Linear mixed-effects modeling approach to FMRI group analysis

Gang Chen, Ziad S. Saad, Jennifer C Britton, Daniel S. Pine, Robert W. Cox

Research output: Contribution to journalArticle

131 Citations (Scopus)

Abstract

Conventional group analysis is usually performed with Student-type t-test, regression, or standard AN(C)OVA in which the variance-covariance matrix is presumed to have a simple structure. Some correction approaches are adopted when assumptions about the covariance structure is violated. However, as experiments are designed with different degrees of sophistication, these traditional methods can become cumbersome, or even be unable to handle the situation at hand. For example, most current FMRI software packages have difficulty analyzing the following scenarios at group level: (1) taking within-subject variability into account when there are effect estimates from multiple runs or sessions; (2) continuous explanatory variables (covariates) modeling in the presence of a within-subject (repeated measures) factor, multiple subject-grouping (between-subjects) factors, or the mixture of both; (3) subject-specific adjustments in covariate modeling; (4) group analysis with estimation of hemodynamic response (HDR) function by multiple basis functions; (5) various cases of missing data in longitudinal studies; and (6) group studies involving family members or twins.Here we present a linear mixed-effects modeling (LME) methodology that extends the conventional group analysis approach to analyze many complicated cases, including the six prototypes delineated above, whose analyses would be otherwise either difficult or unfeasible under traditional frameworks such as AN(C)OVA and general linear model (GLM). In addition, the strength of the LME framework lies in its flexibility to model and estimate the variance-covariance structures for both random effects and residuals. The intraclass correlation (ICC) values can be easily obtained with an LME model with crossed random effects, even at the presence of confounding fixed effects. The simulations of one prototypical scenario indicate that the LME modeling keeps a balance between the control for false positives and the sensitivity for activation detection. The importance of hypothesis formulation is also illustrated in the simulations. Comparisons with alternative group analysis approaches and the limitations of LME are discussed in details.

Original languageEnglish
Pages (from-to)176-190
Number of pages15
JournalNeuroImage
Volume73
DOIs
StatePublished - Jun 1 2013
Externally publishedYes

Fingerprint

Social Adjustment
Longitudinal Studies
Linear Models
Software
Hand
Hemodynamics
Students

Keywords

  • AFNI
  • AN(C)OVA
  • FMRI group analysis
  • GLM
  • ICC
  • LME
  • R

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Linear mixed-effects modeling approach to FMRI group analysis. / Chen, Gang; Saad, Ziad S.; Britton, Jennifer C; Pine, Daniel S.; Cox, Robert W.

In: NeuroImage, Vol. 73, 01.06.2013, p. 176-190.

Research output: Contribution to journalArticle

Chen, Gang ; Saad, Ziad S. ; Britton, Jennifer C ; Pine, Daniel S. ; Cox, Robert W. / Linear mixed-effects modeling approach to FMRI group analysis. In: NeuroImage. 2013 ; Vol. 73. pp. 176-190.
@article{5ad1586967ef4d829ab32a1c78ae921c,
title = "Linear mixed-effects modeling approach to FMRI group analysis",
abstract = "Conventional group analysis is usually performed with Student-type t-test, regression, or standard AN(C)OVA in which the variance-covariance matrix is presumed to have a simple structure. Some correction approaches are adopted when assumptions about the covariance structure is violated. However, as experiments are designed with different degrees of sophistication, these traditional methods can become cumbersome, or even be unable to handle the situation at hand. For example, most current FMRI software packages have difficulty analyzing the following scenarios at group level: (1) taking within-subject variability into account when there are effect estimates from multiple runs or sessions; (2) continuous explanatory variables (covariates) modeling in the presence of a within-subject (repeated measures) factor, multiple subject-grouping (between-subjects) factors, or the mixture of both; (3) subject-specific adjustments in covariate modeling; (4) group analysis with estimation of hemodynamic response (HDR) function by multiple basis functions; (5) various cases of missing data in longitudinal studies; and (6) group studies involving family members or twins.Here we present a linear mixed-effects modeling (LME) methodology that extends the conventional group analysis approach to analyze many complicated cases, including the six prototypes delineated above, whose analyses would be otherwise either difficult or unfeasible under traditional frameworks such as AN(C)OVA and general linear model (GLM). In addition, the strength of the LME framework lies in its flexibility to model and estimate the variance-covariance structures for both random effects and residuals. The intraclass correlation (ICC) values can be easily obtained with an LME model with crossed random effects, even at the presence of confounding fixed effects. The simulations of one prototypical scenario indicate that the LME modeling keeps a balance between the control for false positives and the sensitivity for activation detection. The importance of hypothesis formulation is also illustrated in the simulations. Comparisons with alternative group analysis approaches and the limitations of LME are discussed in details.",
keywords = "AFNI, AN(C)OVA, FMRI group analysis, GLM, ICC, LME, R",
author = "Gang Chen and Saad, {Ziad S.} and Britton, {Jennifer C} and Pine, {Daniel S.} and Cox, {Robert W.}",
year = "2013",
month = "6",
day = "1",
doi = "10.1016/j.neuroimage.2013.01.047",
language = "English",
volume = "73",
pages = "176--190",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Linear mixed-effects modeling approach to FMRI group analysis

AU - Chen, Gang

AU - Saad, Ziad S.

