Functional segmentation of the brain cortex using high model order group-PICA.

Kiviniemi, VJ; Starck, Tuomo; Remes, Jukka; Long, Xiao; Nikkinen, J; Haapea, Marianne; Veijola, Juha; Moilanen, Irma; Isohanni, Matti; Zang, Y-F; Tervonen, Osmo

doi:10.1016/s1053-8119(09)72194-9

Cited by 39 publications

(54 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The component ordering for the figure was based on the relative ranking of the percentage of variance explained by each component. The components observed were highly consistent with those previously reported (van de Ven et al, 2004; Beckmann et al, 2005; DeLuca et al, 2006; Damoiseaux et al, 2006; Smith et al, 2009; Kiviniemi et al, 2009). …”

Section: Resultssupporting

confidence: 92%

“…Each of the three sub-networks exhibited high test-retest reliability. This component-splitting phenomenon previously was thought to be a result of overestimating the model order, but recently was reported to reflect functional segregation or hierarchy within the “default mode” network (Buckner et al, 2008; Kiviniemi et al, 2009; Smith et al, 2009; Uddin et al, 2009; Harrison et al, 2008). Intriguingly, the three sub-networks we observed are well matched with the anatomical and functional theory of the “default mode” network proposed by Buckner et al (2008).…”

Section: Discussionmentioning

confidence: 98%

See 1 more Smart Citation

Reliable intrinsic connectivity networks: Test–retest evaluation using ICA and dual regression approach

et al. 2010

View full text Add to dashboard Cite

Functional connectivity analyses of resting-state fMRI data are rapidly emerging as highly efficient and powerful tools for in vivo mapping of functional networks in the brain, referred to as intrinsic connectivity networks (ICNs). Despite a burgeoning literature, researchers continue to struggle with the challenge of defining computationally efficient and reliable approaches for identifying and characterizing ICNs. Independent component analysis (ICA) has emerged as a powerful tool for exploring ICNs in both healthy and clinical populations. In particular, temporal concatenation group ICA (TC-GICA) coupled with a back-reconstruction step produces participant-level resting state functional connectivity (RSFC) maps for each group-level component. The present work systematically evaluated the test-retest reliability of TC-GICA derived RSFC measures over the short-term (< 45 minutes) and long-term (5 − 16 months). Additionally, to investigate the degree to which the components revealed by TC-GICA are detectable via single-session ICA, we investigated the reproducibility of TC-GICA findings. First, we found moderate-to-high short-and long-term testretest reliability for ICNs derived by combining TC-GICA and dual regression. Exceptions to this finding were limited to physiological-and imaging-related artifacts. Second, our reproducibility analyses revealed notable limitations for template matching procedures to accurately detect TC-GICA based components at the individual scan level. Third, we found that TC-GICA component's reliability and reproducibility ranks are highly consistent. In summary, TC-GICA combined with dual regression is an effective and reliable approach to exploratory analyses of resting state fMRI data.

show abstract

Section: Resultssupporting

confidence: 92%

Section: Discussionmentioning

confidence: 98%

Reliable intrinsic connectivity networks: Test–retest evaluation using ICA and dual regression approach

et al. 2010

View full text Add to dashboard Cite

show abstract

“…ICASSO with 100 re-runs and random initial conditions was used to arrive at a robust decomposition (Himberg et al, 2004), and high model order was used to provide a finer-grained regional separation of components in the cortical and subcortical compartments (Abou-Elseoud et al, 2010; Kiviniemi et al, 2009). The model order was tested according to the spatial map quality as well as the stability and empirically adjusted to 71 components (Abou-Elseoud et al, 2010).…”

Section: Methodsmentioning

confidence: 99%

Reactivity of hemodynamic responses and functional connectivity to different states of alpha synchrony: A concurrent EEG-fMRI study

