پیوندهای مفید
آمار
| تعداد نشریات | 22 |
| تعداد شمارهها | 705 |
| تعداد مقالات | 10,146 |
| تعداد مشاهده مقاله | 71,450,492 |
| تعداد دریافت فایل اصل مقاله | 63,173,980 |
شناسایی سیگنالهای تصور حرکتی با استفاده از یادگیری عمیق | ||
| مدل سازی در مهندسی | ||
| دوره 24، شماره 84، فروردین 1405، صفحه 139-151 اصل مقاله (630.55 K) | ||
| نوع مقاله: مقاله پژوهشی | ||
| شناسه دیجیتال (DOI): 10.22075/jme.2025.38105.2864 | ||
| نویسندگان | ||
| لوئیزا دهیادگاری* 1؛ راضیه راستگو2 | ||
| 1دانشکده مهندسی برق ، دانشگاه صنعتی سیرجان، سیرجان، ایران | ||
| 2دانشکده مهندسی برق و کامپیوتر، دانشگاه سمنان، سمنان، ایران | ||
| تاریخ دریافت: 25 خرداد 1404، تاریخ بازنگری: 31 مرداد 1404، تاریخ پذیرش: 17 شهریور 1404 | ||
| چکیده | ||
| شناسایی سیگنالهای الکتروانسفالوگرافی (EEG) مرتبط با فرآیند تصور حرکت نقش کلیدی در تحلیل و ارزیابی عملکردهای عصبی در سیستمهای رابط مغز و رایانه (BCI) ایفا میکند. با این حال، تفاوتهای فردی قابل توجه در الگوهای EEG چالشی جدی برای طراحی مدلهای دقیق و عمومیپذیر ایجاد کرده است. از سوی دیگر، تشخیص موفق تصور حرکت از بازههای زمانی کوتاهتر سیگنال، تأثیر مستقیمی بر افزایش کارایی و قابلیت استفاده عملی این فناوریها دارد. در این مقاله، یک چارچوب ترکیبی نوآورانه ارائه شده است که به منظور طبقهبندی سیگنالهای EEG ناشی از تصور حرکت، یک معماری دو مرحلهای مبتنی بر ترکیب اینفورمر بهبودیافته و شبکه EEGNet پیشنهاد میکند. در این ساختار ابتدا سیگنالهای EEG پس از استخراج ویژگیهای فرکانسی اولیه، به ماژول اینفورمر بهبودیافته وارد میشوند. این ماژول با بهرهگیری از مکانیزم توجه پراکنده و فیلترهای فرکانسی تطبیقی قادر است وابستگیهای زمانی بلندمدت در دادههای EEG را به طور مؤثر استخراج کند. خروجی اینفورمر سپس به مدل EEGNet منتقل میشود. با طراحی کانولوشنهای خاص (کانولوشن فضایی، کانولوشن عمقی، و کانولوشن تفکیکپذیر زمانی) بهطور هدفمند ویژگیهای فضایی-زمانی سیگنالهای EEG را استخراج میکند و نمایهای فشرده و قدرتمند برای طبقهبندی نهایی تولید میکند. نتایج تجربی حاصل از ارزیابیها نشان میدهد که مدل پیشنهادی در سناریوی میانگروهی و با استفاده از دیتاست استانداردPhysioNet موفق به دستیابی به دقت 85.20 درصد برای بازههای زمانی کوتاه 2 ثانیهای شده است. مقایسه عملکرد با مدلهای پیشرفته موجود نشان میدهد که رویکرد پیشنهادی در مواجهه با دادههای کوتاهتر و تنوع شرکتکنندگان عملکرد رقابتی و بهبودیافتهای دارد. | ||
| کلیدواژهها | ||
| سیگنالهای الکتروانسفالوگرافی؛ تصور حرکتی؛ اینفورمر؛ EEGNet؛ کانولوشن | ||
| عنوان مقاله [English] | ||
| Motor Imagery Signal Recognition Using Deep Learning | ||
| نویسندگان [English] | ||
| Louiza Dehyadegari1؛ Razieh Rastgoo2 | ||
| 1Department of Electronic Engineering, Sirjan University of Technology,Sirjan, Iran | ||
| 2Electrical and Computer Engineering Faculty, Semnan University, Semnan, Iran | ||
| چکیده [English] | ||
| The identification of electroencephalography (EEG) signals related to motor imagery plays a key role in the analysis and evaluation of neural functions in brain-computer interface (BCI) systems. However, considerable individual differences in EEG patterns pose a significant challenge for designing accurate and generalizable models. Moreover, the ability to successfully recognize motor imagery from shorter signal durations has a direct impact on improving the efficiency and practical usability of these technologies. In this study, an innovative hybrid framework is proposed for classifying motor imagery EEG signals, introducing a two-stage architecture based on the combination of an enhanced Informer and the EEGNet model. In this