AU - Britton, Jennifer C

AU - Pine, Daniel S.

AU - Cox, Robert W.

PY - 2013/6/1

Y1 - 2013/6/1

N2 - Conventional group analysis is usually performed with Student-type t-test, regression, or standard AN(C)OVA in which the variance-covariance matrix is presumed to have a simple structure. Some correction approaches are adopted when assumptions about the covariance structure is violated. However, as experiments are designed with different degrees of sophistication, these traditional methods can become cumbersome, or even be unable to handle the situation at hand. For example, most current FMRI software packages have difficulty analyzing the following scenarios at group level: (1) taking within-subject variability into account when there are effect estimates from multiple runs or sessions; (2) continuous explanatory variables (covariates) modeling in the presence of a within-subject (repeated measures) factor, multiple subject-grouping (between-subjects) factors, or the mixture of both; (3) subject-specific adjustments in covariate modeling; (4) group analysis with estimation of hemodynamic response (HDR) function by multiple basis functions; (5) various cases of missing data in longitudinal studies; and (6) group studies involving family members or twins.Here we present a linear mixed-effects modeling (LME) methodology that extends the conventional group analysis approach to analyze many complicated cases, including the six prototypes delineated above, whose analyses would be otherwise either difficult or unfeasible under traditional frameworks such as AN(C)OVA and general linear model (GLM). In addition, the strength of the LME framework lies in its flexibility to model and estimate the variance-covariance structures for both random effects and residuals. The intraclass correlation (ICC) values can be easily obtained with an LME model with crossed random effects, even at the presence of confounding fixed effects. The simulations of one prototypical scenario indicate that the LME modeling keeps a balance between the control for false positives and the sensitivity for activation detection. The importance of hypothesis formulation is also illustrated in the simulations. Comparisons with alternative group analysis approaches and the limitations of LME are discussed in details.

AB - Conventional group analysis is usually performed with Student-type t-test, regression, or standard AN(C)OVA in which the variance-covariance matrix is presumed to have a simple structure. Some correction approaches are adopted when assumptions about the covariance structure is violated. However, as experiments are designed with different degrees of sophistication, these traditional methods can become cumbersome, or even be unable to handle the situation at hand. For example, most current FMRI software packages have difficulty analyzing the following scenarios at group level: (1) taking within-subject variability into account when there are effect estimates from multiple runs or sessions; (2) continuous explanatory variables (covariates) modeling in the presence of a within-subject (repeated measures) factor, multiple subject-grouping (between-subjects) factors, or the mixture of both; (3) subject-specific adjustments in covariate modeling; (4) group analysis with estimation of hemodynamic response (HDR) function by multiple basis functions; (5) various cases of missing data in longitudinal studies; and (6) group studies involving family members or twins.Here we present a linear mixed-effects modeling (LME) methodology that extends the conventional group analysis approach to analyze many complicated cases, including the six prototypes delineated above, whose analyses would be otherwise either difficult or unfeasible under traditional frameworks such as AN(C)OVA and general linear model (GLM). In addition, the strength of the LME framework lies in its flexibility to model and estimate the variance-covariance structures for both random effects and residuals. The intraclass correlation (ICC) values can be easily obtained with an LME model with crossed random effects, even at the presence of confounding fixed effects. The simulations of one prototypical scenario indicate that the LME modeling keeps a balance between the control for false positives and the sensitivity for activation detection. The importance of hypothesis formulation is also illustrated in the simulations. Comparisons with alternative group analysis approaches and the limitations of LME are discussed in details.

KW - AFNI

KW - AN(C)OVA

KW - FMRI group analysis

KW - GLM

KW - ICC

KW - LME

KW - R

UR - http://www.scopus.com/inward/record.url?scp=84874769744&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874769744&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2013.01.047

DO - 10.1016/j.neuroimage.2013.01.047

M3 - Article

C2 - 23376789

AN - SCOPUS:84874769744

VL - 73

SP - 176

EP - 190

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

ER -