臺灣博碩士論文加值系統

腦電波訊號能夠呈現出人類大腦的精神狀態與動作意圖，腦機介面透過分析腦電波訊號，並將其轉換成對外控制的指令，提供人機互動的途徑。運動想像腦電波被廣泛應用在腦機介面的研究領域中，因此如何從複雜的腦電波訊號中正確地辨識出各種特徵與其所對應的動作是一項相當重要的技術。
在目前運動想像的相關研究中，大多數的方法還沒有充分考慮到腦電波訊號在頻率、時間和空間域的特徵訊息，這些模型的結構未能有效地提取具有判別力的特徵，進而導致分類的效能受限。
為了解決此問題，本研究提出一種基於時頻通道加權的運動想像腦電波分類方法。此方法藉由通道注意力模塊，判斷各個EEG通道的重要程度，選擇性的調節不同通道的激勵程度。接著以連續小波轉換將腦電波訊號轉換成二維的時頻形式，並搭配獨立樣本T檢定，檢驗不同運動想像任務之間的特徵差異，找出更具判別力的特徵區段。
本研究使用MI-EEG公共資料庫進行研究。實驗結果顯示，所提出的方法明顯優於其他現有的方法，證實此方法能夠有效地分類運動想像腦電波。

Electroencephalogram (EEG) reflects the mental state and action intentions of the human brain, and the brain-computer interface provides a way for human-computer interaction by analyzing EEG and converting them into externally controlled instructions. Motor Imagery (MI) EEG is widely used in brain-computer interface research, so how to correctly identify various features and their corresponding actions from complex EEG is a very important technique.
In the current research related to MI-EEG, most methods have not fully taken into account the feature of EEG in the frequency, time and space domains, and the structure of these models has not effectively extracted the discriminative features, which in turn leads to limited efficiency of classification.
In order to solve this problem, this study proposes a method of time-frequency-channel weighted MI-EEG classification. This method uses the channel-attention module to judge the importance of each EEG channel and selectively adjust the incentive degree of different channels. Then, Continuous Wavelet Transform (CWT) converts EEG into two-dimensional time-frequency, and with Independent Sample T-test, the feature differences between different MI tasks are examined to find more discriminative feature segments.
This study was conducted using the MI-EEG public database. The experimental results show that the proposed method is significantly superior to other methods, confirming that this method can effectively of MI-EEG classification.

致謝 i
摘要 ii
Abstract iii
目錄 iv
圖目錄 vi
表目錄 viii
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機 2
1.3 研究問題及目的 3
1.4 預期結果及貢獻 4
第二章 文獻探討 5
2.1 腦電波訊號簡介 5
2.2 基於機器學習演算法之相關文獻 10
2.1.1 傅立葉轉換（Fourier transform [FT]） 11
2.1.2 小波轉換（Wavelet Transform [WT]） 12
2.1.3 公共空間模式（Common spatial pattern [CSP]） 12
2.3 基於深度學習演算法之相關文獻 15
第三章 材料與研究方法 20
3.1 實驗材料 20
3.2 研究架構與流程 22
3.3 研究方法之步驟 24
3.3.1 資料預處理 24
3.3.2 卷積運算 24
3.3.3 EEG通道注意力模塊 24
3.3.4 連續小波轉換（Continuous Wavelet Transform [CWT]） 27
3.3.5 獨立樣本T檢定（Independent Samples T-Test） 28
3.3.6 相關係數（Correlation Coefficient） 29
第四章 實驗評估 30
4.1 實驗環境 30
4.2 實驗評估指標 30
4.2.1 專家判斷 30
4.2.2 混淆矩陣（Confusion Matrix） 30
4.2.3 準確率（Accuracy） 31
4.2.3 Kappa係數（Kappa Coefficient） 32
4.2.4 F1分數（F1-score） 32
4.2.5 AUC值（Area Under Curve） 33
4.2.6 三折交叉驗證（3-Fold Cross-Validation） 33
4.3 實驗結果 34
4.3.1 EEG通道注意力模塊加權性能評估 38
4.3.2 獨立樣本T檢定加權性能評估 39
第五章 結論與未來展望 40
5.1 結論 40
5.2 未來展望 40
參考文獻 41

