Using Graph Convolutional Networks to Address fMRI Small Data Problems

2025-02-19

AI Summary

To address poor generalizability in fMRI-based prognostic tasks—such as treatment response prediction—under small-sample regimes caused by data scarcity, this paper proposes a spectral-domain graph convolutional few-shot learning framework. It introduces, for the first time, spectral graph theory–driven graph convolution into fMRI analysis, explicitly modeling functional brain connectivity as a graph structure. To enhance data smoothness within the functional connectivity space while preserving neurobiological interpretability and model robustness, we enforce metric closure via the triangle inequality. Compared to conventional deep learning approaches, our method achieves approximately 12% higher prediction accuracy on identical small-sample fMRI datasets, markedly mitigating overfitting. This work establishes a novel paradigm for few-shot neuroimaging prognostics, combining methodological innovation—particularly in spectral graph representation—with strong potential for clinical translation.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Although great advances in the analysis of neuroimaging data have been made, a major challenge is a lack of training data. This is less problematic in tasks such as diagnosis, where much data exists, but particularly prevalent in harder problems such as predicting treatment responses (prognosis), where data is focused and hence limited. Here, we address the learning from small data problems for medical imaging using graph neural networks. This is particularly challenging as the information about the patients is themselves graphs (regions of interest connectivity graphs). We show how a spectral representation of the connectivity data allows for efficient propagation that can yield approximately 12% improvement over traditional deep learning methods using the exact same data. We show that our method's superior performance is due to a data smoothing result that can be measured by closing the number of triangle inequalities and thereby satisfying transitivity.

Problem

• Graph Convolutional Networks for fMRI
• Small data learning challenges
• Improving treatment response prediction

Innovation

• Graph Convolutional Networks
• Spectral representation
• Data smoothing technique