SWpC: A Sensitive Approach for Detecting Task-Specific Directed Connectivity Dynamics
Poster No:
1233
Submission Type:
Abstract Submission
Authors:
Nan Xu1, Vince Calhoun2, Shella Keilholz3
Institutions:
1University of Maryland, College Park, MD, 2GSU/GATech/Emory, Atlanta, GA, 3Georgia Tech-Emory, Atlanta, GA
First Author:
Co-Author(s):
Introduction:
Functional connectivity (FC) analyses are integral for understanding the temporal dynamics and directional influences among brain regions [1,2]. Traditional correlation-based measures [1–3] provide insight into synchronous neural activity patterns, yet they often lack the ability to capture directional causal relationships [4] and can overlook subtle temporal changes [4,5]. Within the increasingly common framework of time-varying FC, there is a growing need for methods that can disentangle the directionality of information flow and the duration of these directed interactions. To address these gaps, we developed sliding-window prediction correlation (SWpC), a novel approach that combines sliding-window correlation with a linear time-invariant causal model. By applying SWpC to functional magnetic resonance imaging (fMRI) data collected during motor tasks, we aim to validate the method's sensitivity in detecting task-related directed influences and in characterizing how these influences vary over time.
Methods:
SWpC combines sliding-window correlation (SWC) with a linear time-invariant causal model to evaluate time-varying directed functional connectivity (FC). Within each window [t, t+L], SWpC predicts output y from input x to estimate information transfer duration D_yx[t] and directed connectivity strength strength ρ_yx[t]=corr(ý[t], y[t]), where ý[t] is the predicted y[t]. By correlating predicted and observed signals, SWpC dynamically estimates connectivity strength and duration while distinguishing forward and backward connections (ρ_yx[t]≠ρ_xy[t] and D_yx[t]≠D_xy[t] for any t).
To validate SWpC sensitivity, minimally preprocessed Human Connectome Project (HCP) motor task fMRI data from 45 test-retest subjects was analyzed. Subjects performed visually cued movements (left foot, right foot, left hand, right hand, tongue) across two runs, with 12-second task/rest trials and 3-second cues. Task-activated ROIs were identified using a three-level FEAT analysis (Z ≥ 4.42, p < 0.001).
Task-specific differences in FC strength were evaluated by comparing task and rest matrices for SWpC (directed FC) and SWC (standard FC). Significant connections were identified using t-tests (p < 0.01, FDR-corrected) and filtered for low variability (CV < 30%). Overall differences were quantified using the Frobenius norm, and sensitivity was determined by comparing mean Frobenius norms between SWpC and SWC.
To validate SWpC sensitivity, minimally preprocessed Human Connectome Project (HCP) motor task fMRI data from 45 test-retest subjects was analyzed. Subjects performed visually cued movements (left foot, right foot, left hand, right hand, tongue) across two runs, with 12-second task/rest trials and 3-second cues. Task-activated ROIs were identified using a three-level FEAT analysis (Z ≥ 4.42, p < 0.001).
Task-specific differences in FC strength were evaluated by comparing task and rest matrices for SWpC (directed FC) and SWC (standard FC). Significant connections were identified using t-tests (p < 0.01, FDR-corrected) and filtered for low variability (CV < 30%). Overall differences were quantified using the Frobenius norm, and sensitivity was determined by comparing mean Frobenius norms between SWpC and SWC.
Results:
27 ROIs were identified to be mostly responsive to motion tasks, consistent with prior studies [6,7]. Asymmetry measures for directed FC during task were significantly higher than rest, with Cohen's d > 0.8 (Fig 1a). Significant differences in duration of information flow between task and rest conditions were observed in hub-like ROIs within somatomotor areas, showing increased long-range efferent information transfer to these ROIs during hand and foot tasks (Fig. 1c). These regions showed delayed and elevated hemodynamic peaks compared to other ROIs, explaining prolonged durations (Fig. 1b). Regarding the (directed) FC strength, SWpC not only predicted desired directed FC but also demonstrated higher sensitivity in detecting task-specific connections compared to SWC (Fig. 2).
·Fig. 1: Asymmetry in directed functional connectivity and significant changes in information flow duration during task conditions.
·Fig. 2: Enhanced sensitivity of SWpC compared to SWC in detecting task-specific directed functional connectivity.
Conclusions:
In summary, SWpC provides a robust framework for identifying and characterizing task-driven directed functional connectivity changes over time. By leveraging a combination of sliding-window analysis and a linear time-invariant causal model, SWpC uncovers not only the directionality of interactions between regions but also the varying temporal durations of these directed influences. Our results highlight that SWpC is more sensitive than standard correlation-based methods (SWC) in discerning task-specific network dynamics. These findings underscore the utility of SWpC as a valuable tool for researchers seeking a deeper understanding of the causal structure and temporal evolution of brain networks, paving the way for more nuanced interpretations of functional neuroimaging data.
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI) 2
Connectivity (eg. functional, effective, structural) 1
fMRI Connectivity and Network Modeling
Keywords:
ADULTS
Computational Neuroscience
Data analysis
FUNCTIONAL MRI
Informatics
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.
2. Allen, E. A. et al. Tracking whole-brain connectivity dynamics in the resting state. Cereb Cortex 24, 663–676 (2012).
3. Shakil, S., Lee, C. H. & Keilholz, S. D. Evaluation of sliding window correlation performance for characterizing dynamic functional connectivity and brain states. Neuroimage 133, 111–128 (2016).
4. Mitra, A., Snyder, A. Z., Blazey, T. & Raichle, M. E. Lag threads organize the brain’s intrinsic activity. Proc Natl Acad Sci U S A 112, E2235-44 (2015).
5. Xu, N., Doerschuk, P. C., Keilholz, S. D. & Spreng, R. N. Spatiotemporal functional interactivity among large-scale brain networks. Neuroimage 227, 117628 (2021).
6. Tripathi, V. et al. Utilizing connectome fingerprinting functional MRI models for motor activity prediction in presurgical planning: A feasibility study. Hum Brain Mapp 45, (2024).
7. Barch, D. M. et al. Function in the human connectome: task-fMRI and individual differences in behavior. Neuroimage 80, 169–189 (2013).