Poster No:
1264
Submission Type:
Abstract Submission
Authors:
Parsa Oveisi1, Zheng Wang1, Davide Momi2, John Griffiths3
Institutions:
1Centre for Addiction and Mental Health, Toronto, Ontario, 2Stanford University, Stanford, CA, 3University of Toronto, Toronto, Ontario
First Author:
Parsa Oveisi
Centre for Addiction and Mental Health
Toronto, Ontario
Co-Author(s):
Zheng Wang
Centre for Addiction and Mental Health
Toronto, Ontario
Introduction:
Linear dynamical system (LDS) models have emerged as fundamental tools for understanding the complex relationship between brain structural connectivity (SC) and functional connectivity (FC), particularly in whole-brain connectome-based modeling. Traditional LDS approaches operate on a central axiom: that SC and FC share identical eigenvectors and are related through simple mathematical transformations such as inverse-square or negative exponential functions. While these models have provided valuable insights into brain organization, they often yield only modest FC predictions when applied to empirical data. This limitation is particularly evident in their inability to accurately capture inter-hemispheric connections and network modularity patterns observed in human neuroimaging data. The persistent gap between model predictions and empirical observations suggests the need for a more sophisticated approach that preserves the theoretical foundations of LDS while allowing for more flexible structure-function relationships. The challenge lies in improving predictive accuracy without compromising the fundamental mathematical principles that make LDS models theoretically appealing.
Methods:
We present a novel mixed linear modeling approach that optimizes SC eigenvalues while maintaining the core underlying LDS assumptions. Our method begins with eigendecomposition of empirical SC into eigenvectors and eigenvalues, followed by projection of empirical FC onto the SC eigenvector space. This enables optimization of SC eigenvalues while preserving the original eigenvector structure, ensuring that fundamental network structure intact. Using high-quality Human Connectome Project data (n=64), we compared our approach against a traditional LDS model that directly applies the LDS equation to empirical SC. Model performance was evaluated using Pearson correlation between predicted and empirical FC matrices using both the traditional and our mixed approach. Additionally, we systematically assessed model performance under different levels of eigenmode truncation to identify the optimal dimensionality for structure-function mapping.
Results:
Our mixed model significantly outperformed traditional LDS approaches, achieving a 20% improvement in FC prediction accuracy (mean correlation with empirical FC: r=0.421±0.043 vs. r=0.379±0.052, p<0.001). Analysis of eigenvalue scaling relationships revealed that our model better captures empirical FC patterns, particularly in inter-hemispheric connections and network modules. Furthermore, our analysis revealed that approximately the first 20 structural modes are sufficient to capture the essential features of brain functional organization, while the traditional LDS approach showed inconsistent performance across different numbers of modes. This efficiency in using a limited number of eigenmodes suggests low-dimensional principles governing structure-function relationships in brain networks.
Conclusions:
This mathematically principled approach enhances our ability to predict functional brain organization from structural connectivity, with important implications for understanding network disruptions in neurological disorders and developing connectivity-based biomarkers for clinical applications. The success of this approach demonstrates the value of maintaining theoretical constraints while introducing targeted flexibility in structure-function mapping.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 1
fMRI Connectivity and Network Modeling
Methods Development 2
Task-Independent and Resting-State Analysis
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Neuroanatomy Other
Keywords:
Computational Neuroscience
FUNCTIONAL MRI
Machine Learning
Modeling
STRUCTURAL MRI
1|2Indicates the priority used for review
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
I accept
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information.
Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
I am submitting this abstract as an original work to be reproduced. I am available to be the “source party” in an upcoming team and consent to have this work listed on the OSSIG website. I agree to be contacted by OSSIG regarding the challenge and may share data used in this abstract with another team.
Please indicate below if your study was a "resting state" or "task-activation” study.
Resting state
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Healthy subjects
Was this research conducted in the United States?
No
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel?
NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Yes
Were any animal research approved by the relevant IACUC or other animal research panel?
NOTE: Any animal studies without IACUC approval will be automatically rejected.
Not applicable
Please indicate which methods were used in your research:
Functional MRI
Structural MRI
Computational modeling
Provide references using APA citation style.
not applicable
No