Multi-Atlas Brain Network Classification through Consistency Distillation and Complementary Information Fusion

2024-09-28

IEEE journal of biomedical and health informatics

AI Summary

To address cross-atlas inconsistency and ROI-level information fragmentation arising from multi-atlas heterogeneity in rs-fMRI brain network classification, this paper proposes a novel framework. Methodologically: (1) a decoupled Transformer architecture enables cross-atlas consistent knowledge distillation; (2) a master-group dual consistency constraint enforces collaborative alignment across atlases; and (3) an ROI-level cross-atlas message-passing mechanism integrates complementary regional representations. Evaluated on four neuropsychiatric disorder datasets, the method significantly outperforms state-of-the-art approaches in classification accuracy, achieves computational efficiency, and yields neuroscientifically plausible patterns aligned with prior domain knowledge. This work is the first to systematically resolve two core challenges in multi-atlas brain network modeling: (i) establishing consistent cross-atlas representation learning and (ii) enabling fine-grained, ROI-level inter-atlas information interaction.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Responsible Web: Consent frameworks and practices on the web

Abstract

Brain network analysis plays a crucial role in identifying distinctive patterns associated with neurological disorders. Functional magnetic resonance imaging (fMRI) enables the construction of brain networks by analyzing correlations in blood-oxygen-level-dependent (BOLD) signals across different brain regions, known as regions of interest (ROIs). These networks are typically constructed using atlases that parcellate the brain based on various hypotheses of functional and anatomical divisions. However, there is no standard atlas for brain network classification, leading to limitations in detecting abnormalities in disorders. Recent methods leveraging multiple atlases fail to ensure consistency across atlases and lack effective ROI-level information exchange, limiting their efficacy. To address these challenges, we propose the Atlas-Integrated Distillation and Fusion network (AIDFusion), a novel framework designed to enhance brain network classification using fMRI data. AIDFusion introduces a disentangle Transformer to filter out inconsistent atlas-specific information and distill meaningful cross-atlas connections. Additionally, it enforces subject- and population-level consistency constraints to improve cross-atlas coherence. To further enhance feature integration, AIDFusion incorporates an inter-atlas message-passing mechanism that facilitates the fusion of complementary information across brain regions. We evaluate AIDFusion on four resting-state fMRI datasets encompassing different neurological disorders. Experimental results demonstrate its superior classification performance and computational efficiency compared to state-of-the-art methods. Furthermore, a case study highlights AIDFusion's ability to extract interpretable patterns that align with established neuroscience findings, reinforcing its potential as a robust tool for multi-atlas brain network analysis. The code is publicly available at https://github.com/AngusMonroe/AIDFusion.

Problem

• Classifying brain networks using multiple atlases with consistency distillation
• Filtering inconsistent atlas-specific information via disentangle Transformer
• Fusing complementary inter-atlas information through message-passing mechanism

Innovation

• Uses disentangle Transformer to filter inconsistent atlas information
• Applies cross-atlas consistency constraints at multiple levels
• Implements inter-atlas message-passing for complementary information fusion