Correcting brain-wide correlation differences in resting-state FMRI

Ziad S. Saad, Richard C. Reynolds, Hang Joon Jo, Stephen J. Gotts, Gang Chen, Alex Martin, Robert W. Cox

Research output: Contribution to journalArticle

97 Citations (Scopus)

Abstract

Brain function in "resting" state has been extensively studied with functional magnetic resonance imaging (FMRI). However, drawing valid inferences, particularly for group comparisons, is fraught with pitfalls. Differing levels of brain-wide correlations can confound group comparisons. Global signal regression (GSReg) attempts to reduce this confound and is commonly used, even though it differentially biases correlations over brain regions, potentially leading to false group differences. We propose to use average brain-wide correlations as a measure of global correlation (GCOR), and examine the circumstances under which it can be used to identify or correct for differences in global fluctuations. In the process, we show the bias induced by GSReg to be a function only of the data's covariance matrix, and use simulations to compare corrections with GCOR as covariate to GSReg under various scenarios. We find that unlike GSReg, GCOR is a conservative approach that can reduce global variations, while avoiding the introduction of false significant differences, as GSReg can. However, as with GSReg, one cannot escape the interaction effect between the grouping variable and GCOR covariate on effect size. While GCOR is a complementary measure for resting state-FMRI applicable to legacy data, it is a lesser substitute for proper level-I denoising. We also assess the applicability of GCOR to empirical data with motion-based subject grouping and compare group differences to those using GSReg. We find that, while GCOR reduced correlation differences between high and low movers, it is doubtful that motion was the sole driver behind the differences in the first place.

Original languageEnglish
Pages (from-to)339-352
Number of pages14
JournalUnknown Journal
Volume3
Issue number4
DOIs
StatePublished - 2013 Aug 1

Fingerprint

Magnetic Resonance Imaging
Brain

Cite this

Saad, Z. S., Reynolds, R. C., Jo, H. J., Gotts, S. J., Chen, G., Martin, A., & Cox, R. W. (2013). Correcting brain-wide correlation differences in resting-state FMRI. Unknown Journal, 3(4), 339-352. https://doi.org/10.1089/brain.2013.0156
Saad, Ziad S. ; Reynolds, Richard C. ; Jo, Hang Joon ; Gotts, Stephen J. ; Chen, Gang ; Martin, Alex ; Cox, Robert W. / Correcting brain-wide correlation differences in resting-state FMRI. In: Unknown Journal. 2013 ; Vol. 3, No. 4. pp. 339-352.
@article{0dbda7c9168043c08cea73ca34906f1a,
title = "Correcting brain-wide correlation differences in resting-state FMRI",
abstract = "Brain function in {"}resting{"} state has been extensively studied with functional magnetic resonance imaging (FMRI). However, drawing valid inferences, particularly for group comparisons, is fraught with pitfalls. Differing levels of brain-wide correlations can confound group comparisons. Global signal regression (GSReg) attempts to reduce this confound and is commonly used, even though it differentially biases correlations over brain regions, potentially leading to false group differences. We propose to use average brain-wide correlations as a measure of global correlation (GCOR), and examine the circumstances under which it can be used to identify or correct for differences in global fluctuations. In the process, we show the bias induced by GSReg to be a function only of the data's covariance matrix, and use simulations to compare corrections with GCOR as covariate to GSReg under various scenarios. We find that unlike GSReg, GCOR is a conservative approach that can reduce global variations, while avoiding the introduction of false significant differences, as GSReg can. However, as with GSReg, one cannot escape the interaction effect between the grouping variable and GCOR covariate on effect size. While GCOR is a complementary measure for resting state-FMRI applicable to legacy data, it is a lesser substitute for proper level-I denoising. We also assess the applicability of GCOR to empirical data with motion-based subject grouping and compare group differences to those using GSReg. We find that, while GCOR reduced correlation differences between high and low movers, it is doubtful that motion was the sole driver behind the differences in the first place.",
author = "Saad, {Ziad S.} and Reynolds, {Richard C.} and Jo, {Hang Joon} and Gotts, {Stephen J.} and Gang Chen and Alex Martin and Cox, {Robert W.}",
year = "2013",
month = "8",
day = "1",
doi = "10.1089/brain.2013.0156",
language = "English",
volume = "3",
pages = "339--352",
journal = "Unknown Journal",
number = "4",

