A Riemannian Framework for Longitudinal Analysis of Resting-State Functional Connectivity

Zhao, Qingyu; Kwon, Dong-Jin; Pohl, Kilian M.

doi:10.1007/978-3-030-00931-1_17

Cited by 15 publications

(31 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To center each subject's correlation matrices, we took the approach advocated by Zhao et al (2018), which leverages the natural geometry of the space of covariance matrices. We have implemented many of the computations required to replicate the analysis in an R package spdm (symmetric positive-definite matrix), which is freely available from a Git repository at https://gitlab.com/fmriToolkit/spdm.…”

Section: Functional Connectivity Estimation and Comparisonmentioning

confidence: 99%

See 1 more Smart Citation

Muting, not fragmentation, of functional brain networks under general anesthesia

Areshenkoff

Nashed

Hutchison

et al. 2020

Preprint

View full text Add to dashboard Cite

Changes in resting-state functional connectivity (rs-FC) under general anesthesia have been widely studied with the goal of identifying neural signatures of consciousness. This work has commonly revealed an apparent fragmentation of whole-brain network structure during unconsciousness, which has been interpreted as reflecting a break-down in connectivity and a disruption in the brain's ability to integrate information. Here we show, by studying rs-FC under varying depths of isoflurane-induced anesthesia in nonhuman primates, that this apparent fragmentation, rather than reflecting an actual change in network structure, can be simply explained as the result of a global reduction in FC. Specifically, by comparing the actual FC data to surrogate data sets that we derived to test competing hypotheses of how FC changes as a function of dose, we found that increases in whole-brain modularity and the number of network communities -- considered hallmarks of fragmentation -- are artifacts of constructing FC networks by thresholding based on correlation magnitude. Taken together, our findings suggest that deepening levels of unconsciousness are instead associated with the increasingly muted expression of functional networks, an observation that constrains current interpretations as to how anesthesia-induced FC changes map onto existing neurobiological theories of consciousness.

show abstract

Section: Functional Connectivity Estimation and Comparisonmentioning

confidence: 99%

“…where log denotes the matrix logarithm. We then transported each tangent vector to the grand meanS gm using the transport proposed by Zhao et al (2018), obtaining a centered tangent vector…”

Section: Functional Connectivity Estimation and Comparisonmentioning

confidence: 99%

Muting, not fragmentation, of functional brain networks under general anesthesia

Areshenkoff

Nashed

Hutchison

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…We then translated each tangent vector to the grand mean resting state scanR using the transport proposed by Zhao et al (2018), obtaining a centered tangent vector…”

Section: Covariance Estimation Centering and Embeddingmentioning

confidence: 99%

Neural Excursions from Low-Dimensional Manifold Structure Explain Intersubject Variation in Human Motor Learning

Areshenkoff

Gale

Nashed

et al. 2021

Preprint

View full text Add to dashboard Cite

Humans vary greatly in their motor learning abilities, yet little is known about the neural mechanisms that underlie this variability. Recent neuroimaging and electrophysiological studies demonstrate that large-scale neural dynamics inhabit a low-dimensional subspace or manifold, and that learning is constrained by this intrinsic manifold architecture. Here we asked, using functional MRI, whether subject-level differences in neural excursion from manifold structure can explain differences in learning across participants. We had subjects perform a sensorimotor adaptation task in the MRI scanner on two consecutive days, allowing us to assess their learning performance across days, as well as continuously measure brain activity. We find that the overall neural excursion from manifold activity in both cognitive and sensorimotor brain networks is associated with differences in subjects' patterns of learning and relearning across days. These findings suggest that off-manifold activity provides an index of the relative engagement of different neural systems during learning, and that intersubject differences in patterns of learning and relearning across days are related to reconfiguration processes in cognitive and sensorimotor networks during learning.

show abstract

“…These methods usually require fitting probability distributions in a high dimensional space, which is a challenging task. Moreover, it has been found that the underlying distributions of both fMRI measurements [6] and the derived correlation matrices [10] lie on a non-linear latent space. Therefore, modeling Gaussian-mixtures in the original space is suboptimal.…”

Section: Related Workmentioning

confidence: 99%

Variational Autoencoder with Truncated Mixture of Gaussians for Functional Connectivity Analysis

Zhao

Honnorat

Adeli

et al. 2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Resting-state functional connectivity states are often identified as clusters of dynamic connectivity patterns. However, existing clustering approaches do not distinguish major states from rarely occurring minor states and hence are sensitive to noise. To address this issue, we propose to model major states using a non-linear generative process guided by a Gaussian-mixture distribution in a low-dimensional latent space, while separately modeling the connectivity patterns of minor states by a non-informative uniform distribution. We embed this truncated Gaussian-Mixture model in a Variational Autoencoder framework to obtain a general joint clustering and outlier detection approach, tGM-VAE. When applied to synthetic data with known ground-truth, tGM-VAE is more accurate in clustering connectivity patterns than existing approaches. On the rs-fMRI of 593 healthy adolescents, tGM-VAE identifies meaningful major connectivity states. The dwell time of these states significantly correlates with age.

show abstract

A Riemannian Framework for Longitudinal Analysis of Resting-State Functional Connectivity

Cited by 15 publications

References 14 publications

Muting, not fragmentation, of functional brain networks under general anesthesia

Muting, not fragmentation, of functional brain networks under general anesthesia

Neural Excursions from Low-Dimensional Manifold Structure Explain Intersubject Variation in Human Motor Learning

Variational Autoencoder with Truncated Mixture of Gaussians for Functional Connectivity Analysis

Contact Info

Product

Resources

About