تحلیل چنددامنه‌ای سیگنال‌های مغزی تحریک‌شده برای شناسایی محتوای بصری چندکلاسه: رویکردی مبتنی بر ویژگی‌های زمان، فرکانس و فضا

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجوی دکتری، گروه مهندسی پزشکی، دانشکده فنی و مهندسی، واحد مشهد، دانشگاه آزاد اسلامی، مشهد، ایران

2 استادیار گروه مهندسی پزشکی، واحد مشهد، دانشکده فنی و مهندسی، دانشگاه آزاد اسلامی، مشهد، ایران

10.22034/abmir.2025.23463.1147

چکیده

در این مقاله، یک چارچوب جامع برای تحلیل و طبقه‌بندی محتوای بصری چندکلاسه مبتنی بر سیگنال‌های الکتروانسفالوگرافی تحریک‌شده ارائه شده است. با توجه به چالش‌های متعدد در استخراج ویژگی‌های معنادار از سیگنال‌های مغزی، رویکرد پیشنهادی بر تحلیل ترکیبی استخراج ویژگی در حوزه‌های زمانی، فرکانسی، زمان‌ـ‌فرکانس و محلی تمرکز دارد. ویژگی‌های استخراج‌شده با استفاده از روش‌های کلاسیک و ساختاری از سیگنال‌های الکتروانسفالوگرافی در کنار آزمون آماری Kruskal-Wallis انتخاب شده‌اند. سپس با بهره‌گیری از مدل‌های طبقه‌بندی متنوع، از جمله ماشین بردار پشتیبان، K-نزدیک‌ترین همسایه، پرسپترون چندلایه، درخت تصمیم و شبکه‌های عصبی کانولوشنی، عملکرد مدل‌ها ارزیابی گردید. نتایج نشان می‌دهد که ویژگی‌های حوزه زمان‌ـ‌فرکانس و سیگنال ثبت‌شده از کانال Fz، میانگین دقت 92/84٪ را در طبقه‌بندی چهارکلاسه بصری ارائه می‌دهند و بالاترین عملکرد را در میان سایر ترکیب‌ها دارد که نسبت به مراجع گذشته بین 2 تا 8 درصد بهبود داشته است. (با توجه به تعداد کلاس این افزایش مطلوب است) همچنین، روش‌های استخراج ویژگی 1D-LGP و LNDP نیز با دقت‌های بالا (83%) و مقاومت در برابر تغییرات کلاس‌ها، عملکرد قابل‌توجهی نشان دادند. این چارچوب پیشنهادی علاوه بر افزایش دقت طبقه‌بندی، از نظر محاسباتی نیز کارآمد بوده و قابلیت به‌کارگیری در سیستم‌های زمان واقعی را دارد.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

Multidomain Analysis of Visually Evoked Brain Signals for Multiclass Visual Content Classification: An Integrated Framework Combining Temporal, Spectral, and Spatial Features

نویسندگان [English]

  • Hamed Hakkak 1
  • Mohammad Mahdi Khalilzadeh 2
  • Mahdi Azarnoosh 2
  • Hamid Reza Kobravi 2
1 PhD student, Department of Biomedical Engineering, Faculty of Technology and Engineering, Islamic Azad University, Mashhad, Iran
2 Assistant Professor, Department of Biomedical Engineering, Faculty of Technology and Engineering, Islamic Azad University, Mashhad, Iran
چکیده [English]

This study proposes a comprehensive framework for the multidomain analysis and classification of multiclass visual content based on visually evoked electroencephalography (EEG) signals. Addressing the challenges inherent in extracting meaningful features from EEG data, the proposed approach integrates analyses across temporal, spectral, time–frequency, and local structural domains. Feature extraction leverages both classical and structural signal processing techniques, combined with statistical significance testing using the Kruskal–Walli’s method for feature selection. Subsequently, the classification performance of various machine learning models—including Support Vector Machines (SVM), k-Nearest Neighbors (KNN), Multilayer Perceptrons (MLP), and Convolutional Neural Networks (CNN)—was evaluated. The results demonstrate that time–frequency domain features (TFDF), particularly those derived from the Fz channel, achieve a mean accuracy of 84.92% in four-class visual content classification, outperforming other feature combinations. Moreover, local methods such as one-dimensional Local Gabor Patterns (1D-LGP) and Local Neighbor Descriptive Patterns (LNDP) exhibited high classification accuracy and robustness against noise. The proposed framework not only enhances classification accuracy but also maintains computational efficiency, making it a viable solution for real-time systems

کلیدواژه‌ها [English]

  • EEG signals
  • time–frequency domain analysis (TFDF)
  • brain signal feature extraction
  • visual content classification
  • statistical feature selection

