Independent vector analysis (IVA): Multivariate approach for fMRI group study

Lee, Jong Hwan; Lee, Te Won; Jólesz, Ferenc A.; Yoo, Seung Schik

doi:10.1016/j.neuroimage.2007.11.019

Cited by 138 publications

(130 citation statements)

References 55 publications

Supporting

Mentioning

130

Contrasting

Order By: Relevance

“…The inherent spatial homogeneity of the estimates from the GIFT can be traced to the PCAbased dimension reduction scheme across all the subjects' data whereby the estimation of time courses is likely to be biased towards the temporal characteristics from the commonly activated voxels across all the subjects. More detailed explanation on the inherent restriction of the GIFT is described elsewhere (Esposito et al, 2005;Beckmann and Smith, 2005;Lee et al, 2008).…”

Section: Discussionmentioning

confidence: 99%

“…It suggests that IVA may be susceptible to excessive reduction in ICs. Tensorial ICA, as an extension of ICA algorithm (Beckmann and Smith, 2005;Lee et al, 2008), may be used to complement the ICA-based analysis as well as the IVA scheme.…”

Section: Discussionmentioning

confidence: 99%

“…The Kullback-Leibler (KL) divergence between a joint probability density function (p.d.f.) and the factorized marginal p.d.f.s of estimated vector components was employed as a cost function of IVA (Kim et al, 2007;Lee et al, 2008) since KL divergence can measure the mutual information (MI) among vector components (Cichocki and Amari, 2002). The MI using the KL divergence can be defined as, By applying a gradient descent scheme to Eq.…”

Section: A Iva Model and Learning Algorithmmentioning

confidence: 99%

“…(5) to minimize the MI, an iterative learning rule of an unmixing matrix of subject m was obtained. The detailed derivation of learning algorithm and justification of the adopted nonlinear function can be found elsewhere (in the Appendix section, Lee et al 2008). where I is an N×N identity matrix, , and…”

Section: A Iva Model and Learning Algorithmmentioning

confidence: 99%

“…Due to an inherently imposed restriction along the (1) spatial, (2) temporal, and (3) subject domains, the group-inference based-on the ICA approach may not fully analyze the individual-specific activation patterns, especially those rising from small brain areas. The detailed description on the imposed restrictions of group ICA on (1) spatial, (2) temporal, and (3) subject domains was described elsewhere (p.87 & 90 in Lee et al 2008).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Independent vector analysis (IVA) for group fMRI processing of subcortical area

Lee

Lee²,

Jólesz³

et al. 2008

Int J Imaging Syst Tech

Self Cite

View full text Add to dashboard Cite

During functional MRI (fMRI) studies, blood oxygenation-level dependent (BOLD) signal associated with neuronal activity acquired from multiple individuals are subject to the derivation of group-averaged brain activation patterns. Unlike other cortical areas, subcortical areas such as the thalamus and basal ganglia often manifest smaller, biphasic BOLD signal that are aberrant from signals originating from cortices. Independent component analysis (ICA) can offer session/ individual specific brain activation maps without a priori assumptions regarding the timing or pattern of the signal responses. The small activation loci within the subcortical areas are sparsely distributed among the subjects, and a conventional group processing method based on the general linear model (GLM) or ICA may fail to characterize the activation loci. In this paper, we present an independent vector analysis (IVA) to overcome these limitations by offering an analysis of additional dependent components (compared to the ICA-based method) that are assigned for use in the automated grouping of dependent (i.e. similar) activation patterns across subjects. The proposed IVA algorithm was applied to simulated data, and its utility was confirmed from real fMRI data employing a trial-based hand motor task. A GLM and the group ICA of the fMRI toolbox (GIFT) were also applied for comparison. From the analysis of activation patterns within subcortical areas, in which the hemodynamic responses (HRs) often deviate from a canonical, model-driven HR, IVA detected task-related activation loci that were not detected through GLM and GIFT. IVA may offer a unique advantage for inferring group activation originating from subcortical areas.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: A Iva Model and Learning Algorithmmentioning

confidence: 99%

Section: A Iva Model and Learning Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Independent vector analysis (IVA) for group fMRI processing of subcortical area

Lee

Lee²,

Jólesz³

et al. 2008

Int J Imaging Syst Tech

Self Cite

View full text Add to dashboard Cite

show abstract

A method for accurate group difference detection by constraining the mixing coefficients in an ICA framework

Sui¹,

Adalı²,

Pearlson³

et al. 2009

Human Brain Mapping

View full text Add to dashboard Cite

Independent component analysis (ICA) is a promising method that is increasingly used to analyze brain imaging data such as functional magnetic resonance imaging (fMRI), structural MRI, and electroencephalography and has also proved useful for group comparison, e.g., differentiating healthy controls from patients. An advantage of ICA is its ability to identify components that are mixed in an unknown manner. However, ICA is not necessarily robust and optimal in identifying betweengroup effects, especially in highly noisy situations. Here, we propose a modified ICA framework for multi-group data analysis that incorporates prior information regarding group membership as a constraint into the mixing coefficients. Our approach, called coefficient-constrained ICA (CC-ICA), prioritizes identification of components that show a significant group difference. The performance of CC-ICA via synthetic and hybrid data simulations is evaluated under different hypothesis testing assumptions and signal to noise ratios (SNRs). Group analysis is also conducted on real multitask fMRI data. Results show that CC-ICA improves the estimation accuracy of the independent components greatly, especially those that have different patterns for different groups (e.g., patients vs. controls); In addition, it enhances the data extraction sensitivity to group differences by ranking components with P value or J-divergence more consistently with the ground truth. The proposed algorithm performs quite well for both group-difference detection and multitask fMRI data fusion, which may prove especially important for the identification of relevant disease biomarkers.

show abstract

Unbiased group‐level statistical assessment of independent component maps by means of automated retrospective matching

Langers

2009

Human Brain Mapping

View full text Add to dashboard Cite

This report presents and validates a method for the group-level statistical assessment of independent component analysis (ICA) outcomes. The method is based on a matching of individual component maps to corresponding aggregate maps that are obtained from concatenated data. Group-level statistics are derived that include an explicit correction for selection bias. Outcomes were validated by means of calculations with artificial null data. Although statistical inferences were found to be incorrect if bias was neglected, the use of the proposed bias correction sufficed to obtain valid results. This was further confirmed by extensive calculations with artificial data that contained known effects of interest. While uncorrected statistical assessments systematically violated the imposed confidence level thresholds, the corrected method was never observed to exceed the allowed false positive rate. Yet, bias correction was found to result in a reduced sensitivity and a moderate decrease in discriminatory power. The method was also applied to analyze actual fMRI data. Various effects of interest that were detectable in the aggregate data were similarly revealed by the retrospective matching method. In particular, stimulus-related responses were extensive. Nevertheless, differences were observed regarding their spatial distribution. The presented findings indicate that the proposed method is suitable for neuroimaging analyses. Finally, a number of generalizations are discussed. It is concluded that the proposed method provides a framework that may supplement many of the currently available group ICA methods with validated unbiased group inferences.

show abstract

Independent vector analysis (IVA): Multivariate approach for fMRI group study

Cited by 138 publications

References 55 publications

Independent vector analysis (IVA) for group fMRI processing of subcortical area

Independent vector analysis (IVA) for group fMRI processing of subcortical area

A method for accurate group difference detection by constraining the mixing coefficients in an ICA framework

Unbiased group‐level statistical assessment of independent component maps by means of automated retrospective matching

Contact Info

Product

Resources

About