Markov Chain-Guided Graph Construction and Sampling Depth Optimization for EEG-Based Mental Disorder Detection

Yihan Wu,Tao Chang,Peng Xu,Yangsong Zhang

2023-09-19

Abstract:Graph Neural Networks (GNNs) have received considerable attention since its introduction. It has been widely applied in various fields due to its ability to represent graph structured data. However, the application of GNNs is constrained by two main issues. Firstly, the "over-smoothing" problem restricts the use of deeper network structures. Secondly, GNNs' applicability is greatly limited when nodes and edges are not clearly defined and expressed, as is the case with EEG <a class="link-external link-http" href="http://data.In" rel="external noopener nofollow">this http URL</a> this study, we proposed an innovative approach that harnesses the distinctive properties of the graph structure's Markov Chain to optimize the sampling depth of deep graph convolution networks. We introduced a tailored method for constructing graph structures specifically designed for analyzing EEG data, alongside the development of a vertex-level GNN classification model for precise detection of mental disorders. In order to verify the method's performance, we conduct experiments on two disease datasets using a subject-independent experiment scenario. For the Schizophrenia (SZ) data, our method achieves an average accuracy of 100% using only the first 300 seconds of data from each subject. Similarly, for Major Depressive Disorder (MDD) data, the method yields average accuracies of over 99%. These experiments demonstrate the method's ability to effectively distinguish between healthy control (HC) subjects and patients with mental disorders. We believe this method shows great promise for clinical diagnosis.

Neurons and Cognition

What problem does this paper attempt to address?

The paper primarily aims to address the following issues: ### Research Background and Objectives - **Challenges in Diagnosing Mental Disorders**: Mental disorders (such as Schizophrenia (SZ) and Major Depressive Disorder (MDD)) affect a large global population. However, due to the lack of significant biomarkers, clinical diagnosis still relies on the experience of doctors rather than objective indicators. - **Application of Electrophysiological Signals**: Electrophysiological signal technology (such as EEG) has significant value in studying brain neural activity due to its high temporal resolution, non-invasiveness, low cost, and portability. - **Limitations of Existing Methods**: - Methods based on feature engineering highly depend on the experience and prior knowledge of researchers. - Although deep learning methods can automatically extract features, they suffer from the over-smoothing problem, limiting the use of deep network structures. ### Problems Addressed - **Proposing an Innovative Graph Construction Method**: For EEG data, constructing a graph structure to overcome the problem of unclear definitions of nodes and edges. - **Optimizing Deep Graph Convolutional Networks**: Using Markov chain theory to guide the optimization of the sampling depth of the graph structure, solving the over-smoothing problem, and making deep graph convolutional networks more effective in feature extraction. - **Detection of Mental Disorders**: Developing a vertex-level graph neural network classification model to achieve accurate detection of schizophrenia and major depressive disorder. ### Main Contributions - Proposed a new method for organizing EEG signal graphs for the detection of mental disorders. - Developed a high-precision deep graph convolutional method to classify graph data. - Proposed a method to optimize the graph structure and its corresponding sampling depth for deep graph convolutional networks. ### Experimental Results - For the schizophrenia dataset, using only the first 300 seconds of data from each subject, the method achieved an average accuracy of 100%. - For the major depressive disorder dataset, the method's average accuracy exceeded 99%. These results demonstrate the great potential of this method in clinical diagnosis.

Markov Chain-Guided Graph Construction and Sampling Depth Optimization for EEG-Based Mental Disorder Detection

Assisting Schizophrenia Diagnosis Using Clinical Electroencephalography and Interpretable Graph Neural Networks: a Real-World and Cross-Site Study

A Hybrid Graph Neural Network for Enhanced EEG-Based Depression Detection

MGSN: Depression EEG lightweight detection based on multiscale DGCN and SNN for multichannel topology

A novel EEG-based graph convolution network for depression detection: Incorporating secondary subject partitioning and attention mechanism

Graph-based EEG approach for depression prediction: integrating time-frequency complexity and spatial topology

Automatic recognition of schizophrenia from brain-network features using graph convolutional neural network

SLG-NET: Subgraph Neural Network with Local-Global Braingraph Feature Extraction Modules and a Novel Subgraph Generation Algorithm for Automated Identification of Major Depressive Disorder

EEG based depression recognition using improved graph convolutional neural network

GCTNet: a graph convolutional transformer network for major depressive disorder detection based on EEG signals

Multi-view graph network learning framework for identification of major depressive disorder

Graph neural networks in EEG spike detection

Using Dynamic Graph Convolutional Network to Identify Individuals with Major Depression Disorder

Interpretable Graph Neural Networks for Connectome-Based Brain Disorder Analysis

Spatial Temporal Graph Convolution with Graph Structure Self-learning for Early MCI Detection

Parkinson's Disease Detection from Resting State EEG using Multi-Head Graph Structure Learning with Gradient Weighted Graph Attention Explanations

Automated Rest EEG-Based Diagnosis of Depression and Schizophrenia Using a Deep Convolutional Neural Network

Joint Embedding of Structural and Functional Brain Networks with Graph Neural Networks for Mental Illness Diagnosis

Balanced Graph Structure Information for Brain Disease Detection

Classify EEG and Reveal Latent Graph Structure with Spatio-Temporal Graph Convolutional Neural Network

High‐Order line graphs of fMRI data in major depressive disorder