Improving the Accuracy of Motor Imagery in BCI for the Movement of Artificial Prostheses used for Disabled People
محورهای موضوعی : Machine learning
Isaac Jahanbakhshi
1
,
shaghayegh Niavarani
2
,
Fatemeh Shahbazi
3
,
Farahnaz Abdi
4
1 - Department of Computer science, Central Tehran Branch, Islamic Azad University, Tehran, Iran
2 - Department of Computer science, Science and Research Branch, Islamic Azad University, Tehran, Iran
3 - Department of Computer science, Kermanshah Branch, Jahad University, Kermanshah, Iran
4 - Department of Computer science and Software Engineering, University of Western Australia, Perth, Australia
کلید واژه: Brain-Computer Interface, Motor Imagery, CNN, GEO,
چکیده مقاله :
A brain-computer interface (BCI) allows users to communicate directly with an external device, such as a computer, using brain signals. These systems involve signal acquisition, temporal and spatial filtering, feature engineering, and classification before transmitting the control signal to an external device. Brain-computer interfaces based on motor imagery (MI-BCI) hold significant potential for applications in motor enhancement and rehabilitation. However, the control capabilities of MI-BCI can vary among individuals. In this article, we present a motor imagery model that leverages the power of deep learning for feature extraction and optimization methods for feature selection, based on individuals' electroencephalogram (EEG) signals. For this purpose, we employed the convolutional neural network (CNN), the golden eagle optimization (GEO) method, and three hybrid classification models to achieve high accuracy in classifying EEG signals. The innovative classification method presented in the third phase of the proposed method has high flexibility and significant accuracy, which is presented for the first time. This method can be used in all matters of prediction and detection of phenomena to increase accuracy and high scalability. The experimental results demonstrate that the proposed model significantly outperforms baseline approaches across all key performance indicators. In terms of precision, the method achieves an improvement of approximately 12–18% over standard CNN and nearly 20% compared to conventional classifiers. For recall, it yields 10–15% higher performance than CNN and 18–22% higher than classical machine learning methods. Finally, the F-measure results indicate a 12–15% improvement over CNN and about 20% over traditional classifiers. These consistent enhancements confirm the robustness and scalability of the proposed approach for motor imagery-based EEG classification and highlight its potential for practical applications in rehabilitation and prosthetic control systems.
A brain-computer interface (BCI) allows users to communicate directly with an external device, such as a computer, using brain signals. These systems involve signal acquisition, temporal and spatial filtering, feature engineering, and classification before transmitting the control signal to an external device. Brain-computer interfaces based on motor imagery (MI-BCI) hold significant potential for applications in motor enhancement and rehabilitation. However, the control capabilities of MI-BCI can vary among individuals. In this article, we present a motor imagery model that leverages the power of deep learning for feature extraction and optimization methods for feature selection, based on individuals' electroencephalogram (EEG) signals. For this purpose, we employed the convolutional neural network (CNN), the golden eagle optimization (GEO) method, and three hybrid classification models to achieve high accuracy in classifying EEG signals. The innovative classification method presented in the third phase of the proposed method has high flexibility and significant accuracy, which is presented for the first time. This method can be used in all matters of prediction and detection of phenomena to increase accuracy and high scalability. The experimental results demonstrate that the proposed model significantly outperforms baseline approaches across all key performance indicators. In terms of precision, the method achieves an improvement of approximately 12–18% over standard CNN and nearly 20% compared to conventional classifiers. For recall, it yields 10–15% higher performance than CNN and 18–22% higher than classical machine learning methods. Finally, the F-measure results indicate a 12–15% improvement over CNN and about 20% over traditional classifiers. These consistent enhancements confirm the robustness and scalability of the proposed approach for motor imagery-based EEG classification and highlight its potential for practical applications in rehabilitation and prosthetic control systems.
[1] H. Wang, P. Chen, M. Zhang, J. Zhang, X. Sun, M. Li, and Z. Gao, “EEG-Based Motor Imagery Recognition Framework via Multisubject Dynamic Transfer and Iterative Self-training,” IEEE Trans Neural Netw Learn Syst., vol. 35, no. 8, pp. 10698-10712, Aug. 2024. DOI: 10.1109/TNNLS.2023.3243339.
[2] H. ALTAHERI, et al. “Deep learning techniques for classification of electroencephalogram (EEG) motor imagery (MI) signals: A review,” Neural Comput Appl, vol. 35, no 20, pp. 14681-14722, Jul. 2023. DOI: 10.1007/s00521-021-06352-5.