}

Saad, ZS, Reynolds, RC, Jo, HJ, Gotts, SJ, Chen, G, Martin, A & Cox, RW 2013, 'Correcting brain-wide correlation differences in resting-state FMRI', Unknown Journal, vol. 3, no. 4, pp. 339-352. https://doi.org/10.1089/brain.2013.0156

Correcting brain-wide correlation differences in resting-state FMRI. / Saad, Ziad S.; Reynolds, Richard C.; Jo, Hang Joon; Gotts, Stephen J.; Chen, Gang; Martin, Alex; Cox, Robert W.

In: Unknown Journal, Vol. 3, No. 4, 01.08.2013, p. 339-352.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Correcting brain-wide correlation differences in resting-state FMRI

AU - Saad, Ziad S.

AU - Reynolds, Richard C.

AU - Jo, Hang Joon

AU - Gotts, Stephen J.

AU - Chen, Gang

AU - Martin, Alex

AU - Cox, Robert W.

PY - 2013/8/1

Y1 - 2013/8/1

N2 - Brain function in "resting" state has been extensively studied with functional magnetic resonance imaging (FMRI). However, drawing valid inferences, particularly for group comparisons, is fraught with pitfalls. Differing levels of brain-wide correlations can confound group comparisons. Global signal regression (GSReg) attempts to reduce this confound and is commonly used, even though it differentially biases correlations over brain regions, potentially leading to false group differences. We propose to use average brain-wide correlations as a measure of global correlation (GCOR), and examine the circumstances under which it can be used to identify or correct for differences in global fluctuations. In the process, we show the bias induced by GSReg to be a function only of the data's covariance matrix, and use simulations to compare corrections with GCOR as covariate to GSReg under various scenarios. We find that unlike GSReg, GCOR is a conservative approach that can reduce global variations, while avoiding the introduction of false significant differences, as GSReg can. However, as with GSReg, one cannot escape the interaction effect between the grouping variable and GCOR covariate on effect size. While GCOR is a complementary measure for resting state-FMRI applicable to legacy data, it is a lesser substitute for proper level-I denoising. We also assess the applicability of GCOR to empirical data with motion-based subject grouping and compare group differences to those using GSReg. We find that, while GCOR reduced correlation differences between high and low movers, it is doubtful that motion was the sole driver behind the differences in the first place.

AB - Brain function in "resting" state has been extensively studied with functional magnetic resonance imaging (FMRI). However, drawing valid inferences, particularly for group comparisons, is fraught with pitfalls. Differing levels of brain-wide correlations can confound group comparisons. Global signal regression (GSReg) attempts to reduce this confound and is commonly used, even though it differentially biases correlations over brain regions, potentially leading to false group differences. We propose to use average brain-wide correlations as a measure of global correlation (GCOR), and examine the circumstances under which it can be used to identify or correct for differences in global fluctuations. In the process, we show the bias induced by GSReg to be a function only of the data's covariance matrix, and use simulations to compare corrections with GCOR as covariate to GSReg under various scenarios. We find that unlike GSReg, GCOR is a conservative approach that can reduce global variations, while avoiding the introduction of false significant differences, as GSReg can. However, as with GSReg, one cannot escape the interaction effect between the grouping variable and GCOR covariate on effect size. While GCOR is a complementary measure for resting state-FMRI applicable to legacy data, it is a lesser substitute for proper level-I denoising. We also assess the applicability of GCOR to empirical data with motion-based subject grouping and compare group differences to those using GSReg. We find that, while GCOR reduced correlation differences between high and low movers, it is doubtful that motion was the sole driver behind the differences in the first place.

UR - http://www.scopus.com/inward/record.url?scp=84884679700&partnerID=8YFLogxK

U2 - 10.1089/brain.2013.0156

DO - 10.1089/brain.2013.0156

M3 - Article

C2 - 23705677

AN - SCOPUS:84884679700

VL - 3

SP - 339

EP - 352

JO - Unknown Journal

JF - Unknown Journal

IS - 4

ER -