A Privacy-Preserving Domain Adversarial Federated learning for multi-site brain functional connectivity analysis

2025-02-03

AI Summary

To address privacy preservation and poor model generalizability—stemming from non-IID data—in multi-site resting-state fMRI (rs-fMRI) collaborative analysis, as well as weak cross-site interpretability, this paper proposes DAFed, a domain-adversarial federated learning framework. DAFed introduces the first feature-decoupled domain-adversarial architecture, jointly optimizing domain-invariant and domain-specific representations, adversarial transfer, and contrastive learning; it supports semi-supervised training with both labeled and unlabeled data and integrates enhanced Score-CAM for interpretable neurobiomarker discovery. On autism spectrum disorder (ASD) diagnosis, DAFed significantly outperforms state-of-the-art methods. It demonstrates robust cross-disorder generalization to Alzheimer’s disease (AD) and mild cognitive impairment (MCI) classification. Moreover, it successfully identifies cross-site consistent dysfunctional brain regions and aberrant functional connectivity patterns. DAFed establishes a novel paradigm for privacy-preserving, generalizable, and interpretable multi-center brain disorder modeling.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Multi-modal Vision
• Application Domains: Humanities & Computational Social Science

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Resting-state functional magnetic resonance imaging (rs-fMRI) and its derived functional connectivity networks (FCNs) have become critical for understanding neurological disorders. However, collaborative analyses and the generalizability of models still face significant challenges due to privacy regulations and the non-IID (non-independent and identically distributed) property of multiple data sources. To mitigate these difficulties, we propose Domain Adversarial Federated Learning (DAFed), a novel federated deep learning framework specifically designed for non-IID fMRI data analysis in multi-site settings. DAFed addresses these challenges through feature disentanglement, decomposing the latent feature space into domain-invariant and domain-specific components, to ensure robust global learning while preserving local data specificity. Furthermore, adversarial training facilitates effective knowledge transfer between labeled and unlabeled datasets, while a contrastive learning module enhances the global representation of domain-invariant features. We evaluated DAFed on the diagnosis of ASD and further validated its generalizability in the classification of AD, demonstrating its superior classification accuracy compared to state-of-the-art methods. Additionally, an enhanced Score-CAM module identifies key brain regions and functional connectivity significantly associated with ASD and MCI, respectively, uncovering shared neurobiological patterns across sites. These findings highlight the potential of DAFed to advance multi-site collaborative research in neuroimaging while protecting data confidentiality.

Problem

• Develops privacy-preserving fMRI data analysis.
• Addresses non-IID data challenges in neuroimaging.
• Enhances multi-site collaborative model generalizability.

Innovation

• Domain Adversarial Federated Learning framework
• Feature disentanglement for domain-invariant analysis
• Contrastive learning enhances global representation