Exploring the behavioral landscape of functional brain networks
Poster No:
1089
Submission Type:
Abstract Submission
Authors:
Santino Iannone1, AJ Simon1, Kira Tang1, Anja Samardzija1, Saloni Mehta1, Jagriti Arora1, Todd Constable1
Institutions:
1Yale University, New Haven, CT
First Author:
Co-Author(s):
Introduction:
The challenge of predicting clinically relevant behavioral traits from neurobiological measures has been of considerable interest within neuroscience since its inception. Discoveries to this aim have the potential to reveal biological targets for therapeutic development, enhance the fidelity of treatment assignment, uncover the neural correlates of subjective experience, and so on. Attempts in brain-behavior predictive modeling have typically taken a symptom-centric approach, where behavioral constructs of interest are selected and correlated biomarkers are identified. However, most clinically relevant behavioral traits are constituted in internal subjective experience and must be measured through introspective self-reporting. As a consequence, self-report measures often conflate independent, disparate, and even contradictory behavioral constructs, which inhibits predictive performance. Here we demonstrate a network-centric approach for brain-behavioral modeling that reveals the complex symptom profiles of major functional brain networks.
Methods:
A transdiagnostic cohort of participants (n=317) spanning a range of psychopathology were recruited and completed an intake examination consisting of 321 self-reported test items over seven behavioral questionnaires. Participants then underwent eight fMRI runs: six cognitive tasks bookended by two periods of unconstrained rest. Voxels were parcellated into 268 contiguous, non-overlapping regions in accordance with the Shen268 atlas (Shen, 2013) and participant-specific functional connectivity matrices were then generated for each run. Kernel Ridge Regression (KRR) models (He, 2020) were trained on connectivity features within specified networks to predict participant item responses. Training consisted of 100 independent, randomly permuted iterations of 10-fold cross validation (CV) for each item, with prediction performance being calculated as the Spearman correlation coefficient (ρ) of all test predictions across CV folds with the true response values. Item responses were predicted from each of the ten Shen canonical networks. Self-report items were then organized based on statistically significant, rank-order prediction performance for each tested network.
Results:
Overall, 55% of self-report items across the seven measures could be significantly predicted by at least one network (ρ > 0.15), with only 23% of items being predicted by multiple networks (and no items being predicted by 5 or more networks), highlighting network-specificity of item prediction. The observed predictability patterns are not explained by item response statistics such as bias or variance. Items were organized into network-specific groups (top 10 predicted items per network) and displayed little overlap in profile between networks (max 2 items). Top items for each network showed high predictive performance (ρ=0.23-0.31).
·Cumulative rank-order KRR item prediction performance for Shen functional networks
·Top ranked items for Shen networks, colored by measure
Conclusions:
By leveraging a network-centric approach to brain-behavior predictive modeling, unique symptom profiles can be revealed for specific functional brain networks. These distinct profiles can be used to further explore relationships between behavioral constructs and underlying neurobiological systems, discover biomarkers of symptom complexes, and design brain-based behavioral measures that optimally reflect network dysfunction.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 2
Modeling and Analysis Methods:
Classification and Predictive Modeling 1
Connectivity (eg. functional, effective, structural)
fMRI Connectivity and Network Modeling
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
Computational Neuroscience
FUNCTIONAL MRI
Machine Learning
Psychiatric
Psychiatric Disorders
1|2Indicates the priority used for review
Abstract Information
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.
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:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
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.
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.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
He, T., Kong, R., Holmes, A. J., Nguyen, M., Sabuncu, M. R., Eickhoff, S. B., Bzdok, D., Feng, J., & Yeo, B. T. T. (2020). Deep neural networks and kernel regression achieve comparable accuracies for functional connectivity prediction of behavior and demographics. NeuroImage, 206, 116276. https://doi.org/10.1016/j.neuroimage.2019.116276