臺灣博碩士論文加值系統

本研究介紹以一種基於即時性腦機介面(BCI)控制下肢復健外骨骼控制系統。
其中本研究以站立和步態之運動想像(Motor Imagery, MI)任務期間自行誘發的腦電圖 (Electroencephalography , EEG) 訊號，通過共同空間模式(CSP)、功率譜密度(PSD)、離散小波轉換(DWT)取特定之 Mu 頻段(8-12Hz)之訊號結合自回歸模型(AR) 等方法取特徵，特徵被解碼並即時轉換為下肢復健外骨骼(LL)的控制命令，通過支持向量機(SVM)和線性鑑別分析(LDA)進行精度驗證，驅動下肢復健外骨骼系統作相對應命令動作軌跡，以達到即時腦波訊號驅動下肢復健外骨骼控制。
結果顯示，確定了基於 MI 的 EEG 在 BCI 應用在運動命令控制方面的潛力。患者可以從其進步中受益，例如使用腦機介面系統搭配可穿戴下肢復健外骨骼、矯正器、假肢、輪椅和輔助機器人等設備，可幫助或康復行走障礙者做復健，增進行動康復、行動替代、假肢控制與輔助行動等益處。

Abstract—This study introduces a real-time brain-computer interface (BCI)-based control system for lower limb rehabilitation exoskeleton. In this study, the self-evoked electroencephalography (EEG) signals during standing and gait motor imagery (MI) tasks were obtained by common spatial pattern (CSP), power spectral density (PSD), discrete wavelet transforms (DWT) for specific Mu band (8-12 Hz) signals, and combined with auto regressive model (AR). The features are decoded and converted to control commands of the lower limb rehabilitation exoskeleton (LL) in real time, and the accuracy is verified by support vector machine (SVM) and linear discriminant analysis (LDA) to drive the lower limb rehabilitation exoskeleton system to make corresponding command trajectories to achieve real-time brainwave signal-driven lower limb rehabilitation exoskeleton control.
The results showed that the potential of MI-based EEG for BCI applications in motor command control was confirmed. Patients can benefit from its advancements, such as the use of brain-machine interface systems with wearable lower extremity rehabilitation exoskeletons, orthoses, prostheses, wheelchairs and assistive robots for rehabilitation to enhance mobility rehabilitation, mobility replacement, prosthetic control and mobility assistance.

摘要 ii
ABSTRACT iii
致謝 iv
目錄 v
表目錄 ix
圖目錄 xi
1 第一章 緒論 1
1.1 前言 1
1.2 文獻回顧 2
1.2.1 大腦結構 2
1.2.2 腦電圖 3
1.2.3 腦機介面 7
1.2.4 運動想像 10
1.2.5 腦機介面結合復健機系統 13
1.3 研究動機 18
1.4 論文架構 18
2 第二章 實驗設備與系統架構 19
2.1 腦波系統架構與設備介紹 19
2.1.1 腦波訊號擷取器 19
2.1.2 腦波系統環境架設 20
2.1.3 電極位置 21
2.1.4 軟體介面 22
2.1.5 操作流程 22
2.1.6 試驗流程設計(Trial) 26
2.1.7 資料擷取結構 27
2.1.8 實驗架構 29
2.1.9 實驗流程 31
2.2 復健機系統架構與設備介紹 33
2.2.1 復健機系統架構 33
2.2.2 復健機機構 34
2.2.3 懸吊馬達系統 37
2.2.4 跑步機驅動系統 38
2.2.5 LabVIEW復健機介面 39
2.2.6 肌電訊號介紹 43
3 第三章 研究方法與實驗分析 44
3.1 腦波訊號前處理 44
3.1.1 IIR Butterworth 帶通濾波器 44
3.2 特徵選擇 46
3.2.1 共同空間模式(Common Spatial Paternal ,CSP) 46