architecture, the EEG signals, after initial frequency feature extraction, are first fed into the enhanced Informer module. This module, leveraging sparse attention mechanisms and adaptive frequency filters (FAA), effectively captures long-term temporal dependencies within the EEG data. The output of the Informer is then passed to the EEGNet model, which, through its specialized convolutional layers (spatial convolution, depthwise convolution, and separable temporal convolution), purposefully extracts spatial-temporal features from the EEG signals and generates a compact and discriminative representation for final classification. Experimental results demonstrate that the proposed model achieves 85.20% accuracy in cross-subject evaluation on the standard PhysioNet dataset with short 2-second trial durations. Comparative analyses with state-of-the-art models indicate that the proposed approach offers competitive and improved performance, particularly in handling shorter signal durations and participant diversity. | ||
| کلیدواژهها [English] | ||
| Electroencephalography signals, Motor imagery, Informer, EEGNet, Convolution | ||
|
سایر فایل های مرتبط با مقاله
|
||
| مراجع | ||
|
[1] Wolpaw, Jonathan, and Elizabeth W. Wolpaw. "Brain-Computer Interfaces: Principles and Practice." Oxford University Press, USA, 2012. [2] Wolpaw, Jonathan R., Niels Birbaumer, Donald J. McFarland, Gert Pfurtscheller, and Theresa M. Vaughan. "Brain–computer interfaces for communication and control." Clinical Neurophysiology 113, no. 6 (2002): 767-791. [3] Pfurtscheller, Gert, and Christa Neuper. "Motor imagery and direct brain-computer communication." Proceedings of the IEEE 89, no. 7 (2001): 1123-1134. [4] Hamidi, Arezoo, and Kourosh Kiani. "Motor Imagery EEG signals classification using a Transformer-GCN approach." Applied Soft Computing 170 (2025): 112686. [5] Lotte, Fabien, Loïc Bougrain, Andrzej Cichocki, Maud Clerc, Alexandre Congedo, A. Rakotomamonjy, and F. Yger. "A review of classification algorithms for EEG-based brain–computer interfaces: A roadmap to algorithm design." Journal of Neural Engineering 15, no. 6 (2018): 1741-2552. [6] Altaheri, Hassan. "Deep learning techniques for classification of electroencephalogram (EEG) motor imagery (MI) signals: a review." Neural Comput. Appl. 35, no. 20 (2023): 14681–14722. [7] Hamidi, Arezoo, and Kourosh Kiani. "Motor Imagery EEG signals classification using a Transformer-GCN approach." Applied Soft Computing 170, no. 10 (2025): 112686. [8] Ahmadi, Amir Mohammad, Kourosh Kiani, and Razieh Rastgoo. "A Transformer-based model for abnormal activity recognition in video." Journal of Modeling in Engineering 22, no. 76 (2024): 213–221. [9] Esfandiari, Nura, Kourosh Kiani, and Razieh Rastgoo. "A conditional generative chatbot using transformer model." Journal of Modeling in Engineering 23, no. 82 (2025): 99–113. [10] Ma, Zhen. "Transformed common spatial pattern for motor imagery-based brain-computer interfaces." Front. Neurosci 17, no. 3 (2023): 1116721. [11] Hou, Yanlong. "GCNs-net: a graph convolutional neural network approach for decoding time-resolved eeg motor imagery signals." IEEE Trans. Neural Netw. Learn. Syst. 35, no. 6 (2022): 7312 – 7323. [12] Ma, Yang, Yong Song, and Feng Gao. "A novel hybrid CNN-Transformer model for EEG Motor Imagery classification." In Proceedings of the 2022 International Joint