Aggarwal, S., & Chugh, N. (2019). Signal processing techniques for motor imagery brain computer interface: A review. Array, 1, 100003.
Chen, J., Yu, Z., Gu, Z., & Li, Y. (2020). Deep temporal-spatial feature learning for motor imagery-based brain–computer interfaces. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 28(11), 2356-2366.
Duan, L., Li, J., Ji, H., Pang, Z., Zheng, X., Lu, R., ... & Zhuang, J. (2020). Zero-shot learning for EEG classification in motor imagery-based BCI system. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 28(11), 2411-2419.
George, O., Smith, R., Madiraju, P., Yahyasoltani, N., & Ahamed, S. I. (2021, July). Motor Imagery: A review of existing techniques, challenges and potentials. In 2021 IEEE 45th Annual Computers, Software, and Applications Conference (COMPSAC) (pp. 1893-1899). IEEE.
Hsu, W. Y., & Sun, Y. N. (2009). EEG-based motor imagery analysis using weighted wavelet transform features. Journal of neuroscience methods, 176(2), 310-318.
Hsu, W. Y., Lin, C. C., Ju, M. S., & Sun, Y. N. (2007). Wavelet-based fractal features with active segment selection: Application to single-trial EEG data. Journal of neuroscience methods, 163(1), 145-160.
Hu, J., Shen, L., & Sun, G. (2018). Squeeze-and-excitation networks. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 7132-7141).
Iftikhar, M., Khan, S. A., & Hassan, A. (2018, November). A survey of deep learning and traditional approaches for EEG signal processing and classification. In 2018 IEEE 9th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON) (pp. 395-400). IEEE.
Jasper, H. (1958). Report of the committee on methods of clinical examination in electroencephalography. Electroencephalogr Clin Neurophysiol, 10, 370-375.
Khosla, A., Khandnor, P., & Chand, T. (2020). A comparative analysis of signal processing and classification methods for different applications based on EEG signals. Biocybernetics and Biomedical Engineering, 40(2), 649-690.
Lahane, P., Jagtap, J., Inamdar, A., Karne, N., & Dev, R. (2019, February). A review of recent trends in EEG based Brain-Computer Interface. In 2019 International Conference on Computational Intelligence in Data Science (ICCIDS) (pp. 1-6). IEEE.
Li, Y., Guo, L., Liu, Y., Liu, J., & Meng, F. (2021). A temporal-spectral-based squeeze-and-excitation feature fusion network for motor imagery EEG decoding. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 29, 1534-1545.
Li, Y., Zhang, X. R., Zhang, B., Lei, M. Y., Cui, W. G., & Guo, Y. Z. (2019). A channel-projection mixed-scale convolutional neural network for motor imagery EEG decoding. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 27(6), 1170-1180.
Nath, D., Singh, M., Sethia, D., Kalra, D., & Indu, S. (2020, March). A comparative study of subject-dependent and subject-independent strategies for EEG-based emotion recognition using LSTM network. In Proceedings of the 2020 the 4th International Conference on Compute and Data Analysis (pp. 142-147).
Padfield, N., Zabalza, J., Zhao, H., Masero, V., & Ren, J. (2019). EEG-based brain-computer interfaces using motor-imagery: Techniques and challenges. Sensors, 19(6), 1423.
Papanastasiou, G., Drigas, A., Skianis, C., & Lytras, M. (2020). Brain computer interface based applications for training and rehabilitation of students with neurodevelopmental disorders. A literature review. Heliyon, 6(9), e04250.
Pfurtscheller, G., Neuper, C., Flotzinger, D., & Pregenzer, M. (1997). EEG-based discrimination between imagination of right and left hand movement. Electroencephalography and clinical Neurophysiology, 103(6), 642-651.
Schirrmeister, R. T., Springenberg, J. T., Fiederer, L. D. J., Glasstetter, M., Eggensperger, K., Tangermann, M., ... & Ball, T. (2017). Deep learning with convolutional neural networks for EEG decoding and visualization. Human brain mapping, 38(11), 5391-5420.
Soman, S., & Murthy, B. K. (2015). Using brain computer interface for synthesized speech communication for the physically disabled. Procedia Computer Science, 46, 292-298.
Xu, B., Zhang, L., Song, A., Wu, C., Li, W., Zhang, D., ... & Zeng, H. (2018). Wavelet transform time-frequency image and convolutional network-based motor imagery EEG classification. Ieee Access, 7, 6084-6093.
Zhang, Y., Nam, C. S., Zhou, G., Jin, J., Wang, X., & Cichocki, A. (2018). Temporally constrained sparse group spatial patterns for motor imagery BCI. IEEE transactions on cybernetics, 49(9), 3322-3332.
Zhao, X., Zhao, J., Cai, W., & Wu, S. (2019). Transferring common spatial filters with semi-supervised learning for zero-training motor imagery brain-computer interface. IEEE Access, 7, 58120-58130.