Citation

Lew, Kai Liang and Sim, Kok Swee and Ting, Zehong (2024) Deep Learning Approach EEG Signal Classification. JOIV : International Journal on Informatics Visualization, 8 (3-2). pp. 1693-1702. ISSN 2549-9610

Text 2959-9033-1-PB.pdf - Published Version Restricted to Repository staff only Download (3MB)

Abstract

The introduction of deep learning technology has greatly benefited the neuroscience field by improving the electroencephalogram (EEG) signal analysis. These technologies have greatly improved the understanding of complex brain activity by interpreting the signal as normal or abnormal. The EEG signal requires expertise to interpret the pattern, and only then can the EEG signal be differentiated as normal or abnormal. However, some variations always complicate the analysis of the EEG signal by creating noise in the signal. This paper introduces a deep learning model, NeuroNetFlex (NFF), to classify the EEG signal as normal or abnormal. The NNF is designed to classify the EEG signal by using multiple combinations of modules such as one-dimension convolutional neural networks (1D-CNN), Squeeze-and-Excitation (SE) blocks, and the parallel processing fusion of recurrent neural networks (RNN), Long Short Term Memory (LSTM), and Gated Recurrent Unit (GRU) layers are used to analyze the temporal features of the EEG data and learn the signal pattern. The performance of the NNF was evaluated using evaluation metrics such as accuracy, precision, recall, and f1 score. The model achieved an accuracy of 75.33%, a precision of 76.39%, a recall of 75.33%, and an F1 score of 75.08% with a training time of 16.88 minutes, outperforming the existing models. These results demonstrate the promising potential of the NNF to significantly improve the analysis of brain activities.

Item Type:	Article
Uncontrolled Keywords:	EEG signal classification; temporal feature extraction; deep learning; electroencephalogram; parallel processing.
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Engineering and Technology (FET)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	03 Jan 2025 04:16
Last Modified:	03 Jan 2025 04:16
URII:	http://shdl.mmu.edu.my/id/eprint/13285

Downloads

Edit (login required)