Ada-FCN: Adaptive Frequency-Coupled Network for fMRI-Based Brain Disorder Classification

2025-11-06

International Conference on Medical Image Computing and Computer-Assisted Intervention

AI Summary

Existing fMRI-based brain disorder classification methods treat BOLD signals as single-frequency time series, overlooking the intrinsic multi-frequency nature of neural oscillations and relying on predefined frequency bands—limiting adaptability to individual- and disease-specific spectral shifts. To address this, we propose an adaptive cascaded decomposition and band-coupled connectivity learning framework: (1) data-driven adaptive signal decomposition automatically identifies task-relevant frequency bands; (2) a frequency-domain functional connectivity graph is constructed to model cross-band interactions; and (3) a Unified Graph Convolutional Network (Unified-GCN) enables fine-grained regional representation learning via a novel message-passing mechanism. Our approach eliminates reliance on fixed frequency bands and significantly enhances discriminative power. Evaluated on the ADNI and ABIDE datasets, it achieves superior classification accuracy for Alzheimer’s disease and autism spectrum disorder compared to state-of-the-art methods, demonstrating that explicit multi-frequency oscillatory modeling critically improves diagnostic sensitivity and specificity.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

Resting-state fMRI has become a valuable tool for classifying brain disorders and constructing brain functional connectivity networks by tracking BOLD signals across brain regions. However, existing mod els largely neglect the multi-frequency nature of neuronal oscillations, treating BOLD signals as monolithic time series. This overlooks the cru cial fact that neurological disorders often manifest as disruptions within specific frequency bands, limiting diagnostic sensitivity and specificity. While some methods have attempted to incorporate frequency informa tion, they often rely on predefined frequency bands, which may not be optimal for capturing individual variability or disease-specific alterations. To address this, we propose a novel framework featuring Adaptive Cas cade Decomposition to learn task-relevant frequency sub-bands for each brain region and Frequency-Coupled Connectivity Learning to capture both intra- and nuanced cross-band interactions in a unified functional network. This unified network informs a novel message-passing mecha nism within our Unified-GCN, generating refined node representations for diagnostic prediction. Experimental results on the ADNI and ABIDE datasets demonstrate superior performance over existing methods. The code is available at https://github.com/XXYY20221234/Ada-FCN.

Problem

• Classifying brain disorders using fMRI data with multi-frequency neuronal oscillations
• Overcoming limitations of predefined frequency bands in functional connectivity analysis
• Capturing both intra-band and cross-band interactions for improved diagnostic sensitivity

Innovation

• Adaptive Cascade Decomposition learns task-relevant frequency sub-bands
• Frequency-Coupled Connectivity Learning captures cross-band interactions
• Unified-GCN with message-passing generates refined node representations