Abstract

BACKGROUND Linking structural connectivity (SC) to functional connectivity (FC) through mechanistic models remained a fundamental challenge in network neuroscience. It is yet underexplored for translational animal brain models with traumatic brain injury (TBI). In this study, we evaluated the Kuramoto model from multiple network aspects and graph features using swine TBI model.

NEW METHOD We evaluated a structurally constrained Kuramoto phase-oscillator framework to reproduce resting-state FC in normal swine connectome model and to assess its sensitivity to different TBI severities including sham, mild and severe TBI and its longitudinal progression. Unlike previous studies, we implemented a joint optimization procedure to calibrate data-informed natural frequencies and global coupling strength. Further, Kuramoto model with tuned parameters was used to evolve phases of oscillators from SC. To reproduce blood oxygenated level dependent (BOLD) signals, the Balloon-Windkessel hemodynamic model was then incorporated.

RESULTS Averaged simulated FC (sFC) demonstrated significant edgewise correspondence with averaged empirical FC(eFC, r = 0.61, p < 0.001). This correspondence exceeded the null model benchmarks based on degree-preserving randomized SC. A longitudinal comparison of Graph topological properties between eFC and sFC across connection densities (30-50%) demonstrated strong agreement for global efficiency, characteristic path length, and clustering coefficient, while modularity and small-worldness showed deviations with TBI severity-dependence.

COMPARISON TO EXISTING METHODS Data-informed calibration of natural frequencies leveraged the model to capture both edge-based and topological properties of eFC while preserving inter-subject heterogeneity.

CONCLUSION The framework provides a scalable, mechanistically interpretable platform for investigating structure-function coupling and its perturbation using the translational swine TBI model.