Uncovering Latent Communication Patterns in Brain Networks via Adaptive Flow Routing

TL;DR

AFR-Net reframes SC–FC fusion as a latent, energy-efficient flow routing problem on the structural connectome. It combines a physics-informed graph construction with a differentiable, closed-form flow solver to compute edge-wise information flow guided by FC demands, then uses a pattern-guided attention mechanism to improve disease classification. The approach yields state-of-the-art accuracy across multiple brain-disease benchmarks and provides biologically interpretable routing patterns, such as pathway hubs and topological detours, aligning with known neuroscientific findings. This work advances both predictive performance and mechanistic understandability in multimodal connectomics, with potential for mechanistic biomarker discovery and improved clinical insight.

Abstract

Unraveling how macroscopic cognitive phenotypes emerge from microscopic neuronal connectivity remains one of the core pursuits of neuroscience. To this end, researchers typically leverage multi-modal information from structural connectivity (SC) and functional connectivity (FC) to complete downstream tasks. Recent methodologies explore the intricate coupling mechanisms between SC and FC, attempting to fuse their representations at the regional level. However, lacking fundamental neuroscientific insight, these approaches fail to uncover the latent interactions between neural regions underlying these connectomes, and thus cannot explain why SC and FC exhibit dynamic states of both coupling and heterogeneity. In this paper, we formulate multi-modal fusion through the lens of neural communication dynamics and propose the Adaptive Flow Routing Network (AFR-Net), a physics-informed framework that models how structural constraints (SC) give rise to functional communication patterns (FC), enabling interpretable discovery of critical neural pathways. Extensive experiments demonstrate that AFR-Net significantly outperforms state-of-the-art baselines. The code is available at https://anonymous.4open.science/r/DIAL-F0D1.

Figures (5)

• Figure 1: Neural communication models can be categorized along a spectrum ranging from diffusion to routing, based on the extent of global topological information available to the network. We hypothesize that biological neural communication operates in an intermediate regime, utilizing multiple alternative paths to ensure robustness and efficiency. In the illustration, the thickness of each path represents its information flow capacity, which is derived from the reciprocal of the connection strength.
• Figure 2: The overall framework of our proposed method AFR-Net. The model explicitly fuses two modalities of structural and functional connectivity via three integrated phases: (1) Physics-Informed Graph Construction, which initializes structure-aware node representations using effective resistance distance (ERD) and constructs a dynamic flow network with learnable edge capacities calculated by edge gate; (2) Differentiable Information Flow Solver, which computes a closed-form solution for global information traffic driven by functional demands ($\mathbf{L}_{\text{fc}}$) through structural constraints ($\mathbf{L}_{\text{flow}}$); and (3) Pattern-Guided Aggregation, which utilizes the learned flow patterns to guide message passing for downstream classification tasks.
• Figure 3: (a) Visualization of the top 100 edges with the highest information flow intensity ($\Phi_{ij}$) averaged across all subjects. The model autonomously identifies the visual and somatomotor networks as the structural core of brain communication. (b) Circle plot showing edges with significant group differences ($p < 0.05$, FDR corrected). Red lines denote edges where the patient exhibits higher information flow intensity than HC, whereas blue lines indicate the opposite. The left hippocampus (L_H, marked with ***) emerges as a pathological hub, exhibiting dense hyper-connectivity with the default mode network and subcortical regions in patient group.
• Figure 4: Accuracy vs. Time Cost on ABCD dataset over 50 epochs for each method. Our method achieves the best trade-off between performance and efficiency.
• Figure 5: (a) The Informational Backbone. Visualization of the top 100 edges with the highest information flow intensity averaged across all subjects in the OCD dataset. Consistent with the main findings, the model identifies the visual and somatomotor networks as the stable structural core. (b) Pathological Hyper-connectivity in OCD. A circle plot showing edges with significant group differences ($p < 0.05$, FDR corrected). Red lines indicate hyper-active flow (Patient $>$ HC). The Primary Visual Cortex (V1/V2) and Hippocampus (H, marked with ***) emerge as pathological hubs, exhibiting dense hyper-connectivity between sensory processing and memory encoding circuits.

Theorems & Definitions (2)

• Theorem 2.1: Closed-Form Solution of Total Information Flow
• proof