مراجع

[1]     L. Spillmann, & Werner, J. S. (Eds.), “Visual perception: the neurophysiological foundations,” Elsevier, 2012.
[2]     E. N. Bruce, “Biomedical signal processing and signal modeling,” (No Title), 2001.
[3]     H. Saadi, M. Ferroukhi, & M. Attari, “Development of wireless high immunity EEG recording system,” In 2011 International Conference on Electronic Devices, Systems and Applications (ICEDSA) (pp. 120-124). IEEE, 2011, April.
[4]     M. D. McDonnell, & L. M. Ward, “The benefits of noise in neural systems: bridging theory and experiment,” Nature Reviews Neuroscience, 12(7), 415-425, 2011.
[5]     D. Mahmood, H. Nisar, & Y. V. Voon, “Removal of physiological artifacts from electroencephalogram signals: a review and case study,” In 2021 IEEE 9th Conference on Systems, Process and Control (ICSPC 2021) (pp. 141-146). IEEE, 2021, December.
[6]     S. Vaid, P. Singh, & C. Kaur, “EEG signal analysis for BCI interface: A review,” In 2015 fifth international conference on advanced computing & communication technologies (pp. 143-147). IEEE, 2015, February.
[7]     W. R. Rudnicki, M. Wrzesie, & W. Paja, “Feature Selection for Data and Pattern Recognition, ” Studies in Computational Intelligence, 584, 2015.
[8]     H. Yaacob, F. Hossain, S. Shari, S. K. Khare, C. P. Ooi, & U. R. Acharya, “Application of artificial intelligence techniques for brain–computer interface in mental fatigue detection: A systematic review (2011–2022), ” IEEE Access, 11, 74736-74758, 2023.
[9]     Z. L. Li, H. Cao, & J. S. Zhang, “Emotion Recognition in EEG Based on Multilevel Multidomain Feature Fusion, ” IEEE Access, 2024.
[10] S. Zhu, Z. Ye, Q. Ai, & Y. Liu, “EEG-ImageNet: An Electroencephalogram Dataset and Benchmarks with Image Visual Stimuli of Multi-Granularity Labels, ” arXiv preprint arXiv:2406.07151, 2024.
[11] X. Wang, Y. Pei, Z. Luo, S. Zhao, L. Xie, Y. Yan, ... & D. Ming, “Fusion of multi-domain EEG signatures improves emotion recognition, ” Journal of Integrative Neuroscience, 23(1), 18, 2024.
[12] Z. Lu, & J. Wang, “A novel and efficient multi-scale feature extraction method for EEG classification, ” AIMS Mathematics, 9(6), 16605-16622, 2024.
[13] Y. J. Chen, S. C. Chen, & C. M. Wu, “Multi-Domain Features and Multi-Task Learning for Steady-State Visual Evoked Potential-Based Brain–Computer Interfaces, ” Applied Sciences, 15(4), 2176, 2025.
[14] Z. T. Al-Qaysi, A. S. Albahri, M. A. Ahmed, R. A. Hamid, M. A. Alsalem, O. S. Albahri, ... & A. M. Duhaim, “A comprehensive review of deep learning power in steady-state visual evoked potentials, ” Neural Computing and Applications, 36(27), 16683-16706, 2024.
[15] L. Ghosh, P. Rakshit, & A. Konar, “Working memory modeling using inverse fuzzy relational approach, ” Applied Soft Computing, 83, 105591, 2019.
[16] N. Ji, L. Ma, H. Dong, & X. Zhang, “EEG signals feature extraction based on DWT and EMD combined with approximate entropy, ” Brain sciences, 9(8), 201, 2019.
[17] C. Zhang, Y. K. Kim, & A. Eskandarian, “EEG-inception: an accurate and robust end-to-end neural network for EEG-based motor imagery classification, ” Journal of Neural Engineering, 18(4), 046014, 2021.
[18] Y. Kaya, M. Uyar, R. Tekin, & S. Yıldırım, “1D-local binary pattern-based feature extraction for classification of epileptic EEG signals, ” Applied Mathematics and Computation, 243, 209-219, 2014.
[19] S. R. Das, D. Mishra, & M. Rout, “Stock market prediction using Firefly algorithm with evolutionary framework optimized feature reduction for OSELM method, ” Expert Systems with Applications: X, 4, 100016, 2019.
[20] Sengur, “An expert system based on linear discriminant analysis and adaptive neuro-fuzzy inference system to diagnosis heart valve diseases, ” Expert Systems with Applications, 35(1-2), 214-222, 2008.
[21] Jun, I. Choi, & D. Kim, “Local transform features and hybridization for accurate face and human detection, ” IEEE transactions on pattern analysis and machine intelligence, 35(6), 1423-1436, 2012.
[22] K. Jaiswal, & H. Banka, “Local pattern transformation-based feature extraction techniques for classification of epileptic EEG signals, ” Biomedical Signal Processing and Control, 34, 81-92, 2017.
[23] W. Zhang, D. Wu, J. Cao, L. Jiang, & T. Jiang, “Multibit local neighborhood difference pattern optimization for seizure detection of west syndrome EEG signals, ” IEEE Sensors Journal, 23(19), 22693-22703, 2023.
[24] M. Verma, & B. Raman, “Local neighborhood difference pattern: A new feature descriptor for natural and texture image retrieval, ” Multimedia Tools and Applications, 77(10), 11843-11866, 2018.