3.2.2 小波轉換(Wavelet transform, WT) 48
3.2.3 功率譜密度（Power Spectral Density, PSD） 52
3.2.4 自我回歸模型 (Auto Regressive Model, AR) 53
3.3 分類與驗證準確率 54
3.3.1 支持向量機(Support Vector Machine, SVM) 54
3.3.2 線性鑑別分析(Linear Discriminant Analysis ,LDA) 55
3.3.3 ROC曲線（Receiver operating characteristic curve） 56
3.3.4 K折交叉驗證(K-fold cross-validation) 57
3.4 EEG腦波訊號特徵離線分類分析 58
3.4.1 三種資料數據結構 58
3.4.2 三種資料結構訓練準確率 59
3.4.3 個人數據訓練混合測試準確率 68
3.4.4 多次實驗資料訓練準確率 72
3.5 Baseline方法提升準確率 75
3.6 EMG肌電訊號特徵離線分類分析 79
3.6.1 腿部肌肉 79
3.6.2 電極貼片位置 80
3.6.3 肌電訊號特徵值 81
3.6.4 肌電訊號分類 82
4 第四章 實驗結果與分析 84
4.1 實驗總架構 84
4.2 下肢復健機程式撰寫 85
4.2.1 復健機系統 85
4.2.2 跑步機系統 87
4.2.3 TCP接收EEG/EMG控制命令 88
4.2.4 主要驅動程式 89
4.3 EEG腦波訊號驅動下肢復健機實驗 91
4.3.1 腦機介面結合復健機系統總架構 91
4.3.2 實驗模式 92
4.3.3 線上測試準確率 96
4.4 EMG肌電訊號驅動下肢復健機實驗 100
4.4.1 EMG驅動機構流程圖 100
5 第五章 結論與未來展望 103
5.1 實驗結論 103
5.2 未來展望 104
6 文獻參考 105

[1]N. Padfield, J. Zabalza, H. Zhao, V. Masero, and J. Ren, "EEG-based brain-computer interfaces using motor-imagery: Techniques and challenges," Sensors, vol. 19, no. 6, p. 1423, 2019.
[2]陳怡岑 and 陳永昇, "運動想像腦電波之連續式辨別," 碩士, 資訊工程學系碩士班, 國立交通大學, 新竹, 2004.
[3]廖倍瑜 and 林灶生, "植基於 SSVEP 之腦機介面電話撥打系統," 2013.
[4]王俊偉, "使用最大對比光束集成濾波器之腦活動多類動作想像腦機介面系統," 2017.
[5]陳柏中. "腦機介面趨勢發展分析." https://portal.stpi.narl.org.tw/index/article/10227 (accessed.
[6]徐. S.-H. S. Hsu. "腦機介面 (Brain-Computer Interface) 專題 (上)." https://investigator.tw/1226/%E8%85%A6%E6%A9%9F%E4%BB%8B%E9%9D%A2-brain-computer-interface-%E5%B0%88%E9%A1%8C-%E4%B8%8A/ (accessed.
[7]盧士暐, "使用運動想像腦電波之適應性腦機介面," 碩士, 資訊科學與工程研究所, 國立交通大學, 新竹市, 2007. [Online]. Available: https://hdl.handle.net/11296/gb45bk
[8]"脊髓小腦共濟失調 ─SCAs的遺傳及症狀." https://highscope.ch.ntu.edu.tw/wordpress/?p=24616 (accessed.
[9]"脊髓損傷 / Spinal Cord Injury." https://www.christopherreeve.org/zh/international/top-paralysis-topics-in-chinese/spinal-cord-injury (accessed.
[10]"布羅德曼分區系統." https://zh.wikipedia.org/zh-tw/%E5%B8%83%E7%BD%97%E5%BE%B7%E6%9B%BC%E5%88%86%E5%8C%BA%E7%B3%BB%E7%BB%9F (accessed.
[11]方振隆, "以腦波控制之主動式義手," 碩士, 機械工程學系碩博士班, 國立成功大學, 台南市, 2004. [Online]. Available: https://hdl.handle.net/11296/e25367
[12]"3D人體大透視／你懂大腦多少？." https://health.udn.com/health/story/6008/357634 (accessed.
[13]"大腦結構圖." https://blog.xuite.net/fong551/twblog/124597583 (accessed.
[14]"大腦區域功能圖." http://www.liteoncf.org.tw/dm/136/136.html (accessed.
[15]"大腦運動區域." http://fms.hsc.edu.tw/sysdata/doc/c/c81927d832c398ce/pdf.pdf (accessed.