Conference on Neural Networks (IJCNN), IEEE, 2022. [13] Barnova, Klara, Martina Mikolasova, Renata V. Kahankova, Renata Jarosa, Aliaksandr K. Sterniuk, Vlastimil Snasel, Seyed Mirjalili, Marek Pelcb, and Radek Martineka. "Implementation of artificial intelligence and machine learning-based methods in brain–computer interaction." Computers in Biology and Medicine 163 (2023): 107135. [14] Luo, Ting-Jie, Chun Zhou, and Fenlin Chao. "Exploring spatial-frequency-sequential relationships for motor imagery classification with recurrent neural network." BMC Bioinforma 19 (2018): 344. [15] Tortora, Simone, Simone Ghidoni, Carmela Chisari, Silvestro Micera, and Francesco Artoni. "Deep learning-based BCI for gait decoding from EEG with LSTM recurrent neural network." Journal of Neural Engineering 17 (2020): 046011. [16] Bang, Jong Sik, and Seung Woo Lee. "Motor imagery classification based on CNN-GRU network with spatio-temporal feature representation." In: Wallraven, C., Liu, Q., Nagahara, H. (eds) Pattern Recognition. ACPR. Lecture Notes in Computer Science, 13188. (2022): 104–115. [17] Dai, Yanning, Xiaowei Deng, Xiaoli Fu, and Yunjie Zhao. "Periodicity-based multi-dimensional interaction convolution network with multi-scale feature fusion for motor imagery EEG classification." Journal of Neuroscience Methods 415 (2025): 110356. [18] Bagherzadeh, Fahimeh, and Razieh Rastgoo. "Deepfake image detection using a deep hybrid convolutional neural network." Journal of Modeling in Engineering 21, no. 75 (2023): pp. 19–28. [19] Chowdhurya, Riya Sh, Sh. Boseb, S. Ghosha, and A. Konara. "Attention Induced Dual Convolutional-Capsule Network (AIDC-CN): A deep learning framework for motor imagery classification." Computers in Biology and Medicine 183 (2024): 109260. [20] Wang, Xiaoli, Yu Wang, Wei Qia, Dawei Kong, and Wei Wang. "Brain GridNet: A two-branch depth wise CNN for decoding EEG-based multi-class motor imagery." Neural Networks 170 (2024): 312-324. [21] Luo, Jian, Yong Wang, Shiqiang Xia, Ning Lu, Xiaoming Ren, Zhifeng Shi, and Xiaodong Hei. "A shallow mirror transformer for subject-independent motor imagery BCI." Computers in Biology and Medicine 164 (2023): 107254. [22] Thangaraj, Thenmozhi, Helen Rajendran, and S. Mythili. "Classification of motor imagery EEG with ensemble RNCA model." Behavioural Brain Research 479 (2025): 115345. [23] Saibenea, Aurora, Hafez Ghaemi, and Eda Dagdevir. "Deeplearning in motor imagery EEG signal decoding: A Systematic Review." Neurocomputing 610 (2024): 128577. [24] Yang, Yong, Xia Zhang, Xia Zhang, and Cha. Yu. "MCMT Net: Advanced network architectures for EEG-based motor imagery classification." Neurocomputing 620 (2025): 129255. [25] Gao, Yan, Wen Xie, Zh. Luo, M. Houston, and Y. Zhang. "Multi-domain feature analysis of MI-EEG signals using tensor train decomposition and projected gradient Non-negative Matrix Factorization." Neurocomputing 623, no. 28 (2025): 129410. [26] Jiang, Xiaojun, Ling Meng, Xin Chen, Yiming Xu, and Dong Wu. "CSP-Net: Common spatial pattern empowered neural networks for EEG-based motor imagery classification." Knowledge-Based Systems 305, no. 3 (2024): 112668. [27] Karimian-Kelishadrokhi, Mohammad, and Fatemeh Safi-Esfahani. "Application of the time-distributed layer in the controller of memory-augmented neural networks to classify brain activities into motor imagery and motor execution." Applied