[25] Hudgins, P. Parker, & R. N. Scott, “A new strategy for multifunction myoelectric control, ” IEEE transactions on biomedical engineering, 40(1), 82-94, 1993.
[26] H. P. Huang, & C. Y. Chen, “Development of a myoelectric discrimination system for a multi-degree prosthetic hand, ” In Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No. 99CH36288C) (Vol. 3, pp. 2392-2397). IEEE, 1999, May.
[27] R. Boostani, & M. H. Moradi, “Evaluation of the forearm EMG signal features for the control of a prosthetic hand, ” Physiological measurement, 24(2), 309, 2003.
[28] S. Du, & M. Vuskovic, “Temporal vs. spectral approach to feature extraction from prehensile EMG signals, ” In Proceedings of the 2004 IEEE International Conference on Information Reuse and Integration, IRI 2004. (pp. 344-350). IEEE, 2004, November.
[29] Keil, EM. Bernat, MX. Cohen, M. Ding, M. Fabiani, G. Gratton, ES. Kappenman, E. Maris, KE. Mathewson, RT. Ward, N. Weisz, “Recommendations and publication guidelines for studies using frequency domain and time‐frequency domain analyses of neural time series, ” Psychophysiology.;59(5):e14052, 2022 May.
[30] F. Lotte, L. Bougrain, A. Cichocki, M. Clerc, M. Congedo, A. Rakotomamonjy, & F. Yger, “A review of classification algorithms for EEG-based brain–computer interfaces: a 10-year update, ” Journal of neural engineering, 15(3), 031005, 2018.
[31] V. Bajaj, S. Taran, S. K. Khare, & A. Sengur, “Feature extraction method for classification of alertness and drowsiness states EEG signals, ” Applied Acoustics, 163, 107224, 2020.
[32] S. Razavi, A. Jakeman, A. Saltelli, C. Prieur, B. Iooss, E. Borgonovo, ... & H. R. Maier, “The future of sensitivity analysis: an essential discipline for systems modeling and policy support, ” Environmental Modelling & Software, 137, 104954, 2021.
[33] T. Zhang, W. Chen, & M. Li, “Fuzzy distribution entropy and its application in automated seizure detection technique, ” Biomedical Signal Processing and Control, 39, 360-377, 2018.
[34] T. J. Luo, “Parallel genetic algorithm based common spatial patterns selection on time–frequency decomposed EEG signals for motor imagery brain-computer interface, ” Biomedical Signal Processing and Control, 80, 104397, 2023.
[35] P. Agarwal, & S. Kumar, “Electroencephalography based imagined alphabets classification using spatial and time‐domain features, ” International Journal of Imaging Systems and Technology, 32(1), 111-122, 2022.
[36] E. Gokgoz, & A. Subasi, “Comparison of decision tree algorithms for EMG signal classification using DWT, ” Biomedical signal processing and control, 18, 138-144, 2015.
[37]  X. Wang, V. Liesaputra, Z. Liu, Y. Wang, Z. Huang, “An in-depth survey on Deep Learning-based Motor Imagery Electroencephalogram (EEG) classification, ” Artificial Intelligence in Medicine. 2:102738, 2023 Dec.
[38] X. Tang, C. Yang, X. Sun, M. Zou, & H. Wang, “Motor imagery EEG decoding based on multi-scale hybrid networks and feature enhancement, ” IEEE Transactions on Neural Systems and Rehabilitation Engineering, 31, 1208-1218, 2023.
[39] X. Wang, X. Dai, Y. Liu, X. Chen, Q. Hu, R. Hu, & M. Li, “Motor imagery electroencephalogram classification algorithm based on joint features in the spatial and frequency domains and instance transfer, ” Frontiers in Human Neuroscience, 17, 1175399, 2023.
[40] A. K. Singh, & S. Krishnan, “Trends in EEG signal feature extraction applications, ” Frontiers in Artificial Intelligence, 5, 1072801, 2023.
[41] S. Maddury, “The performance of domain-based feature extraction on EEG, ECG, and fNIRS for Huntington’s disease diagnosis via shallow machine learning, ” Frontiers in Signal Processing, 4, 1321861, 2024.
[42] X. Wang, X. Dai, Y. Liu, X. Chen, Q. Hu, R. Hu, & M. Li, “Motor imagery electroencephalogram classification algorithm based on joint features in the spatial and frequency domains and instance transfer, ” Frontiers in Human Neuroscience, 17, 1175399, 2023
[43] X. Wang, X. Dai, Y. Liu, X. Chen, Q. Hu, R. Hu, & M. Li, “Motor imagery electroencephalogram classification algorithm based on joint features in the spatial and frequency domains and instance transfer, ” Frontiers in Human Neuroscience, 17, 1175399, 2023.
[44] Mahmoudi, N. S. Samani, A. Toumajian, “Enhancing the accuracy of hyperspectral image classification using an advanced convolutional neural network and deep learning. ” Journal of Spatial Information Technology Engineering. 2022 Mar 10;9(4):109-25.
پژوهش های نظری و کاربردی هوش ماشینی
دوره 4، شماره 1
بهار و تابستان 1405
اردیبهشت 1405
صفحه 1-14

فایل ها

هم رسانی

ارجاع به این مقاله

آمار