Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Yue Xun, Jiaxing Xu, Wenbo Gao, Chen Yang, Shujun Wang

TL;DR

Ada-FCN tackles the challenge of multi-frequency information in fMRI by learning task-relevant frequency sub-bands per brain region and modeling both intra-band and cross-band interactions within a unified graph. The approach combines an Adaptive Cascade Decomposer with a Frequency-Coupled Connectivity Learning framework and a cross-band attentive mechanism inside a Unified-GCN, optimized with a multi-term loss that promotes band diversity and sparsity. Experiments on ADNI and ABIDE show state-of-the-art accuracy and AUROC, with ablations confirming the contribution of dynamic thresholding, cross-band attention, and frequency-aware pooling. The results highlight the importance of frequency-specific patterns in neurological disorder classification and offer a framework for interpretable, frequency-aware brain network analysis.

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.

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

TL;DR

Ada-FCN tackles the challenge of multi-frequency information in fMRI by learning task-relevant frequency sub-bands per brain region and modeling both intra-band and cross-band interactions within a unified graph. The approach combines an Adaptive Cascade Decomposer with a Frequency-Coupled Connectivity Learning framework and a cross-band attentive mechanism inside a Unified-GCN, optimized with a multi-term loss that promotes band diversity and sparsity. Experiments on ADNI and ABIDE show state-of-the-art accuracy and AUROC, with ablations confirming the contribution of dynamic thresholding, cross-band attention, and frequency-aware pooling. The results highlight the importance of frequency-specific patterns in neurological disorder classification and offer a framework for interpretable, frequency-aware brain network analysis.

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.

Paper Structure

This paper contains 20 sections, 10 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Method
  3. Adaptive Cascade Decomposer
  4. Frequency-Coupled Connectivity Learning
  5. Intra-Band Connectivity via Dynamic Thresholding.
  6. Cross-Band Coupling with Dual-Projection Bilinear Attention.
  7. Unified Graph Convolution.
  8. Classification Head and Loss Function
  9. Classification Head.
  10. Loss Function.
  11. Experiments and Results
  12. Experiment Settings
  13. Datasets.
  14. Implementation details.
  15. Results
  16. ...and 5 more sections

Figures (2)

  • Figure 1: The framework of Ada-FCN for fMRI-based brain disorder classification.
  • Figure 2: Distinct frequency-coupled connectivity patterns revealed by group-averaged $A_{\text{unified}}$ matrices for CN, MCI, and AD.