Poster No:
1810
Submission Type:
Abstract Submission
Authors:
Shreyas Harita1,2, Davide Momi2, John Griffiths2,3
Institutions:
1Institute of Medical Science, University of Toronto, Toronto, ON, Canada, 2Centre for Addiction and Mental Health, Toronto, ON, Canada, 3Department of Psychiatry, University of Toronto, Toronto, ON, Canada
First Author:
Shreyas Harita
Institute of Medical Science, University of Toronto|Centre for Addiction and Mental Health
Toronto, ON, Canada|Toronto, ON, Canada
Co-Author(s):
Davide Momi
Centre for Addiction and Mental Health
Toronto, ON, Canada
John Griffiths
Centre for Addiction and Mental Health|Department of Psychiatry, University of Toronto
Toronto, ON, Canada|Toronto, ON, Canada
Introduction:
Introduction
Resting-state brain activity, as observed through functional magnetic resonance imaging (fMRI), reveals structured neural patterns and resting-state networks (RSNs) (1). These RSNs, such as visual (VIS), somatomotor (SMT), dorsal attention (DAN), ventral attention (VAN), limbic (LIM), frontoparietal (FPN), and default mode (DMN) networks, consistently emerge in individuals across different contexts (2). While RSNs' functional connectivity varies, structural connectivity (SC) remains relatively stable and underlies neural coordination (3).
Present Study
This study explores the influence of brain structural connectivity (SC), on the existence and interactions of RSNs. Specifically, we examine how inter-network connections affect functional connectivity within and between these networks and investigate how these networks communicate based on their SC.
Methods:
Methods
We computed SC matrices for 200 randomly selected subjects from the Human Connectome Project (4) using diffusion-weighted imaging data and preprocessing software, including FSL (5), MRtrix3 (6), and FreeSurfer (7). To model mesoscopic brain activity, we utilized the dynamic mean field (DMF) model, characterizing each network node with excitatory and inhibitory neural masses (8, 9). We implemented our brain network model using the WhoBPyT library (10), enabling gradient-based parameter optimization via the ADAM algorithm. This approach bridges physiological brain network models with deep recurrent neural networks. After confirming the replication of resting-state fMRI (rs-fMRI) time series, we selectively lesioned structural connectivity to isolate the seven RSNs. Our goal was to explore how network isolation impacts the relationships between networks, primarily focusing on average functional connectivity (FC) and average SC communicability.
Results:
Results
Our results reveal a clear distinction between two network groups: lower order networks (LONs) and higher order networks (HONs). LONs encompass VIS, SMT, DAN, and VAN, while HONs include LIM, FPN, and DMN. Following LON lesions, average FC within LONs decreased significantly by 29.2%, with a 12.4% decrease in FC between LONs. Changes between LONs and HONs were not significant. For HON lesions, there was a significant 8.9% decrease in average FC within HONs, while FC between HON networks remained stable. LIM and FPN lesions had no significant effect on FC between HONs and LONs. However, a DMN lesion led to a significant 2.3% increase in average FC within and between LON networks.
Regarding SC communicability, we observed a significant reduction within and between LONs and HONs, regardless of whether a LON or HON was lesioned. Following LON lesions, average SC communicability within LONs decreased by 6.9%, with a 2.5% decrease between LONs and a 1.9% decrease between LONs and HONs. HON lesions resulted in an 8.4% decrease in average SC communicability within HONs, along with a 4.0% decrease between HONs and a 2.9% decrease between HONs and LONs.
Conclusions:
Conclusions
Our findings provide insights into the emergence and interactions of RSNs in the context of structural connectivity. LON networks exhibit strong interconnections, leading to reduced FC and SC communicability when one LON network is lesioned. Conversely, HON networks appear to operate independently and display limited interdependence among HON networks. This divergence may be attributed to the nature of LON and HON networks. LONs process unimodal information, fostering interactions with similar networks, while HONs are multimodal, processing diverse information (11). Furthermore, this decrease in FC following network lesioning,particularly for LONs, can also be attributed to reduced SC communicability, affecting efficient information propagation. Our study introduces a novel hypothesis regarding the role of structural connectivity in shaping RSN existence and interactions in fMRI data, contributing to the understanding of brain network dynamics.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 2
fMRI Connectivity and Network Modeling 1
Keywords:
Computational Neuroscience
FUNCTIONAL MRI
Modeling
1|2Indicates the priority used for review
Provide references using author date format
Biswal B, Yetkin FZ, Haughton VM, Hyde JS. Functional connectivity in the motor cortex of resting human brain using echo-planar MRI. Magn Reson Med. 1995;34(4):537-541.
[2] Rosazza C, Minati L. Resting-state brain networks: literature review and clinical applications. Neurol Sci. 2011;32(5):773-785.
[3] Seguin C, Sporns O, Zalesky A. Brain network communication: concepts, models and applications. Nat Rev Neurosci. Published online July 12, 2023. doi:10.1038/s41583-023-00718-5
[4] Van Essen DC, Ugurbil K, Auerbach E, Barch D, Behrens TEJ, Bucholz R, et al. The Human Connectome Project: a data acquisition perspective. Neuroimage. 2012;62(4):2222-2231.
[5] Andersson JLR, Sotiropoulos SN. An integrated approach to correction for off-resonance effects and subject movement in diffusion MR imaging. Neuroimage. 2016;125:1063-1078.
[6] Tournier JD, Calamante F, Connelly A. MRtrix: Diffusion tractography in crossing fiber regions. Int J Imaging Syst Technol. 2012;22(1):53-66.
[7] Fischl B. FreeSurfer. Neuroimage. 2012;62(2):774-781.
[8] Deco G, Ponce-Alvarez A, Mantini D, Romani GL, Hagmann P, Corbetta M. Resting-state functional connectivity emerges from structurally and dynamically shaped slow linear fluctuations. J Neurosci. 2013;33(27):11239-11252.
[9] Deco G, Ponce-Alvarez A, Hagmann P, Romani GL, Mantini D, Corbetta M. How local excitation-inhibition ratio impacts the whole brain dynamics. J Neurosci. 2014;34(23):7886-7898.
[10] Griffiths JD, Wang Z, Ather SH, Momi D, Rich S, Diaconescu A, et al. Deep Learning-Based Parameter Estimation for Neurophysiological Models of Neuroimaging Data. bioRxiv. Published online May 19, 2022:2022.05.19.492664. doi:10.1101/2022.05.19.492664
[11] Margulies DS, Ghosh SS, Goulas A, Falkiewicz M, Huntenburg JM, Langs G, et al. Situating the default-mode network along a principal gradient of macroscale cortical organization. Proc Natl Acad Sci U S A. 2016;113(44):12574-12579.