Intelligent diagnosis of major depressive disorder with edge convolution and contrastive learning

we propose a novel deep learning framework, Graph Contrastive Learning based on Edge Convolution (EC-GCL), to analyze resting-state fMRI data from 1,160 participants, including 597 patients with MDD and 563 healthy controls. The network integrates a learnable graph augmentation in adversarial graph contrastive learning with edge convolution encoder to enhance feature extraction and classification accuracy.