in data representation. Our method exploits the sparsity of fiber orientations, therefore facilitating inference from under-sampled data. Experimental results show improved accuracy and decreased uncertainty in fiber orientation estimates. For under-sampled data, the proposed method is also shown to produce more robust estimates of fiber orientations.",
keywords = "Brain imaging, Compressive sensing, Diffusion MRI, Fiber orientation, Linear un-mixing, Sparse bayesian inference",
author = "Pisharady, {Pramod K} and Duarte-Carvajalino, {Julio M.} and Sotiropoulos, {Stamatios N.} and Guillermo Sapiro and Christophe Lenglet",
year = "2015",
month = "1",
day = "1",
doi = "10.1007/978-3-319-24553-9_15",
language = "English (US)",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer- Verlag",
pages = "117--124",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - CHAP

T1 - Sparse bayesian inference of white matter fiber orientations from compressed multi-resolution diffusion MRI

AU - Pisharady, Pramod K

AU - Duarte-Carvajalino, Julio M.

AU - Sotiropoulos, Stamatios N.

AU - Sapiro, Guillermo

AU - Lenglet, Christophe

PY - 2015/1/1

Y1 - 2015/1/1

N2 - The RubiX [1] algorithm combines high SNR characteristics of low resolution data with high spacial specificity of high resolution data, to extract microstructural tissue parameters from diffusion MRI. In this paper we focus on estimating crossing fiber orientations and introduce sparsity to the RubiX algorithm, making it suitable for reconstruction from compressed (under-sampled) data. We propose a sparse Bayesian algorithm for estimation of fiber orientations and volume fractions from compressed diffusion MRI. The data at high resolution is modeled using a parametric spherical deconvolution approach and represented using a dictionary created with the exponential decay components along different possible directions. Volume fractions of fibers along these orientations define the dictionary weights. The data at low resolution is modeled using a spatial partial volume representation. The proposed dictionary representation and sparsity priors consider the dependence between fiber orientations and the spatial redundancy in data representation. Our method exploits the sparsity of fiber orientations, therefore facilitating inference from under-sampled data. Experimental results show improved accuracy and decreased uncertainty in fiber orientation estimates. For under-sampled data, the proposed method is also shown to produce more robust estimates of fiber orientations.

AB - The RubiX [1] algorithm combines high SNR characteristics of low resolution data with high spacial specificity of high resolution data, to extract microstructural tissue parameters from diffusion MRI. In this paper we focus on estimating crossing fiber orientations and introduce sparsity to the RubiX algorithm, making it suitable for reconstruction from compressed (under-sampled) data. We propose a sparse Bayesian algorithm for estimation of fiber orientations and volume fractions from compressed diffusion MRI. The data at high resolution is modeled using a parametric spherical deconvolution approach and represented using a dictionary created with the exponential decay components along different possible directions. Volume fractions of fibers along these orientations define the dictionary weights. The data at low resolution is modeled using a spatial partial volume representation. The proposed dictionary representation and sparsity priors consider the dependence between fiber orientations and the spatial redundancy in data representation. Our method exploits the sparsity of fiber orientations, therefore facilitating inference from under-sampled data. Experimental results show improved accuracy and decreased uncertainty in fiber orientation estimates. For under-sampled data, the proposed method is also shown to produce more robust estimates of fiber orientations.

KW - Brain imaging

KW - Compressive sensing

KW - Diffusion MRI

KW - Fiber orientation

KW - Linear un-mixing

KW - Sparse bayesian inference

UR - http://www.scopus.com/inward/record.url?scp=84947436505&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947436505&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-24553-9_15

DO - 10.1007/978-3-319-24553-9_15

M3 - Chapter

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 117

EP - 124

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer- Verlag

ER -