Soft Computing 162 (2024): 111771. [28] Abenna, Said, Mohammed Nahid, Hamid Bouyghf, Brahim Ouacha. "An enhanced motor imagery EEG signal prediction system in real-time based on delta rhythm." Biomedical Signal Processing and Control 79, no. 2 (2023): 104210. [29] Cao, Jian, Gang Li, Jin Shen, and Chuang Dai. "IFBCL Net: Spatio-temporal frequency feature extraction-based MI-EEG classification convolutional network." Biomedical Signal Processing and Control 92 (2024): 106092. [30] Lawhern, Vernon J., Amelia J. Solon, Nicholas R. Waytowich, Stephen M. Gordon, Chou P. Hung and Brent J Lance. "EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces." Journal of Neural Engineering 15, no. 5 (2018): 056013. [31] Zhou, Haoyi, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, Wancai Zhang. "Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting." The Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI-21), pp. 11106-11115, 2021. [32] Vaswani, Ashish, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. "Attention Is All You Need." arXiv:1706.03762, 2023. [33] Zarbafi, Sahar, Kourosh Kiani, and Razieh Rastgoo. "Spoken Persian digits recognition using deep learning." Journal of Modeling in Engineering 21, no. 74 (2023): pp. 163–172. [34] Rastgoo, Razieh, and Kourosh Kiani. "Face recognition using fine-tuning of Deep Convolutional Neural Network and transfer learning." Journal of Modeling in Engineering 17, no. 58 (2019): pp. 103–111. [35] Schalk, Gerwin, Dennis J. McFarland, Thilo Hinterberger, Niels Birbaumer, and Jonathan R. Wolpaw. "BCI2000: a general-purpose brain-computer interface (BCI) system." IEEE Transactions on biomedical engineering 51, no. 6 (2004): 1034-1043. [36] Xie, Jin, Jie Zhang, Jiayao Sun, Zheng Ma, Liuni Qin, Guanglin Li, Huihui Zhou, and Yang Zhan. "A transformer-based approach combining deep learning network and spatial-temporal information for raw EEG classification." IEEE Transactions on Neural Systems and Rehabilitation Engineering 30 (2022): 2126-2136. [37] Wang, Xiaying, Michael Hersche, Batuhan Tömekce, Burak Kaya, Michele Magno, Luca Benini. "An accurate eegnet-based motor-imagery brain–computer interface for low-power edge computing." in 2020 IEEE International Symposium on Medical Measurements and Applications (MeMeA). 2020. IEEE. [38] Chowdhury, Radia Rayan, Yar Muhammad, and Usman Adeel. "Enhancing cross-subject motor imagery classification in EEG-based brain–computer interfaces by using multi-branch CNN." Sensors 23, no. 18 (2023): 7908. [39] Moaveninejad, Sadaf, Valentina D'Onofrio, Franca Tecchio, Francesco Ferracuti, Sabrina Iarlori, Andrea Monteriù, and Camillo Porcaro. "Fractal Dimension as a discriminative feature for high accuracy classification in motor imagery EEG-based brain-computer interface." Computer Methods and Programs in Biomedicine 244 (2024): 107944. [40] Ali, Omair, Muhammad Saif-ur-Rehman, Tobias Glasmachers, Ioannis Iossifidis, and Christian Klaes. "ConTraNet: A hybrid network for improving the classification of EEG and EMG signals with limited training data." Computers in biology and medicine 168 (2024): 107649. [41] Hmaidi, Arezoo, Kourosh Kiani, Razieh Rastgoo." A Fusion of Transformer and EEGnet Models for Motor Imagery EEG Decoding." Journal of Modeling in Engineering 23 (2025): 191-202. | ||
|
آمار تعداد مشاهده مقاله: 405 تعداد دریافت فایل اصل مقاله: 101 |
||
