Comparison between gradients and parcellations for functional connectivity prediction of behavior

Kong, Ru; Tan, Yan Zhi; Wulan, Naren; Ooi, Leon Qi Rong; Farahibozorg, Seyedeh-Rezvan; Harrison, Sarah M.; Bijsterbosch, Janine; Bernhardt, Boris C.; Eickhoff, Simon B.; Yeo, B.T. Thomas

doi:10.1016/j.neuroimage.2023.120044

Cited by 24 publications

(15 citation statements)

References 74 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our research not only replicated previous findings on the predictive capacity of FC edges for the cognitive factor [10][11][12]20] but also expanded them by pinpointing a specific set of FC edges within association cortex networks like the DMN, FPCN, and attention network. This discovery aligns with existing findings on the necessity of coordinated activities across distributed brain regions, primarily within the association cortex, for numerous cognitive functions [6,9,45,46].…”

Section: Discussionsupporting

confidence: 86%

“…To identify the FC edges that have a significant influence on predicting cognitive factor score, we employed ridge regression to examine the relationship between all FC edges and cognitive factor score obtained from the CFA model [11,17,20,37]. We set the second-order cognitive factor score as the dependent variable, enabling us to calculate the regression coefficient (β value) for each edge corresponding to the cognitive factor score.…”

Section: Methodsmentioning

confidence: 99%

“…This parcellation divides the whole brain into 400 parcels and has been extensively utilized [30][31][32]. Previous studies showed commendable performance when using this parcellation in the context of FC network analysis to predict cognitive functions [11,33,34]. We evaluated FC by calculating the Pearson correlation coefficient between the timeseries of parcels.…”

Section: (B) the Pre-processing Of Fmri Data And Functional Connectiv...mentioning

confidence: 99%

“…For instance, Dubois et al [10] utilized exploratory factor analysis (EFA) to establish a classical cognitive ontology entity, namely the g-factor, and demonstrated its predictability through distributed FC networks. In contrast, Kong et al [11] applied factor analysis to develop a different cognitive ontology entity, the cognitive factor, without examining its predictability through FC networks. This disparity highlights the need for a coherent understanding of FC networks’ role in cognition prediction.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Unveiling the Core Functional Networks of Cognition: An Ontology-Guided Machine Learning Approach

Wu,

Cui,

Wang

et al. 2024

Preprint

View full text Add to dashboard Cite

Deciphering the functional architecture that underpins diverse cognitive functions is fundamental quest in neuroscience. In this study, we employed an innovative machine learning framework that integrated cognitive ontology with functional connectivity analysis to identify brain networks essential for cognition. We identified a core assembly of functional connectomes, primarily located within the association cortex, which showed superior predictive performance compared to two conventional methods widely employed in previous research across various cognitive domains. Our approach achieved a mean prediction accuracy of 0.13 across 16 cognitive tasks, including working memory, reading comprehension, and sustained attention, outperforming the traditional methods' accuracy of 0.08. In contrast, our method showed limited predictive power for sensory, motor, and emotional functions, with a mean prediction accuracy of 0.03 across 9 relevant tasks, slightly lower than the traditional methods' accuracy of 0.04. These cognitive connectomes were further characterized by distinctive patterns of resting-state functional connectivity, structural connectivity via white matter tracts, and gene expression, highlighting their neurogenetic underpinnings. Our findings reveal a domain-general functional network fingerprint that pivotal to cognition, offering a novel computational approach to explore the neural foundations of cognitive abilities.

show abstract

Section: Discussionsupporting

confidence: 86%

Section: Methodsmentioning

confidence: 99%

Section: (B) the Pre-processing Of Fmri Data And Functional Connectiv...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Unveiling the Core Functional Networks of Cognition: An Ontology-Guided Machine Learning Approach

