English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Brain functional connectivity (FC) reveals biomarkers for identification ofvarious neuropsychiatric disorders. Recent application of deep neural networks(DNNs) to connectome-based classification mostly relies on traditionalconvolutional neural networks using input connectivity matrices on a regularEuclidean grid. We propose a graph deep learning framework to incorporate thenon-Euclidean information about graph structure for classifying functionalmagnetic resonance imaging (fMRI)-derived brain networks in major depressivedisorder (MDD). We design a novel graph autoencoder (GAE) architecture based onthe graph convolutional networks (GCNs) to embed the topological structure andnode content of large-sized fMRI networks into low-dimensional latentrepresentations. In network construction, we employ the Ledoit-Wolf (LDW)shrinkage method to estimate the high-dimensional FC metrics efficiently fromfMRI data. We consider both supervised and unsupervised approaches for thegraph embedding learning. The learned embeddings are then used as featureinputs for a deep fully-connected neural network (FCNN) to discriminate MDDfrom healthy controls. Evaluated on two resting-state fMRI (rs-fMRI) MDDdatasets, results show that the proposed GAE-FCNN model significantlyoutperforms several state-of-the-art methods for brain connectomeclassification, achieving the best accuracy using the LDW-FC edges as nodefeatures. The graph embeddings of fMRI FC networks learned by the GAE alsoreveal apparent group differences between MDD and HC. Our new frameworkdemonstrates feasibility of learning graph embeddings on brain networks toprovide discriminative information for diagnosis of brain disorders.

Graph Autoencoders for Embedding Learning in Brain Networks and Major Depressive Disorder Identification