Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Graph Neural Networks for Brain Graph Learning: A Survey

Xuexiong Luo, Jia Wu, Jian Yang, Shan Xue, Amin Beheshti, Quan Z. Sheng, David McAlpine, Paul Sowman, Alexis Giral, Philip S. Yu

TL;DR

This survey compiles and organizes work on learning brain representations from graph‑structured brain data with Graph Neural Networks. It categorizes approaches into static, dynamic, and multi‑modal brain graphs, each paired with prediction and interpretation objectives, and catalogs representative methods, datasets, and implementations. The paper highlights gaps in graph construction quality, multi‑scale fusion, and integration of prior neuroscience knowledge, and outlines concrete directions such as subgraph extraction and diffusion‑based augmentation. By providing a structured overview and practical resources, it aims to accelerate brain disorder analysis through reliable, interpretable brain graph learning.

Abstract

Exploring the complex structure of the human brain is crucial for understanding its functionality and diagnosing brain disorders. Thanks to advancements in neuroimaging technology, a novel approach has emerged that involves modeling the human brain as a graph-structured pattern, with different brain regions represented as nodes and the functional relationships among these regions as edges. Moreover, graph neural networks (GNNs) have demonstrated a significant advantage in mining graph-structured data. Developing GNNs to learn brain graph representations for brain disorder analysis has recently gained increasing attention. However, there is a lack of systematic survey work summarizing current research methods in this domain. In this paper, we aim to bridge this gap by reviewing brain graph learning works that utilize GNNs. We first introduce the process of brain graph modeling based on common neuroimaging data. Subsequently, we systematically categorize current works based on the type of brain graph generated and the targeted research problems. To make this research accessible to a broader range of interested researchers, we provide an overview of representative methods and commonly used datasets, along with their implementation sources. Finally, we present our insights on future research directions. The repository of this survey is available at \url{https://github.com/XuexiongLuoMQ/Awesome-Brain-Graph-Learning-with-GNNs}.

Graph Neural Networks for Brain Graph Learning: A Survey

TL;DR

This survey compiles and organizes work on learning brain representations from graph‑structured brain data with Graph Neural Networks. It categorizes approaches into static, dynamic, and multi‑modal brain graphs, each paired with prediction and interpretation objectives, and catalogs representative methods, datasets, and implementations. The paper highlights gaps in graph construction quality, multi‑scale fusion, and integration of prior neuroscience knowledge, and outlines concrete directions such as subgraph extraction and diffusion‑based augmentation. By providing a structured overview and practical resources, it aims to accelerate brain disorder analysis through reliable, interpretable brain graph learning.

Abstract

Exploring the complex structure of the human brain is crucial for understanding its functionality and diagnosing brain disorders. Thanks to advancements in neuroimaging technology, a novel approach has emerged that involves modeling the human brain as a graph-structured pattern, with different brain regions represented as nodes and the functional relationships among these regions as edges. Moreover, graph neural networks (GNNs) have demonstrated a significant advantage in mining graph-structured data. Developing GNNs to learn brain graph representations for brain disorder analysis has recently gained increasing attention. However, there is a lack of systematic survey work summarizing current research methods in this domain. In this paper, we aim to bridge this gap by reviewing brain graph learning works that utilize GNNs. We first introduce the process of brain graph modeling based on common neuroimaging data. Subsequently, we systematically categorize current works based on the type of brain graph generated and the targeted research problems. To make this research accessible to a broader range of interested researchers, we provide an overview of representative methods and commonly used datasets, along with their implementation sources. Finally, we present our insights on future research directions. The repository of this survey is available at \url{https://github.com/XuexiongLuoMQ/Awesome-Brain-Graph-Learning-with-GNNs}.
Paper Structure (19 sections, 2 equations, 2 figures, 2 tables)

This paper contains 19 sections, 2 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Graph Definitions
  4. Graph Neural Networks
  5. Brain Graph Construction
  6. Research Problems
  7. Taxonomy of Brain Graph Learning Approaches
  8. Static Brain Graph Learning (SBGL)
  9. Dynamic Brain Graph Leaning (DBGL)
  10. Multi-Modal Brain Graph Learning (MBGL)
  11. Brain Graph Datasets
  12. Future Research Directions
  13. Reliable Brain Graph Construction
  14. Multi-Scale Brain Graph Fusion
  15. Prior Domain Knowledge Fusion
  16. ...and 4 more sections

Figures (2)

  • Figure 1: An illustration of brain graph learning framework with GNNs, where different types of neuroimaging are preprocessed to generate corresponding connectivity matrixes. Then, brain graphs are fed into GNNs to learn brain graph representations for disorder analysis.
  • Figure 2: A toy example of different types of brain graph generated.