2010

View full text Add to dashboard Cite

Concurrent EEG-fMRI studies have provided increasing details of the dynamics of intrinsic brain activity during the resting state. Here, we investigate a prominent effect in EEG during relaxed resting, i.e. the increase of the alpha power when the eyes are closed compared to when the eyes are open. This phenomenon is related to changes in thalamo-cortical and cortico-cortical synchronization. In order to investigate possible changes to EEG-fMRI coupling and fMRI functional connectivity during the two states we adopted a data-driven approach that fuses the multimodal data on the basis of parallel ICA decompositions of the fMRI data in the spatial domain and of the EEG data in the spectral domain. The power variation of a posterior alpha component was used as a reference function to deconvolve the hemodynamic responses from occipital, frontal, temporal, and subcortical fMRI components. Additionally, we computed the functional connectivity between these components. The results showed widespread alpha hemodynamic responses and high functional connectivity during eyes-closed (EC) rest, while eyes open (EO) resting abolished many of the hemodynamic responses and markedly decreased functional connectivity. These data suggest that generation of local hemodynamic responses is highly sensitive to state changes that do not involve changes of mental effort or awareness. They also indicate the localized power differences in posterior alpha between EO and EC in resting state data are accompanied by spatially widespread amplitude changes in hemodynamic responses and inter-regional functional connectivity, i.e. low frequency hemodynamic signals display an equivalent of alpha reactivity.

show abstract

“…However, these methods may not be ideal in the sense that observable anatomical boundaries do not necessarily correspond to functional units . For example, using ICA, it has been observed that in low dimensional parcellations, each component typically represents an extended "entire" brain network, whereas in high-dimensionality parcellations the obtained components are smaller and they are more likely to represent subparts of networks (Abou Elseoud et al, 2011;Kiviniemi et al, 2009). In these, nodes are defined from the analysis of blood-oxygen-level dependent (BOLD) activations obtained from a localizer or task-driven fMRI scan (Lashkari et al, 2012;Thirion, Varoquaux, Dohmatob, & Poline, 2014), or by independent component analysis (ICA) of resting fMRI data (Marrelec & Fransson, 2011;Smith et al, 2011).…”

Section: Introductionmentioning

confidence: 99%

Spatial parcellations, spectral filtering, and connectivity measures in fMRI: Optimizing for discrimination

Sala‐Llonch

Smith

Woolrich

et al. 2018

Human Brain Mapping

View full text Add to dashboard Cite

The analysis of Functional Connectivity (FC) is a key technique of fMRI, having been used to distinguish brain states and conditions. While many approaches to calculating FC are available, there have been few assessments of their differences, making it difficult to choose approaches, and compare results. Here, we assess the impact of methodological choices on discriminability, using a fully controlled data set of continuous active states involving basic visual and motor tasks, providing robust localized FC changes. We tested a range of anatomical and functional parcellations, including the AAL atlas, parcellations derived from the Human Connectome Project and Independent Component Analysis (ICA) of many dimensionalities. We measure amplitude, covariance, correlation, and regularized partial correlation under different temporal filtering choices. We evaluate features derived from these methods for discriminating states using MVPA. We find that multidimensional parcellations derived from functional data performed similarly, outperforming an anatomical atlas, with correlation and partial correlation (p < .05, FDR). Partial correlation, with appropriate regularization, outperformed correlation. Amplitude and covariance generally discriminated less well, although gave good results with high‐dimensionality ICA. We found that discriminative FC properties are frequency specific; higher frequencies performed surprisingly well under certain configurations of atlas choices and dependency measures, with ICA‐based parcellations revealing greater discriminability at high frequencies compared to other parcellations. Methodological choices in FC analyses can have a profound impact on results and can be selected to optimize accuracy, interpretability, and sharing of results. This work contributes to a basis for consistent selection of approaches to estimating and analyzing FC.

show abstract

Functional segmentation of the brain cortex using high model order group-PICA.

Cited by 39 publications

References 0 publications

Reliable intrinsic connectivity networks: Test–retest evaluation using ICA and dual regression approach

Reliable intrinsic connectivity networks: Test–retest evaluation using ICA and dual regression approach

Reactivity of hemodynamic responses and functional connectivity to different states of alpha synchrony: A concurrent EEG-fMRI study

Spatial parcellations, spectral filtering, and connectivity measures in fMRI: Optimizing for discrimination

Contact Info

Product

Resources

About