BrainOOD: Out-of-distribution Generalizable Brain Network Analysis

2025-02-02

AI Summary

Multi-site brain network data suffer from distributional shifts, leading to poor out-of-distribution (OOD) generalization and limited interpretability of disease-relevant brain regions. Method: We propose the first OOD-generalizable analytical framework for brain networks, featuring a dual-module graph neural network (GNN) architecture that jointly performs feature selection and structural extraction. We introduce an enhanced Graph Information Bottleneck (GIB) to recover causal subgraphs and employ multi-auxiliary loss optimization. Contribution/Results: We establish the first brain network OOD benchmark; our method achieves state-of-the-art performance across 16 baselines, with up to 8.5% improvement in OOD generalization. Identified brain region patterns align with established neuroscientific knowledge, significantly enhancing both model interpretability and clinical relevance.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Humans and AI: Brain-Sensing and Analysis
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

In neuroscience, identifying distinct patterns linked to neurological disorders, such as Alzheimer's and Autism, is critical for early diagnosis and effective intervention. Graph Neural Networks (GNNs) have shown promising in analyzing brain networks, but there are two major challenges in using GNNs: (1) distribution shifts in multi-site brain network data, leading to poor Out-of-Distribution (OOD) generalization, and (2) limited interpretability in identifying key brain regions critical to neurological disorders. Existing graph OOD methods, while effective in other domains, struggle with the unique characteristics of brain networks. To bridge these gaps, we introduce BrainOOD, a novel framework tailored for brain networks that enhances GNNs' OOD generalization and interpretability. BrainOOD framework consists of a feature selector and a structure extractor, which incorporates various auxiliary losses including an improved Graph Information Bottleneck (GIB) objective to recover causal subgraphs. By aligning structure selection across brain networks and filtering noisy features, BrainOOD offers reliable interpretations of critical brain regions. Our approach outperforms 16 existing methods and improves generalization to OOD subjects by up to 8.5%. Case studies highlight the scientific validity of the patterns extracted, which aligns with the findings in known neuroscience literature. We also propose the first OOD brain network benchmark, which provides a foundation for future research in this field. Our code is available at https://github.com/AngusMonroe/BrainOOD.

Problem

• Enhances OOD generalization in brain network analysis
• Improves interpretability of key brain regions
• Addresses distribution shifts in multi-site data

Innovation

• Graph Neural Networks enhancement
• Improved Graph Information Bottleneck
• Feature and structure extractor