Abstract
Resting-state functional brain imaging studies of network connectivity have long assumed that functional connections are stationary on the timescale of a typical scan. Interest in moving beyond this simplifying assumption has emerged only recently. The great hope is that training the right lens on time-varying properties of whole-brain network connectivity will shed additional light on previously concealed brain activation patterns characteristic of serious neurological or psychiatric disorders. We present evidence that multiple explicitly dynamical properties of time-varying whole-brain network connectivity are strongly associated with schizophrenia, a complex mental illness whose symptomatic presentation can vary enormously across subjects. As with so much brain-imaging research, a central challenge for dynamic network connectivity lies in determining transformations of the data that both reduce its dimensionality and expose features that are strongly predictive of important population characteristics. Our paper introduces an elegant, simple method of reducing and organizing data around which a large constellation of mutually informative and intuitive dynamical analyses can be performed. This framework combines a discrete multidimensional data-driven representation of connectivity space with four core dynamism measures computed from large-scale properties of each subject’s trajectory, ie., properties not identifiable with any specific moment in time and therefore reasonable to employ in settings lacking inter-subject time-alignment, such as resting-state functional imaging studies. Our analysis exposes pronounced differences between schizophrenia patients (Nsz = 151) and healthy controls (Nhc = 163). Time-varying whole-brain network connectivity patterns are found to be markedly less dynamically active in schizophrenia patients, an effect that is even more pronounced in patients with high levels of hallucinatory behavior. To the best of our knowledge this is the first demonstration that high-level dynamic properties of whole-brain connectivity, generic enough to be commensurable under many decompositions of time-varying connectivity data, exhibit robust and systematic differences between schizophrenia patients and healthy controls.
Highlights
We present evidence that multiple explicitly dynamical properties of time-varying whole-brain network connectivity are strongly associated with schizophrenia, a complex mental illness whose symptomatic presentation can vary enormously across subjects
We introduce a set of simple dynamism measures calculated from subject trajectories through the induced discrete five-dimensional statespace, finding consistent, significant and replicable differences in connectivity dynamics between schizophrenia patients and healthy controls (Fig 2B and 2D)
For consistency with the other decompositions we report the results from regressing windowed functional network connectivity matrices (wFNC) data on k-means Correlation Patterns (CPs), but results using the alternative weighting system presented the same directionality and significance
Summary
Cognitive and psychiatric disorders have been shown to affect connectivity between functional brain networks [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24] even in so-called "resting" conditions where subjects are not engaged in a task. Network connectivity is typically assessed as a stationary feature of the data, inferred from the correlation or mutual information between pairs of network activation timecourses that extend through the duration of the scan. Recent investigations of dynamic connectivity have shown that connections are varying through time [25,26,27,28,29,30,31,32,33,34,35,36], but that this variation takes different forms in different demographic [35] and diagnostic [16, 26, 30, 32, 33, 37–
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.