[3] J. RAHUL, D. SHARMA, L. D. SHARMA, U. Nanda, and A. K. Sarkar, “A systematic review of EEG based automated schizophrenia classification through machine learning and deep learning,” Front Hum Neurosci, vol. 18, p. 1347082, Feb. 2024. DOI: 10.3389/fnhum.2024.1347082.
[4] A. Nandhini and J. Sangeetha, “A Review on Deep Learning Approaches for Motor Imagery EEG Signal Classification for Brain–Computer Interface Systems,” in Proc. Computational Vision and Bio-Inspired Computing. Advances in Intelligent Systems and Computing, Singapore, vol. 1439, pp. 353-365, Apr. 2023. DOI: 10.1007/978-981-19-9819-5_27.
[5] I. Raoof, and M. K. Gupta, “Domain-independent short-term calibration-based hybrid approach for motor imagery electroencephalograph classification: a comprehensive review,” Multimed Tools Appl, vol. 83, no 3, pp. 9181-9226, Jan. 2024. DOI: 10.1007/s11042-023-15900-1.
[6] I. ABDULLAH FAYE, and M. R. ISLAM, “EEG channel selection techniques in motor imagery applications: a review and new perspectives,” Bioengineering, vol. 9, no 12, p. 726. Nov. 2022. DOI: 10.3390/bioengineering9120726.
[7] V. S. KARDAM, S. TARAN and A. PANDEY, “Motor imagery tasks-based electroencephalogram signals classification using data-driven features,” Neurosci Inform, vol. 3, no 2, p. 100128, Jun. 2023. DOI: 10.1016/j.neuri.2023.100128.
[8] D. JAIPRIYA and K. C. SRIHARIPRIYA, “Brain computer interface-based signal processing techniques for feature extraction and classification of motor imagery using EEG: A literature review,” Biomed Mater Devices, vol. 2, no 2, pp. 601-613, Sep. 2024. DOI: 10.1007/s44174-023-00082-z.
[9] P. ARPAIA, F. DONNARUMMA, A. ESPOSITO and M Parvis, “Channel selection for optimal EEG measurement in motor imagery-based brain-computer interfaces,” Int J Neural Syst, vol. 31, no 03, p. 2150003, 2021. DOI: 10.1142/S0129065721500039.
[10] C. Zuo, et al, “Cluster decomposing and multi-objective optimization based-ensemble learning framework for motor imagery-based brain–computer interfaces,” J Neural Eng, vol. 18, no. 2, p. 026018, 2021. DOI: 10.1088/1741-2552/abe20f.
[11] Y. Chen, et al. “A novel transfer support matrix machine for motor imagery-based brain computer interface,” Front Neurosci, vol. 14, p. 606949, Nov. 2020. DOI: 10.3389/fnins.2020.606949.
[12] J. Jin, et al. “Bispectrum-based channel selection for motor imagery based brain-computer interfacing,” in Proc. IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 28, no. 10, pp. 2153-2163, Oct. 2020. DOI: 10.1109/TNSRE.2020.3020975.
[13] J. V. Riquelme-Ros, G. Rodríguez-Bermúdez, I. Rodríguez-Rodríguez, J. V. Rodríguez and J. M. Molina-García-Pardo, “On the better performance of pianists with motor imagery-based brain-computer interface systems,” Sensors, vol.20, no. 16, p. 4452, Aug 2020. DOI: 10.3390/s20164452.
[14] O. P. Idowu, P. Fang and G. Li, “Bio-inspired algorithms for optimal feature selection in motor imagery-based brain-computer interface,” in Proc. 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Montreal, QC, Canada, pp. 519-522, Jul. 2020, DOI: 10.1109/EMBC44109.2020.9176244.
[15] C. Liu, et al. “Distinguishable spatial-spectral feature learning neural network framework for motor imagery-based brain–computer interface,” J Neural Eng, vol. 18, no. 4, p. 0460e4, Aug. 2021. DOI: 10.1088/1741-2552/ac1d36.
[16] R. Dhiman, “Motor imagery signal classification using Wavelet packet decomposition and modified binary grey wolf optimization,” Measurement: Sensors, vol. 24, p. 100553, Dec 2022. DOI: 10.1016/j.measen.2022.100553.
