Evaluating Graph Signal Processing for Neuroimaging Through Classification and Dimensionality Reduction

TL;DR

This paper assesses the effectiveness of Graph Signal Processing techniques in neuroimaging, demonstrating that graph-based dimensionality reduction, especially mixed graphs, outperforms classical methods like PCA and ICA in decoding brain activity.

Contribution

It introduces and compares seven graph-based models for neuroimaging data, highlighting the superior performance of mixed graphs in classification tasks.

Findings

Mixed graphs yield the best classification accuracy.

Graph sampling surpasses PCA and ICA in dimension reduction.

Graph-based methods effectively decode brain activity.

Abstract

Graph Signal Processing (GSP) is a promising framework to analyze multi-dimensional neuroimaging datasets, while taking into account both the spatial and functional dependencies between brain signals. In the present work, we apply dimensionality reduction techniques based on graph representations of the brain to decode brain activity from real and simulated fMRI datasets. We introduce seven graphs obtained from a) geometric structure and/or b) functional connectivity between brain areas at rest, and compare them when performing dimension reduction for classification. We show that mixed graphs using both a) and b) offer the best performance. We also show that graph sampling methods perform better than classical dimension reduction including Principal Component Analysis (PCA) and Independent Component Analysis (ICA).