[16]陳志瑋, "研究以小波神經網路作mu波即時鑑別," 碩士, 機械工程學系碩博士班, 國立成功大學, 台南市, 2002. [Online]. Available: https://hdl.handle.net/11296/8n79e4
[17]湯雅雯, "腦波量測系統之研製與腦波信號之非線性分析," 碩士, 電機工程學系碩博士班, 國立成功大學, 台南市, 2005. [Online]. Available: https://hdl.handle.net/11296/j52g37
[18]"神經元." https://www.ehanlin.com.tw/keywordPool/wordPage.html?key=%E7%A5%9E%E7%B6%93%E5%85%83&subject=H-BI (accessed.
[19]"腦波." https://zh.m.wikipedia.org/zh-tw/%E8%85%A6%E6%B3%A2 (accessed.
[20]黃育瑜, "基於獨立EEG成分的運動執行、運動想像和運動觀察的源識別及比較乾/濕電極的獨立成分差異," 碩士, 認知與神經科學研究所, 國立中央大學, 桃園縣, 2019. [Online]. Available: https://hdl.handle.net/11296/c5wwd2
[21]林三永. "10-20系統和乾式電極." https://sa.ylib.com/MagArticle.aspx?id=1821 (accessed.
[22]"人機介面." https://en.wikipedia.org/wiki/Brain%E2%80%93computer_interface (accessed.
[23]徐. S.-H. S. Hsu. "腦機介面 (Brain-Computer Interface) 專題 (下)." https://investigator.tw/1237/%E8%85%A6%E6%A9%9F%E4%BB%8B%E9%9D%A2-brain-computer-interface-%E5%B0%88%E9%A1%8C-%E4%B8%8B/ (accessed.
[24]"Vision Quest." https://www.wired.com/2002/09/vision/ (accessed.
[25]G. Deuschl et al., "A Randomized Trial of Deep-Brain Stimulation for Parkinson's Disease," New England Journal of Medicine, vol. 355, no. 9, pp. 896-908, 2006/08/31 2006, doi: 10.1056/NEJMoa060281.
[26]M. Vidailhet et al., "Bilateral Deep-Brain Stimulation of the Globus Pallidus in Primary Generalized Dystonia," New England Journal of Medicine, vol. 352, no. 5, pp. 459-467, 2005/02/03 2005, doi: 10.1056/NEJMoa042187.
[27]P. Boon et al., "Deep Brain Stimulation in Patients with Refractory Temporal Lobe Epilepsy," Epilepsia, vol. 48, no. 8, pp. 1551-1560, 2007, doi: 10.1111/j.1528-1167.2007.01005.x.
[28]K. M. Szostak, L. Grand, and T. G. Constandinou, "Neural interfaces for intracortical recording: Requirements, fabrication methods, and characteristics," Frontiers in Neuroscience, vol. 11, p. 665, 2017.
[29]施明志, "應用於人腦電腦介面之主動式手部輔具," 碩士, 機械工程學系碩博士班, 國立成功大學, 台南市, 2006. [Online]. Available: https://hdl.handle.net/11296/nx3cc5
[30]黃繪禎, "EEG腦機介面控制肩肘機器人於中風病患復健研究," 碩士, 機械工程學系碩博士班, 國立成功大學, 台南市, 2012. [Online]. Available: https://hdl.handle.net/11296/9q9pzr
[31]"心理神經肌肉理論." http://www.tswong.net/hkpe/running/psychological_aspects/psychoneuromuscular_theory.htm (accessed.
[32]Z. Tang, S. Sun, S. Zhang, Y. Chen, C. Li, and S. Chen, "A Brain-Machine Interface Based on ERD/ERS for an Upper-Limb Exoskeleton Control," (in eng), Sensors (Basel), vol. 16, no. 12, p. 2050, 2016, doi: 10.3390/s16122050.
[33]M. Tariq, P. M. Trivailo, and M. Simic, "Detection of knee motor imagery by Mu ERD/ERS quantification for BCI based neurorehabilitation applications," in 2017 11th Asian Control Conference (ASCC), 17-20 Dec. 2017 2017, pp. 2215-2219, doi: 10.1109/ASCC.2017.8287519.
[34]Y. Jiang, N. T. Hau, and W. Y. Chung, "Semiasynchronous BCI Using Wearable Two-Channel EEG," IEEE Transactions on Cognitive and Developmental Systems, vol. 10, no. 3, pp. 681-686, 2018, doi: 10.1109/TCDS.2017.2716973.
