Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Identifying Influential nodes in Brain Networks via Self-Supervised Graph-Transformer

Yanqing Kang, Di Zhu, Haiyang Zhang, Enze Shi, Sigang Yu, Jinru Wu, Xuhui Wang, Xuan Liu, Geng Chen, Xi Jiang, Tuo Zhang, Shu Zhang

TL;DR

A Self-Supervised Graph Reconstruction framework based on Graph-Transformer (SSGR-GT) to identify I-nodes, which has three main characteristics, which enhance the understanding of the I-nodes within the brain network, and provide new insights for future research in further understanding the brain working mechanisms.

Abstract

Studying influential nodes (I-nodes) in brain networks is of great significance in the field of brain imaging. Most existing studies consider brain connectivity hubs as I-nodes. However, this approach relies heavily on prior knowledge from graph theory, which may overlook the intrinsic characteristics of the brain network, especially when its architecture is not fully understood. In contrast, self-supervised deep learning can learn meaningful representations directly from the data. This approach enables the exploration of I-nodes for brain networks, which is also lacking in current studies. This paper proposes a Self-Supervised Graph Reconstruction framework based on Graph-Transformer (SSGR-GT) to identify I-nodes, which has three main characteristics. First, as a self-supervised model, SSGR-GT extracts the importance of brain nodes to the reconstruction. Second, SSGR-GT uses Graph-Transformer, which is well-suited for extracting features from brain graphs, combining both local and global characteristics. Third, multimodal analysis of I-nodes uses graph-based fusion technology, combining functional and structural brain information. The I-nodes we obtained are distributed in critical areas such as the superior frontal lobe, lateral parietal lobe, and lateral occipital lobe, with a total of 56 identified across different experiments. These I-nodes are involved in more brain networks than other regions, have longer fiber connections, and occupy more central positions in structural connectivity. They also exhibit strong connectivity and high node efficiency in both functional and structural networks. Furthermore, there is a significant overlap between the I-nodes and both the structural and functional rich-club. These findings enhance our understanding of the I-nodes within the brain network, and provide new insights for future research in further understanding the brain working mechanisms.

Identifying Influential nodes in Brain Networks via Self-Supervised Graph-Transformer

TL;DR

A Self-Supervised Graph Reconstruction framework based on Graph-Transformer (SSGR-GT) to identify I-nodes, which has three main characteristics, which enhance the understanding of the I-nodes within the brain network, and provide new insights for future research in further understanding the brain working mechanisms.

Abstract

Studying influential nodes (I-nodes) in brain networks is of great significance in the field of brain imaging. Most existing studies consider brain connectivity hubs as I-nodes. However, this approach relies heavily on prior knowledge from graph theory, which may overlook the intrinsic characteristics of the brain network, especially when its architecture is not fully understood. In contrast, self-supervised deep learning can learn meaningful representations directly from the data. This approach enables the exploration of I-nodes for brain networks, which is also lacking in current studies. This paper proposes a Self-Supervised Graph Reconstruction framework based on Graph-Transformer (SSGR-GT) to identify I-nodes, which has three main characteristics. First, as a self-supervised model, SSGR-GT extracts the importance of brain nodes to the reconstruction. Second, SSGR-GT uses Graph-Transformer, which is well-suited for extracting features from brain graphs, combining both local and global characteristics. Third, multimodal analysis of I-nodes uses graph-based fusion technology, combining functional and structural brain information. The I-nodes we obtained are distributed in critical areas such as the superior frontal lobe, lateral parietal lobe, and lateral occipital lobe, with a total of 56 identified across different experiments. These I-nodes are involved in more brain networks than other regions, have longer fiber connections, and occupy more central positions in structural connectivity. They also exhibit strong connectivity and high node efficiency in both functional and structural networks. Furthermore, there is a significant overlap between the I-nodes and both the structural and functional rich-club. These findings enhance our understanding of the I-nodes within the brain network, and provide new insights for future research in further understanding the brain working mechanisms.
Paper Structure (22 sections, 12 figures, 1 table)

This paper contains 22 sections, 12 figures, 1 table.

Table of Contents

  1. INTRODUCTION
  2. METHOD
  3. Overview
  4. Data and Preprocessing
  5. Self-supervised Graph Reconstruct Framework based on Graph-Transformer (SSGR-GT Framework)
  6. Graph Generation
  7. GR-GT Module
  8. Ranking ROIs via MScore Pool
  9. Evaluating the Hub Attributes of the High-Score ROIs
  10. Analyzing Foundational Elements for Coordinated Structure and Multi-tasks
  11. Evaluating Reproducibility of I-nodes on New Brain Atlas
  12. EXPERIMENTS AND RESULTS
  13. Experimental Setup and Performance
  14. Experimental Setup
  15. Model Performance
  16. ...and 7 more sections

Figures (12)

  • Figure 1: The pipeline of the proposed SSGR-GT framework. (A) represents the graph generation of brain. (B) represents the GR-GT module used to reconstruct the graph and obtain node scores. The module consists of three parts, including the Encoder module, MScore Pool, and Decoder module.
  • Figure 2: The detail of the GR-GT module. (A) represents the network architecture of the module. (B) represents the data representation in the module process.
  • Figure 3: Model Performance for the RS. (A) Model reconstruction quality. (B) Comparison of the original and reconstructed matrices for a randomly selected individual. (C) Changes in scores for selected ROIs and their visualization across brain regions.
  • Figure 5: Group-level scores of ROIs. (A) represents the histogram of inter-individual ROIs scores similarity of eight sets of experiments. (B) represents the scores of all ROIs and the division of ROIs at RS and S.
  • Figure 6: The Scale-1 ROIs in RS and S.
  • ...and 7 more figures