Similarity Learning with Higher-Order Proximity for Brain Network Analysis

Recently, learning a similarity metric has gained much attention, where the goal is to learn a function that maps input patterns to a target space while preserving the semantic distance in the input space. While most related work focused on images, in this work, we focus instead on learning a similarity metric for neuroimages, such as fMRI images and DTI images. Functional magnetic resonance imaging (fMRI) is widely used in cognitive neuroscience, medical, and clinical applications. In this work, we focus on the similarity learning for fMRI brain network analysis. We present a framework called Higher-order Siamese GCN for similarity learning on graphs extracted from fMRI data. Our proposed framework captures the higher-order structure of the data, which leads to more accurate results compared to baseline methods. To the best of our knowledge, this is the first community-preserving approach for brain network analysis. We evaluate the proposed Higher-order Siamese GCN framework on four real fMRI brain network datasets for similarity learning with respect to brain health status and cognitive abilities. Our proposed method achieves an average AUC gain of 82.6% compared to PCA, and an average AUC gain of 42% compared to S-GCN across a variety of datasets, indicating its promising learning ability for clinical investigation and brain disease diagnosis.