[35]C. Lindig-León, S. Rimbert, and L. Bougrain, "Multiclass Classification Based on Combined Motor Imageries," (in English), Frontiers in Neuroscience, Original Research vol. 14, 2020-November-19 2020, doi: 10.3389/fnins.2020.559858.
[36]N. A. Grigorev et al., "A BCI-Based Vibrotactile Neurofeedback Training Improves Motor Cortical Excitability During Motor Imagery," IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 29, pp. 1583-1592, 2021, doi: 10.1109/TNSRE.2021.3102304.
[37]G. Zeilig, H. Weingarden, M. Zwecker, I. Dudkiewicz, A. Bloch, and A. Esquenazi, "Safety and tolerance of the ReWalk™ exoskeleton suit for ambulation by people with complete spinal cord injury: a pilot study," (in eng), J Spinal Cord Med, vol. 35, no. 2, pp. 96-101, Mar 2012, doi: 10.1179/2045772312y.0000000003.
[38]G. Colombo, M. Joerg, R. Schreier, and V. Dietz, "Treadmill training of paraplegic patients using a robotic orthosis," (in eng), J Rehabil Res Dev, vol. 37, no. 6, pp. 693-700, Nov-Dec 2000.
[39]M. Tariq, P. M. Trivailo, and M. Simic, "EEG-based BCI control schemes for lower-limb assistive-robots," Frontiers in human neuroscience, vol. 12, p. 312, 2018.
[40]"為中風病患開啟復健新路　成大工學院與成醫合作開發腦機介面復健機器人." https://news-secr.ncku.edu.tw/p/404-1037-68211.php (accessed.
[41]G. Yu, J. Wang, W. Chen, and J. Zhang, "EEG-based brain-controlled lower extremity exoskeleton rehabilitation robot," in 2017 IEEE International Conference on Cybernetics and Intelligent Systems (CIS) and IEEE Conference on Robotics, Automation and Mechatronics (RAM), 2017: IEEE, pp. 763-767.
[42]X. Long, D.-X. Liu, S. Liang, Z. Yan, and X. Wu, "An eeg-based bci system for controlling lower exoskeleton to step over obstacles in realistic walking situation," in 2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV), 2018: IEEE, pp. 1609-1614.
[43]J. Choi and H. Kim, "Real-time Decoding of EEG Gait Intention for Controlling a Lower-limb Exoskeleton System," in 2019 7th International Winter Conference on Brain-Computer Interface (BCI), 18-20 Feb. 2019 2019, pp. 1-3, doi: 10.1109/IWW-BCI.2019.8737311.
[44]S. Y. Gordleeva et al., "Real-Time EEG–EMG Human–Machine Interface-Based Control System for a Lower-Limb Exoskeleton," IEEE Access, vol. 8, pp. 84070-84081, 2020, doi: 10.1109/ACCESS.2020.2991812.
[45]J. Choi, K. T. Kim, J. Lee, S. J. Lee, and H. Kim, "Robust Semi-synchronous BCI Controller for Brain-Actuated Exoskeleton System," in 2020 8th International Winter Conference on Brain-Computer Interface (BCI), 26-28 Feb. 2020 2020, pp. 1-3, doi: 10.1109/BCI48061.2020.9061658.
[46]黃浩權, "基於多通道濾波共同空間模式之憂鬱症腦波分析," 碩士, 機電整合研究所, 國立臺北科技大學, 台北市, 2017. [Online]. Available: https://hdl.handle.net/11296/9x26c7
[47]"自迴歸模型." https://zh-yue.wikipedia.org/wiki/%E8%87%AA%E8%BF%B4%E6%AD%B8%E6%A8%A1%E5%9E%8B (accessed.
[48]"迴歸分析." https://zh.wikipedia.org/zh-tw/%E8%BF%B4%E6%AD%B8%E5%88%86%E6%9E%90 (accessed.
[49]"支持向量機." https://zh.wikipedia.org/zh-tw/%E6%94%AF%E6%8C%81%E5%90%91%E9%87%8F%E6%9C%BAhttps://zh.wikipedia.org/zh-tw/%E6%94%AF%E6%8C%81%E5%90%91%E9%87%8F%E6%9C%BA (accessed.
[50]"線性判別分析." https://zh.wikipedia.org/zh-tw/%E7%B7%9A%E6%80%A7%E5%88%A4%E5%88%A5%E5%88%86%E6%9E%90 (accessed.