Related Experiment Video
Updated: Jun 29, 2026
06:35
Basics of Multivariate Analysis in Neuroimaging Data
Published on: July 24, 2010
16.8K
TGNet: tensor-based graph convolutional networks for multimodal brain network analysis
Zhaoming Kong1, Rong Zhou2, Xinwei Luo2
1School of Software Engineering, South China University of Technology, 382 Waihuan Dong Road, Guangzhou, 510006, China.
Biodata Mining
|December 5, 2024
Summary
This study introduces TGNet, a novel framework for multimodal brain network analysis. TGNet effectively classifies neurological disorders, outperforming existing methods, especially with limited data.
Area of Science:
- Neuroscience
- Artificial Intelligence
- Medical Imaging
Background:
- Multimodal brain network analysis is crucial for understanding neurological disorders.
- Current methods face challenges in modeling complex multimodal brain network structures.
Purpose of the Study:
- To propose a novel tensor-based graph convolutional network (TGNet) framework.
- To effectively model homogeneity and intricate structures in multimodal brain networks.
Main Methods:
- Developed a tensor-based graph convolutional network (TGNet) framework.
- Combined tensor decomposition with multi-layer GCNs.
- Evaluated TGNet on HIV, Bipolar Disorder, PPMI, and ADNI datasets.
Main Results:
- TGNet significantly outperforms existing methods in disease classification.
- Demonstrated superior performance, particularly with limited sample sizes.
- Showcased robustness and effectiveness in multimodal brain network analysis.
Conclusions:
- TGNet offers a powerful approach for multimodal brain network analysis.
- The framework has potential for advancing the diagnosis and understanding of neurological disorders.
- TGNet shows promise for applications in clinical settings and research.
