SABER: A Semantic-Aligned Brain Network Analysis Framework via Multi-scale Hypergraphs

2026-07-02

AI Summary

Existing approaches to brain disorder diagnosis struggle to effectively integrate the semantic knowledge of large language models (LLMs), limiting model stability and robustness. This work proposes a semantic-aligned brain network analysis framework that, for the first time, deeply embeds LLM-derived semantics throughout the entire modeling pipeline. The framework employs multiscale hypergraphs to capture functional subnetworks and high-order interactions among multiple regions of interest (ROIs), while introducing semantic guidance at both node and decision levels. A novel decision-level semantic alignment mechanism is designed to preserve the original network structure without perturbation. By integrating global self-attention with functional connectivity analysis, the method significantly enhances diagnostic performance, stability, robustness, and interpretability in small-sample settings, as demonstrated on the ABIDE and ADHD-200 datasets.

Abstract

Effective brain disease diagnosis requires the synergy of brain connectivity patterns and high-level semantic knowledge. Existing methods, however, largely treat semantics from large language models (LLMs) as auxiliary features or supervision, limiting their direct role in decision-making and constraining classification stability and robustness. To overcome this, we propose a semantic-aligned brain network framework that actively integrates LLM-derived semantics into the prediction process. Specifically, ROI-level semantics are first incorporated via global self-attention to enrich node representations and provide whole-brain context. Multi-scale hypergraphs are then constructed to explicitly model functional subnetworks and multi-ROI interactions, addressing the locality limitations of traditional GNNs and capturing high-order dependencies. Finally, a decision-level semantic alignment mechanism selectively injects patient-specific textual embeddings into graph representations, enabling semantics to directly guide predictions without perturbing the underlying network structure. Experiments on public brain network datasets ABIDE and ADHD-200 demonstrate state-of-the-art performance, enhanced stability, and improved interpretability, particularly in small-sample settings.

Problem

• brain disease diagnosis
• semantic knowledge
• large language models
• decision-making
• classification stability

Innovation

• semantic alignment
• multi-scale hypergraphs
• brain network analysis
• large language models
• graph representation learning