P300 EEG SIGNAL CLASSIFICATION USING RNN FOR COGNITIVE RESPONSE DETECTION
Abstract
P300-based EEG signals are a promising method for detecting brain responses to target and non-target stimuli. This study implements the Recurrent Neural Network (RNN) method to analyze and classify EEG signals recorded using a 19-channel EEG device based on the international 10–20 system standard. The preprocessing stage includes filtering to remove environmental noise and biological artifacts, followed by Independent Component Analysis (ICA) to ensure signal quality relevant for analysis. The results indicate that target stimuli produce higher P300 amplitudes and shorter latencies compared to non-target stimuli, with the Pz channel serving as the primary detection point. The RNN model achieved an average accuracy of 87.5%, precision of 88%, recall of 87.4%, and an F1-score of 87.7%. These findings confirm the reliability of RNN in capturing the temporal patterns of EEG signals and highlight its potential applications in neurotechnology, such as early detection of cognitive disorders and the development of neurofeedback systems.
Full Text:
PDFReferences
A. M. Albert and C. Cohen, “The Test for Severe Impairment: an instrument for the assessment of patients with severe cognitive dysfunction,” Journal of the American Geriatrics Society, vol. 40, no. 5, pp. 449–53, 1992.
A. M. Nawi et al., “Risk and adolescents: a systematic review,” BMC Public Health, vol. 21, no. 1, pp. 1–15, 2021.
A. Turnip et al., “Detection of Drug Effects on Brain Activity using EEG-P300 with Similar Stimuli,” IOP Conf. Ser.: Mater. Sci. Eng., vol. 220, no. 012042, 2017.
EMP Pusiknas Bareskrim Polri, "For Drug Dealers and Dealers, BNN: Don't Play with the State," [Online]. Available: https://pusiknas.polri.go.id/detail_artikel/untuk_para_bandar_dan_pengedar_narkoba,_bnn:_jangan_mainmain_dengan_negara#:~:text=Adapun%20jumlah%20orang%20yang%20dilaporkan%20terkait%20kasus%20narkoba%20sebanyak%204.865,total%20terlapor%20di%20September%202024.
A. Turnip et al., “Brain Mapping of Drug Addiction in Withdrawal Condition Based P300 Signals,” J. Phys.: Conf. Ser., vol. 1007, no. 012060, 2018.
A. Turnip et al., “Analysis of Neural Patterns and Cognitive Responses of Drug Users,” JEEE, vol. 476, no. 182, 2018.
M. S. Alhajeri et al., “Model Predictive Control of Nonlinear Processes using Transfer Learning-Based Recurrent Neural Networks,” Chem. Eng. Res. Des., vol. 205, pp. 1–12, 2024.
H. Shahinzadeh et al., “Deep Learning: A Overview of Theory and Architectures,” in Proc. 2024 20th CSI Int. Symp. Artificial Intelligence and Signal Processing (AISP), Babol, Iran, Feb. 2024, pp. 1–11.
R. D. Baruah and M. M. Organero, “Explicit Context Integrated Recurrent Neural Network for Applications in Smart Environments,” Expert Syst. Appl., vol. 255, no. 124752, 2024.
A. Turnip et al., “Brain Activity Analysis Using EEG with Machine Learning Approaches,” J. Phys.: Conf. Ser., vol. 987, no. 012065, 2020.
Wahyu Caesarendra, Mochammad Ariyanto, Kharisma A. Pambudi, M. Faizal Amri, Arjon Turnip, "Classification of EEG Signals for Eye Focuses Using Artificial Neural Network," ISSN: 1942-9703/© 2017 IIJ, Vol. 9/No. 1, 2017.
S. Liu and X. Chen, “Applying moving target defense against data theft ransomware on windows os,” 2023.
J. Rafapa and A. Konokix, “Ransomware detection using aggregated random forest technique with recent variants,” 2024.
Eton Blue, Gregory Campbell, Andrew Stokes, Lawrence Thompson, James Clarke, “Ransomware Detection on Linux Operating System Using Recurrent Neural Networks with Binary Opcode Analysis,” 2024.
Refbacks
- There are currently no refbacks.
Publisher:
Department of Electrical Engineering
Universitas Padjadjaran
Jl. Ir. Soekarno km.21, Jatinangor, Sumedang, Jawa Barat 45363