Journal
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
Volume -, Issue -, Pages -Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TNNLS.2023.3252569
Keywords
Electroencephalography; Brain modeling; Feature extraction; Pattern classification; Deep learning; Convolutional neural networks; Computational modeling; Alternating direction method of multipliers (ADMM); electroencephalogram (EEG) signal classification; graph neural network (GNN); nonconvextiy; weight pruning
Ask authors/readers for more resources
In this article, a sparse spectra graph convolutional network (SSGCNet) is proposed for epileptic electroencephalogram (EEG) signal classification. The aim is to develop a lightweight deep learning model that maintains a high level of classification accuracy. To achieve this, a weighted neighborhood field graph (WNFG) is introduced to represent EEG signals. The WNFG reduces redundant edges and has lower graph generation time and memory usage. The sequential graph convolutional network is then extended from the WNFG using sparse weight pruning and the alternating direction method of multipliers (ADMM). Compared to the state-of-the-art method, our approach achieves the same classification accuracy with a ten times smaller connection rate on the Bonn public dataset and the spikes and slow waves (SSW) clinical real dataset.
In this article, we propose a sparse spectra graph convolutional network (SSGCNet) for epileptic electroencephalogram (EEG) signal classification. The goal is to develop a lightweighted deep learning model while retaining a high level of classification accuracy. To do so, we propose a weighted neighborhood field graph (WNFG) to represent EEG signals. The WNFG reduces redundant edges between graph nodes and has lower graph generation time and memory usage than the baseline solution. The sequential graph convolutional network is further developed from a WNFG by combining sparse weight pruning and the alternating direction method of multipliers (ADMM). Compared with the state-of-the-art method, our method has the same classification accuracy on the Bonn public dataset and the spikes and slow waves (SSW) clinical real dataset when the connection rate is ten times smaller.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available