FMRI analysis through bayesian variable selection with a spatial prior

Jing Xia; Feng Liang; Yongmei Michelle Wang

doi:10.1109/ISBI.2009.5193147

FMRI analysis through bayesian variable selection with a spatial prior

Jing Xia, Feng Liang, Yongmei Michelle Wang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper presents a novel spatial Bayesian method for simultaneous activation detection and hemodynamic response function (HRF) estimation of functional magnetic resonance imaging (fMRI) data. A Bayesian variable selection approach is used to induce shrinkage and sparsity, with a spatial prior on latent variables representing activated hemodynamic response components. Then, the activation map is generated from the full spectrum of posterior inference constructed through a Markov chain Monte Carlo scheme, and HRFs at different voxels are estimated non-parametrically with information pooling from neighboring voxels. By integrating functional activation detection and HRFs estimation in a unified framework, our method is more robust to noise and less sensitive to model mis-specification.

Original language	English (US)
Title of host publication	Proceedings - 2009 IEEE International Symposium on Biomedical Imaging
Subtitle of host publication	From Nano to Macro, ISBI 2009
Pages	714-717
Number of pages	4
DOIs	https://doi.org/10.1109/ISBI.2009.5193147
State	Published - 2009
Event	2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009 - Boston, MA, United States Duration: Jun 28 2009 → Jul 1 2009

Publication series

Name	Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009

Other

Other	2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009
Country/Territory	United States
City	Boston, MA
Period	6/28/09 → 7/1/09

Keywords

Activation detection
Bayesian variable selection
Hemodynamic response function
Markov chain Monte Carlo (MCMC)
Spatial prior

ASJC Scopus subject areas

Biomedical Engineering
Radiology Nuclear Medicine and imaging

Online availability

10.1109/ISBI.2009.5193147

Library availability

Discover UIUC Full Text

Cite this

Xia, J., Liang, F., & Wang, Y. M. (2009). FMRI analysis through bayesian variable selection with a spatial prior. In Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009 (pp. 714-717). Article 5193147 (Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009). https://doi.org/10.1109/ISBI.2009.5193147

FMRI analysis through bayesian variable selection with a spatial prior. / Xia, Jing; Liang, Feng ; Wang, Yongmei Michelle.
Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009. 2009. p. 714-717 5193147 (Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Xia, J, Liang, F & Wang, YM 2009, FMRI analysis through bayesian variable selection with a spatial prior. in Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009., 5193147, Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009, pp. 714-717, 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009, Boston, MA, United States, 6/28/09. https://doi.org/10.1109/ISBI.2009.5193147

@inproceedings{dd191be9cd13477fb50572145b5f27f5,

title = "FMRI analysis through bayesian variable selection with a spatial prior",

abstract = "This paper presents a novel spatial Bayesian method for simultaneous activation detection and hemodynamic response function (HRF) estimation of functional magnetic resonance imaging (fMRI) data. A Bayesian variable selection approach is used to induce shrinkage and sparsity, with a spatial prior on latent variables representing activated hemodynamic response components. Then, the activation map is generated from the full spectrum of posterior inference constructed through a Markov chain Monte Carlo scheme, and HRFs at different voxels are estimated non-parametrically with information pooling from neighboring voxels. By integrating functional activation detection and HRFs estimation in a unified framework, our method is more robust to noise and less sensitive to model mis-specification.",

keywords = "Activation detection, Bayesian variable selection, Hemodynamic response function, Markov chain Monte Carlo (MCMC), Spatial prior",

author = "Jing Xia and Feng Liang and Wang, {Yongmei Michelle}",

year = "2009",

doi = "10.1109/ISBI.2009.5193147",

language = "English (US)",

isbn = "9781424439324",

series = "Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009",

pages = "714--717",

booktitle = "Proceedings - 2009 IEEE International Symposium on Biomedical Imaging",

note = "2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009 ; Conference date: 28-06-2009 Through 01-07-2009",

}

TY - GEN

T1 - FMRI analysis through bayesian variable selection with a spatial prior

AU - Xia, Jing

AU - Liang, Feng

AU - Wang, Yongmei Michelle

PY - 2009

Y1 - 2009

N2 - This paper presents a novel spatial Bayesian method for simultaneous activation detection and hemodynamic response function (HRF) estimation of functional magnetic resonance imaging (fMRI) data. A Bayesian variable selection approach is used to induce shrinkage and sparsity, with a spatial prior on latent variables representing activated hemodynamic response components. Then, the activation map is generated from the full spectrum of posterior inference constructed through a Markov chain Monte Carlo scheme, and HRFs at different voxels are estimated non-parametrically with information pooling from neighboring voxels. By integrating functional activation detection and HRFs estimation in a unified framework, our method is more robust to noise and less sensitive to model mis-specification.

AB - This paper presents a novel spatial Bayesian method for simultaneous activation detection and hemodynamic response function (HRF) estimation of functional magnetic resonance imaging (fMRI) data. A Bayesian variable selection approach is used to induce shrinkage and sparsity, with a spatial prior on latent variables representing activated hemodynamic response components. Then, the activation map is generated from the full spectrum of posterior inference constructed through a Markov chain Monte Carlo scheme, and HRFs at different voxels are estimated non-parametrically with information pooling from neighboring voxels. By integrating functional activation detection and HRFs estimation in a unified framework, our method is more robust to noise and less sensitive to model mis-specification.

KW - Activation detection

KW - Bayesian variable selection

KW - Hemodynamic response function

KW - Markov chain Monte Carlo (MCMC)

KW - Spatial prior

UR - http://www.scopus.com/inward/record.url?scp=70449363842&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449363842&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2009.5193147

DO - 10.1109/ISBI.2009.5193147

M3 - Conference contribution

AN - SCOPUS:70449363842

SN - 9781424439324

T3 - Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009

SP - 714

EP - 717

BT - Proceedings - 2009 IEEE International Symposium on Biomedical Imaging

T2 - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009

Y2 - 28 June 2009 through 1 July 2009

ER -

FMRI analysis through bayesian variable selection with a spatial prior

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this