Electroencephalography decoding model based on fusion deep graph convolutional neural network for spinal cord injury

• Published in: Healthcare and Rehabilitation Vol. 1; no. 3; p. 100039
• Main Authors: Lou, Tianwei, Zhang, Xinting, Jiang, Lei, Chen, Lei, Gao, Licai, Lun, Zhixiao, Li, Jincheng, Zhang, Yang, Xu, Fangzhou, Jung, Tzyy-Ping
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.07.2025
• Subjects: Combined graph convolution network; Electroencephalogram; Fusion wavelet; Motor imagery; Spinal cord injury; CNN; F-GCN; Electroencephalogram; ERS; Fusion wavelet; LDA; SVM; Spinal cord injury; Combined graph convolution network; RNN; GFT; MI; EC; EMG; PSD; CSP; CWT; EEG; C2CM; BCI; EO; Motor imagery; coif4; tSCI; AR; GCN; ERD; db4; SCI; CapsNet; sym4; LOSO-CV
• ISSN: 3050-6131
• DOI: 10.1016/j.hcr.2025.100039

Electroencephalography (EEG) signals can be used to measure neuronal activity in different regions of the brain through electrodes. To enhance the decoding of motor imagery (MI) EEG signals in spinal cord injury (SCI) patients, this study proposes a feature fusion graph convolutional neural network (F-GCN) model that integrates wavelet-based time-frequency features and functional topological relationships among EEG electrodes, aiming to improve classification accuracy and provide guidance for rehabilitation. This study included 10 patients with spinal cord injuries as the experimental group, and 10 healthy individuals as the control group. After the experiment began, the subjects underwent 2-min recordings of their EEG signals in resting states with eyes open or closed, with records for each state repeated twice. The participants were then asked to imagine the movements of their left hand, and right hand. The entire process of MI consists of four task stages, with each stage containing three tasks. Each task randomly appears 10 times. Time–frequency features of MI-EEG signals were extracted using a continuous wavelet transform to enhance the effectiveness of decoding raw EEG signals. Functional and statistical analyses of brain regions during MI were conducted based on the extracted time–frequency features. Based on this, the motor intentions of patients with SCI were decoded using a GCN that integrates the functional topological relationships of the electrodes. The proposed network achieved a classification accuracy of 92.44 % for MI task recognition. Furthermore, the fusion of wavelet features demonstrated superior performance in classification and recognition. The results of this study confirm the efficacy of wavelet fusion in advancing MI feature decoding, enhancing the understanding of neurological conditions, such as SCI, and offering promising prospects for improving rehabilitation methods.