Related Experiment Video
Updated: Jun 7, 2025
05:19
Author Spotlight: Therapeutic Benefit of Closed-Loop Deep Brain Stimulation in Depression Treatment
Published on: July 7, 2023
2.2K
An adaptive multi-graph neural network with multimodal feature fusion learning for MDD detection.
Tao Xing1,2, Yutao Dou2, Xianliang Chen3
1College of Computer Science and Engineering, Guilin University of Technology, Guilin, 541006, China.
Scientific Reports
|November 17, 2024
Summary
This study introduces EMO-GCN, a novel multimodal approach for detecting Major Depressive Disorder (MDD) using electroencephalograms and audio data. The method achieves high accuracy, offering improved insights into mental health diagnostics.
Area of Science:
- Neuroscience
- Computer Science
- Psychiatry
Background:
- Major Depressive Disorder (MDD) significantly impacts quality of life and increases suicide risk.
- Current MDD detection methods often rely on single data types or fail to integrate multimodal information effectively.
- Overlooking inter-modal relationships can limit the depth of depression detection.
Purpose of the Study:
- To develop an advanced multimodal depression detection system.
- To effectively leverage both electroencephalograms and patient interview audios for MDD identification.
- To address limitations in existing methods by better utilizing diverse data modalities.
Main Methods:
- Proposed EMO-GCN, an adaptive multi-graph neural network for multimodal analysis.
- Utilized graph-based techniques to model and extract features from electroencephalogram and audio data.
- Investigated correlations and differences between various data modalities.
Main Results:
- Achieved outstanding performance on the MODMA dataset with 96.30% accuracy (ACC).
- Ablation studies validated the efficacy of individual EMO-GCN components.
- Demonstrated the potential of graph-based multimodal analysis in mental health.
Conclusions:
- EMO-GCN offers a promising approach for accurate MDD detection.
- Graph-based multimodal analysis provides new avenues for mental health research.
- The findings highlight the importance of integrating diverse data sources for improved diagnostic capabilities.
