Identifying resting‐state effective connectivity abnormalities in drug‐naïve major depressive disorder diagnosis via graph convolutional networks

Abstract

Major depressive disorder (MDD) is a leading cause of disability; its symptoms interfere with social, occupational, interpersonal, and academic functioning. However, the diagnosis of MDD is still made by phenomenological approach. The advent of neuroimaging techniques allowed numerous studies to use resting‐state functional magnetic resonance imaging (rs‐fMRI) and estimate functional connectivity for brain‐disease identification. Recently, attempts have been made to investigate effective connectivity (EC) that represents causal relations among regions of interest. In the meantime, to identify meaningful phenotypes for clinical diagnosis, graph‐based approaches such as graph convolutional networks (GCNs) have been leveraged recently to explore complex pairwise similarities in imaging/nonimaging features among subjects. In this study, we validate the use of EC for MDD identification by estimating its measures via a group sparse representation along with a structured equation modeling approach in a whole‐brain data‐driven manner from rs‐fMRI. To distinguish drug‐naïve MDD patients from healthy controls, we utilize spectral GCNs based on a population graph to successfully integrate EC and nonimaging phenotypic information. Furthermore, we devise a novel sensitivity analysis method to investigate the discriminant connections for MDD identification in our trained GCNs. Our experimental results validated the effectiveness of our method in various scenarios, and we identified altered connectivities associated with the diagnosis of MDD.

Using graph convolutional network to characterize individuals with major depressive disorder across multiple imaging sites

“…Although one previous study has demonstrated the capability of GCN for distinguishing individuals with MDD from HC, 39 our work has significant advantages. First, the previous effort only included 29 individuals with MDD and 44 HC.…”

Graph Convolutional Networks Reveal Network-Level Functional Dysconnectivity in Schizophrenia

“…Motivated by convolutional neural network (CNN), GCN was designed to perform convolution operation on graph structure to aggregate local and neighboring information to generate new feature maps. The GCN has been successfully applied in previous publications to characterize autism spectrum disorder, 30 Alzheimer’s disease, 31 depression, 32 and sex. 33 However, most current application of GCN is limited to small dataset, and the model performance could be potentially unreliable.…”

Application of Artificial Intelligence in the MRI Classification Task of Human Brain Neurological and Psychiatric Diseases: A Scoping Review

“…The discriminator and generator of the proposed GAN model both have four fully-connected layers. Jun et al [111] used spectral GCNs based on a population graph to successfully integrate effective connectivity (EC) and non-imaging phenotypic information. The above papers on the application of AI methods in MDD diagnosis are summarized in Table 4.…”