atp-bme / lgmf-gnn

Code of paper "An Objective Quantitative Diagnosis of Depression Using Local-to-global Multi-modal Fusion Graph Neural Network"

LGMF-GNN

Code of paper "An Objective Quantitative Diagnosis of Depression Using Local-to-global Multi-modal Fusion Graph Neural Network"

Overview

Content includes source code for the implementation of local-to-global multi-modal fusion graph neural network on large multi-site MDD fMRI datasets for major depressive disorder diagnosis.

Requirements

• python >= 3.8
• torch = 1.13.0
• torch-cluster = 1.6.1+pt113cu117
• torch-geometric = 2.3.1
• torch-scatter = 2.1.1+pt113cu117
• torch-sparse = 0.6.17+pt113cu117

datasets

• SRPBS dataset
• REST-meta-MDD dataset
• OpenNeuro dataset
• Anding dataset

Toolboxes

• DPARSF
• NeuroCombat-sklearn

lgmf-gnn's Issues

paper

Thank you very much for the code! But I didn't find your paper, can you provide the paper, thanks!