Using Graph Convolutional Networks to Address fMRI Small Data Problems

TL;DR

This paper introduces a graph convolutional network approach for small data neuroimaging tasks, demonstrating a 12% accuracy improvement by leveraging spectral representations and data smoothing techniques.

Contribution

It presents a novel application of spectral graph convolutional networks to small data neuroimaging problems, improving performance over traditional methods.

Findings

Approximately 12% accuracy improvement over traditional methods.

Spectral representation enables efficient information propagation.

Data smoothing via triangle inequalities enhances model performance.

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.