Spatial-Temporal Graph Convolutional Network for Insomnia Classification via Brain Functional Connectivity Imaging of rs-fMRI

Chronic Insomnia Disorder (CID) is a prevalent sleep disorder characterized by persistent difficulties in initiating or maintaining sleep, leading to significant impairment in daily functioning and quality of life. Accurate classification of CID patients is crucial for effective treatment and personalized care. However, existing approaches face challenges in capturing the complex spatio-temporal patterns inherent in rs-fMRI data, limiting their classification performance. In this study, we propose a novel approach utilizing the Spatial-Temporal Graph Convolutional Network (ST-GCN) for classification of CID patients. Our method aims to address the limitations of existing approaches by leveraging the graph convolutional framework to model the spatio-temporal dynamics in rs-fMRI data. Specifically, this method first pre-processes the raw rs-fMRI images and divides the brain into several regions of interest using a brain template. Next, it utilizes the ST-GCN network to integrate spatio-temporal features. Finally, the extracted features are utilized into a fully connected network for classification. Comparative experiment results show that the Accuracy and Specificity of the proposed method reach 98.90%, 99.08% respectively, which are better than the state-of-the-art methods.