[17] X. Yu, M. Z. Aziz, M.T. Sadiq, Z. Fan and G. Xiao, “A new framework for automatic detection of motor and mental imagery EEG signals for robust BCI systems,” IEEE Trans Instrum Meas, vol. 70, pp. 1-12, 2021. DOI: 10.1109/TIM.2021.3069026.
[18] A. Ak, V. Topuz and I. Midi, “Motor imagery EEG signal classification using image processing technique over GoogLeNet deep learning algorithm for controlling the robot manipulator,” Biomed Signal Process Control, vol. 72, p. 103295, Feb. 2022. DOI: 10.1016/j.bspc.2021.103295.
[19] Z. Khademi, F. Ebrahimi and H. M. Kordy, “A transfer learning-based CNN and LSTM hybrid deep learning model to classify motor imagery EEG signals,” Comput Biol Med, vol. 143, p. 105288, Apr. 2022. DOI: 10.1016/j.compbiomed.2022.105288.
[20] N. Tibrewal, N. Leeuwis and M. Alimardani, “Classification of motor imagery EEG using deep learning increases performance in inefficient BCI users,” Plos one, vol. 17, no. 7, p. e0268880, Jul. 2022. DOI: 10.1371/journal.pone.0268880.
[21] X. Meng, S. Qiu, S. Wan, K. Cheng and L. Cui, “A motor imagery EEG signal classification algorithm based on recurrence plot convolution neural network,” Pattern Recognit Lett, vol. 146, pp. 134-141, Jun. 2021. DOI: 10.1016/j.patrec.2021.03.023.
[22] M. N. Cherloo, H. K. Amiri and M. R. Daliri, “Ensemble regularized common spatio-spectral pattern (ensemble RCSSP) model for motor imagery-based EEG signal classification,” Comput Biol Med, vol. 135, p. 104546, Aug. 2021. DOI: 10.1016/j.compbiomed.2021.104546.
[23] X. Wang, R. Yang and M. Huang, “An unsupervised deep-transfer-learning-based motor imagery EEG classification scheme for brain–computer interface,” Sensors, vol. 22, no. 6, p. 2241, Feb. 2022. DOI: 10.3390/s22062241.
[24] H. Li, M. Ding, R. Zhang and C. Xiu, “Motor imagery EEG classification algorithm based on CNN-LSTM feature fusion network,” Biomed Signal Process Control, vol. 72, p. 103342, Feb. 2022. DOI: 10.1016/j.bspc.2021.103342.
[25] J. Uddin, "An Autoencoder based Emotional Stress State Detection Approach by using Electroencephalography Signals", Journal of Information Systems and Telecommunication (JIST) RICEST, vol. 11, no. 41, pp. 24-30, Jun. 2023. DOI: 10.52547/jist.32267.11.41.24.
[26] S. Motamed and E. Askari, “Recognition of Attention Deficit/Hyperactivity Disorder (ADHD) Based on Electroencephalographic Signals Using Convolutional Neural Networks (CNNs),” Journal of Information Systems and Telecommunication (JIST). Vol. 10, No. 39, pp. 222-228, Aug. 2022. DOI: 10.52547/jist.16399.10.39.222.
[27] A. Mohammadi-Balani, M. D. Nayeri, A. Azar and M. Taghizadeh-Yazdi, “Golden eagle optimizer: A nature-inspired metaheuristic algorithm,” Comput Ind Eng, vol. 152, p. 107050, Feb. 2021. DOI: 10.1016/j.cie.2020.107050.
[28] Y. Xie and S. Oniga, “Classification of Motor Imagery EEG Signals Based on Data Augmentation and Convolutional Neural Networks,” Sensors, vol. 23, no. 4, p. 1932, Feb. 2023. DOI: doi.org/10.3390/s23041932.
[29] L. Abualigah, A. Diabat, S. Mirjalili, M. Abd Elaziz and A. H. Gandomi, “The Arithmetic Optimization Algorithm,” Comput Methods Appl Mech Eng, vol. 376, p. 113609, Apr. 2021. DOI: 10.1016/j.cma.2020.113609.
[30] B. Abdollahzadeh, F. S. Gharehchopogh and S. Mirjalili, “African vultures optimization algorithm: A new nature-inspired metaheuristic algorithm for global optimization problems,” Comput Ind Eng, vol. 158, p. 107408, Aug. 2021. DOI: 10.1016/j.cie.2021.107408.