Estimation and classification of fMRI hemodynamic response patterns

Robert D. Gibbons, Nicole A. Lazar, Dulal K. Bhaumik, Stanley L. Sclove, Hua Yun Chen, Keith R. Thulborn, John A. Sweeney, Kwan Hur, Dave Patterson

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

In this paper, we propose an approach to modeling functional magnetic resonance imaging (fMRI) data that combines hierarchical polynomial models, Bayes estimation, and clustering. A cubic polynomial is used to fit the voxel time courses of event-related design experiments. The coefficients of the polynomials are estimated by Bayes estimation, in a two-level hierarchical model, which allows us to borrow strength from all voxels. The voxel-specific Bayes polynomial coefficients are then transformed to the times and magnitudes of the minimum and maximum points on the hemodynamic response curve, which are in turn used to classify the voxels as being activated or not. The procedure is demonstrated on real data from an event-related design experiment of visually guided saccades and shown to be an effective alternative to existing methods.

Original languageEnglish (US)
Pages (from-to)804-814
Number of pages11
JournalNeuroImage
Volume22
Issue number2
DOIs
StatePublished - Jun 2004

Fingerprint

Hemodynamics
Magnetic Resonance Imaging
Saccades
Statistical Models
Cluster Analysis

Keywords

  • Bayes estimation
  • Hemodynamic response
  • Voxels

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Gibbons, R. D., Lazar, N. A., Bhaumik, D. K., Sclove, S. L., Chen, H. Y., Thulborn, K. R., ... Patterson, D. (2004). Estimation and classification of fMRI hemodynamic response patterns. NeuroImage, 22(2), 804-814. https://doi.org/10.1016/j.neuroimage.2004.02.003

Estimation and classification of fMRI hemodynamic response patterns. / Gibbons, Robert D.; Lazar, Nicole A.; Bhaumik, Dulal K.; Sclove, Stanley L.; Chen, Hua Yun; Thulborn, Keith R.; Sweeney, John A.; Hur, Kwan; Patterson, Dave.

In: NeuroImage, Vol. 22, No. 2, 06.2004, p. 804-814.

Research output: Contribution to journalArticle

Gibbons, RD, Lazar, NA, Bhaumik, DK, Sclove, SL, Chen, HY, Thulborn, KR, Sweeney, JA, Hur, K & Patterson, D 2004, 'Estimation and classification of fMRI hemodynamic response patterns', NeuroImage, vol. 22, no. 2, pp. 804-814. https://doi.org/10.1016/j.neuroimage.2004.02.003
Gibbons RD, Lazar NA, Bhaumik DK, Sclove SL, Chen HY, Thulborn KR et al. Estimation and classification of fMRI hemodynamic response patterns. NeuroImage. 2004 Jun;22(2):804-814. https://doi.org/10.1016/j.neuroimage.2004.02.003
Gibbons, Robert D. ; Lazar, Nicole A. ; Bhaumik, Dulal K. ; Sclove, Stanley L. ; Chen, Hua Yun ; Thulborn, Keith R. ; Sweeney, John A. ; Hur, Kwan ; Patterson, Dave. / Estimation and classification of fMRI hemodynamic response patterns. In: NeuroImage. 2004 ; Vol. 22, No. 2. pp. 804-814.
@article{ac51dff051f6462e96cfd27e61427880,
title = "Estimation and classification of fMRI hemodynamic response patterns",
abstract = "In this paper, we propose an approach to modeling functional magnetic resonance imaging (fMRI) data that combines hierarchical polynomial models, Bayes estimation, and clustering. A cubic polynomial is used to fit the voxel time courses of event-related design experiments. The coefficients of the polynomials are estimated by Bayes estimation, in a two-level hierarchical model, which allows us to borrow strength from all voxels. The voxel-specific Bayes polynomial coefficients are then transformed to the times and magnitudes of the minimum and maximum points on the hemodynamic response curve, which are in turn used to classify the voxels as being activated or not. The procedure is demonstrated on real data from an event-related design experiment of visually guided saccades and shown to be an effective alternative to existing methods.",
keywords = "Bayes estimation, Hemodynamic response, Voxels",
author = "Gibbons, {Robert D.} and Lazar, {Nicole A.} and Bhaumik, {Dulal K.} and Sclove, {Stanley L.} and Chen, {Hua Yun} and Thulborn, {Keith R.} and Sweeney, {John A.} and Kwan Hur and Dave Patterson",
year = "2004",
month = "6",
doi = "10.1016/j.neuroimage.2004.02.003",
language = "English (US)",
volume = "22",
pages = "804--814",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "2",

}

TY - JOUR

T1 - Estimation and classification of fMRI hemodynamic response patterns

AU - Gibbons, Robert D.

AU - Lazar, Nicole A.

AU - Bhaumik, Dulal K.

AU - Sclove, Stanley L.

AU - Chen, Hua Yun

AU - Thulborn, Keith R.

AU - Sweeney, John A.

AU - Hur, Kwan

AU - Patterson, Dave

PY - 2004/6

Y1 - 2004/6

N2 - In this paper, we propose an approach to modeling functional magnetic resonance imaging (fMRI) data that combines hierarchical polynomial models, Bayes estimation, and clustering. A cubic polynomial is used to fit the voxel time courses of event-related design experiments. The coefficients of the polynomials are estimated by Bayes estimation, in a two-level hierarchical model, which allows us to borrow strength from all voxels. The voxel-specific Bayes polynomial coefficients are then transformed to the times and magnitudes of the minimum and maximum points on the hemodynamic response curve, which are in turn used to classify the voxels as being activated or not. The procedure is demonstrated on real data from an event-related design experiment of visually guided saccades and shown to be an effective alternative to existing methods.

AB - In this paper, we propose an approach to modeling functional magnetic resonance imaging (fMRI) data that combines hierarchical polynomial models, Bayes estimation, and clustering. A cubic polynomial is used to fit the voxel time courses of event-related design experiments. The coefficients of the polynomials are estimated by Bayes estimation, in a two-level hierarchical model, which allows us to borrow strength from all voxels. The voxel-specific Bayes polynomial coefficients are then transformed to the times and magnitudes of the minimum and maximum points on the hemodynamic response curve, which are in turn used to classify the voxels as being activated or not. The procedure is demonstrated on real data from an event-related design experiment of visually guided saccades and shown to be an effective alternative to existing methods.

KW - Bayes estimation

KW - Hemodynamic response

KW - Voxels

UR - http://www.scopus.com/inward/record.url?scp=2942585718&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2942585718&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2004.02.003

DO - 10.1016/j.neuroimage.2004.02.003

M3 - Article

C2 - 15193609

AN - SCOPUS:2942585718

VL - 22

SP - 804

EP - 814

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 2

ER -