中風患者大多有運動功能障礙導致無法自主的運動，經由先前的研究發現，常人經過訓練可以調節其μ波，因此若能將人腦電腦介面和復健機器人結合，有可能發展新的復健方法以激發中風病患活動相依大腦重塑(activity-dependent brain plasticity)藉此增進其運動學習能力。
本研究修改前人之即時處理腦波方法並以本實驗室發展的復健機械手為基礎，設計一套雙手模式的新型復健機械手，將其結合成一套復健系統，並設計功能性磁振造影實驗以評估受測者在手指動作時大腦皮質神經元之活化；一位中風病患與一位常人參與本研究設計之復健系統長期訓練。中風患者兩側想像成功率最佳皆可達90%，患側最佳平均持續想像時間可於22.4秒完成拉伸手指的動作，常人受測者慣用側與非慣用側最佳想像成功率分別為90%與80%，非慣用側最佳平均持續想像時間可於21.6秒完成拉伸手指的動作；功能性磁振造影評估結果發現兩位受測者經長期訓練後大腦活化區域皆有減少且集中的趨勢，證實利用腦機介面結合復健機器人做神經復健訓練能造成人腦神經元活化之改變。

Stroke patients suffer from movement disabilities as the results of neurological injuries. Previous studies showed that healthy subjects can learn to regulate their mu waves. By integrating the EEG-based BCI and the rehabilitation robot, a new therapeutic method may be developed for enhancing the activity-dependent brain plasticity in stroke patients.
A real-time electroencephalogram(EEG) classification system and an orthotic hand, developed in our laboratory, were used to construct a new rehabilitation system. Functional magnetic resonance imaging(fMRI) experiments were set up to assess the therapeutic effect on the sensori-motor cortices. One stroke patient and one healthy subject were recruited. The patient was trained by the rehabilitation system for eight weeks and the healthy subject for four weeks. After training, the successful imagination rate of the stroke patient at both sides could reach 90%. The successful imagination rate of the healthy subject at dominant side could reach 90% and the non-dominant side for 80%. The maximum continuous imagination time of the affected side of the stroke patient and non-dominant side of the healthy subject could reach 22.4 sec and 21.6 sec, respectively. The results of fMRI show more focused activation on the motor area after BCI-based training. The results may provide an evidence of benefit of integrating the EEG-based BCI and rehabilitation robots.

[1]C. Starr, C.A. Evers and L. Starr(2006), “Biology:concepts and application”, Thomsom Brooks/Cole, Australia, pp. 498-507.

[2]G. Pfurtscheller (1977), “Graphical display and statistical evaluation of event-related desynchronization (ERD),” Electroencephalography and Clinical Neurophysiology, 43 pp. 757-760.

[3]G. Pfurtscheller and F.H. Lopes da Silva (1999), “Event-related EEG/MEG synchronization and desynchronization: basic principles,” Clinical Neurophysiology, 110 pp. 1842-1857.

[4]G. Pfurscheller and A. Berghold (1989), “Patterns of cortical activation during planning of voluntary movement,” Electroencephalography and Clinical Neurophysiology, 72 pp. 250-258.

[5]G. Pfurscheller, C. Brunner, A. Schlogl, F.H Lopes da Silva (2006), “Mu rhythm (de)synchronization and EEG single-trial classification of different motor imagery tasks,” NeuroImage, 31 pp. 153-159.

[6]J.R. Wolpaw, D.J. McFarland, G.W. Neat and C.A. Forneris (1991), “An EEG-based brain-computer interface for cursor control,” Electroencephalography and Clinical Neurophysiology, 103 pp. 252-259.

[7]J.R. Wolpaw and D.J. McFarland (2004), “Control of a two-dimensional movement signal by a noninvasive brain-computer interface in humans,” Proceedings of the National Academy of Sciences of the United States of America, 101 pp. 17849-17854.

[8]G. Pfurscheller, C. Guger, G. M. Ller et al. (2000), “Brain oscillations control hand orthosis in a tetraplegic,” Neuroscience Letters, 292 pp. 211-214.

[9]C. Brunner, Brendan Z. Allison, Dean J. Krusienski et al. (2010), “Improved signal processing approaches in an offline simulation of a hybrid brain–computer interface,” Neuroscience Methods, 188 pp. 165-173.

[10]黃志旭 (2000), 拇指被動運動對μ波的影響, 碩士論文, 國立成功大學機械工程學系.

[11]陳志瑋 (2002), 研究以小波神經網路做μ波即時鑑別, 碩士論文, 國立成功大學機械工程學系.

[12]陳志瑋 (2009), 以mu波為基礎之大腦電腦介面之實現與強化,博士論文, 國立成功大學機械工程學系.

[13]李昆祐 (2009), 腦波觸發控制機器人於中風病人膝髖關節復健之研究,碩士論文, 國立成功大學機械工程學系.

[14]呂俊霆 (2010), 發展以簡化希爾伯黃轉換為基礎之即時處理腦波訊號方法及其應用,碩士論文, 國立成功大學機械工程學系.

[15]B.H. Dobkin (2007), “Brain-computer interface technology as a tool to augment plasticity and outcomes for neurological rehabilitation,” Physiol, 579 (3) pp. 637-642.

[16]J.J. Daly, J.R. Wolpaw (2008), “Brain–computer interfaces in neurological rehabilitation,” Lancet Neurol, 7 pp. 1032-1043.

[17]M. Butz, F. Worgotter, A. van Ooyen (2009), “Activity-dependent structural plasticity,” Brain research reviews, 60 pp. 287-305.

[18]F. Pichiorri, F. De. Vico Fallani, F. Cincotti, F. Babiloni, M. Molinari, S.C. Kleih, C. Neuper, A. Kubler, D. Mattia (2011), “Sensorimotor rhythm-based brain-computer interface training: the impact on motor cortical responsiveness,” Neural Engineering, 8 pp. 25020-25028.

[19]P. Ritter, A. Villringer (2006), “Simultaneous EEG-fMRI,” Neuroscience and Biobehavioral Reviews, 30 pp. 823-838.

[20]P.J. van Hodt, J.C. de Munck, M. Zijlmans, G. Huiskamp, Frans S.S. Leijten, Paul A.J.M. Boon, Pauly .P.W. Ossenblok (2010), “Comparison of analytical strategies for EEG-correlated fMRI data in patients with epilepsy,” Magnetic Resonance Imaging, 28 pp. 1078-1086.

[21]T. Ueno, M. Inoue, T. Matsuoka, T. Abe, H. Maeda, K. Morita (2010), “ Comparison between a real sequential finger and imagery ovements: An fMRI Study revisited,” Brain Imaging and Behavior, 4 pp. 80-85.

[22]C.C.K. Lin, Y.N. Sun, C.I. Huang, C.Y. Yu, M.S Ju (2010), “Cortical activation by tactile stimulation to face and anterior neck areas: An fMRI Study with Three Analytic Methods,” Human Brain Mapping, 31 pp. 1876-1885.

[23]劉威呈 (2006), 模糊邏輯於功能性神經電刺激控制器之應用,碩士論文, 國立成功大學機械工程學系.

[24]Schaechter, JD; Kraft, E; Hilliard, TS, et al. (2002) “Motor recovery and cortical reorganization after constraint-induced movement therapy in stroke patients: A preliminary study”, J. Neurorehabilitation and Neural Repair, 16, pp. 326-338.

[25]Caria, A; Weber, C; Brotz, D, et al. (2011) “Chronic stroke recovery after combined BCI training and physiotherapy: A case report”, J. Psychophysiology, 48, pp. 578-582.