Wu,

Cui,

Wang

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Thus, the individual elements of the data (i.e., voxels or vertices) are preserved. Despite conceptual and mathematical differences between gradients and parcellations (Hong et al, 2020), it has been shown that the principal gradients of connectivity performed similarly to parcellations as behavioral predictors (Kong et al, 2023). Notably, gradients of functional connectivity complement parcellation methods by capturing a smoother transition between different functional regions, offering potential insight into hierarchical information processing in the brain (Margulies et al, 2016).…”

Section: Introductionmentioning

confidence: 99%

Methods for decoding cortical gradients of functional connectivity

Peraza

Salo

Riedel³

et al. 2023

Preprint

View full text Add to dashboard Cite

Macroscale gradients have emerged as a central principle for understanding functional brain organization. Previous studies have demonstrated that a principal gradient of connectivity in the human brain exists, with unimodal primary sensorimotor regions situated at one end, and transmodal regions associated with the default mode network and representative of abstract functioning at the other. The functional significance and interpretation of macroscale gradients remains a central topic of discussion in the neuroimaging community, with some studies demonstrating that gradients may be described using meta-analytic functional decoding techniques. However, additional methodological development is necessary to more fully leverage available meta-analytic methods and resources and quantitatively evaluate their relative performance. Here, we conducted a comprehensive series of analyses to investigate and improve the framework of data-driven, meta-analytic methods, thereby establishing a principled approach for gradient segmentation and functional decoding. We found that a small number of segments determined by a K-means segmentation approach and an LDA-based meta-analysis combined with the NeuroQuery database was the optimal combination of methods for decoding functional connectivity gradients. Taken together, the current work aims to provide recommendations on best practices, along with flexible methods, for gradient-based functional decoding of fMRI data.

show abstract

Functional gradients reveal cortical hierarchy changes in multiple sclerosis

De Rosa,

d'Ambrosio,

Bisecco

et al. 2024

Human Brain Mapping

View full text Add to dashboard Cite

Functional gradient (FG) analysis represents an increasingly popular methodological perspective for investigating brain hierarchical organization but whether and how network hierarchy changes concomitant with functional connectivity alterations in multiple sclerosis (MS) has remained elusive. Here, we analyzed FG components to uncover possible alterations in cortical hierarchy using resting‐state functional MRI (rs‐fMRI) data acquired in 122 MS patients and 97 healthy control (HC) subjects. Cortical hierarchy was assessed by deriving regional FG scores from rs‐fMRI connectivity matrices using a functional parcellation of the cerebral cortex. The FG analysis identified a primary (visual‐to‐sensorimotor) and a secondary (sensory‐to‐transmodal) component. Results showed a significant alteration in cortical hierarchy as indexed by regional changes in FG scores in MS patients within the sensorimotor network and a compression (i.e., a reduced standard deviation across all cortical parcels) of the sensory‐transmodal gradient axis, suggesting disrupted segregation between sensory and cognitive processing. Moreover, FG scores within limbic and default mode networks were significantly correlated (, p < .005 after Bonferroni correction for both) with the symbol digit modality test (SDMT) score, a measure of information processing speed commonly used in MS neuropsychological assessments. Finally, leveraging supervised machine learning, we tested the predictive value of network‐level FG features, highlighting the prominent role of the FG scores within the default mode network in the accurate prediction of SDMT scores in MS patients (average mean absolute error of 1.22 ± 0.07 points on a hold‐out set of 24 patients). Our work provides a comprehensive evaluation of FG alterations in MS, shedding light on the hierarchical organization of the MS brain and suggesting that FG connectivity analysis can be regarded as a valuable approach in rs‐fMRI studies across different MS populations.

show abstract

Comparison between gradients and parcellations for functional connectivity prediction of behavior

Cited by 24 publications

References 74 publications

Unveiling the Core Functional Networks of Cognition: An Ontology-Guided Machine Learning Approach

Unveiling the Core Functional Networks of Cognition: An Ontology-Guided Machine Learning Approach

Methods for decoding cortical gradients of functional connectivity

Functional gradients reveal cortical hierarchy changes in multiple sclerosis

Contact Info

Product

Resources

About