臺灣博碩士論文加值系統

　　嚴重殘疾的患者在生活上會遭遇許多的問題，像是表達與行動時所遭遇的問題，且在使用傳統輔具上有相當的難度。雖然現在有許多科技應用在分析人類大腦的生物訊號例如:腦機介面(brain-computer interface ,BCI)，相關的研究上已經有相當的發展，其中以穩態視覺誘發電位(steady-state visual evoked potentials, SSVEP)的準確度較佳，透過視覺刺激來誘發腦部反射的方式能夠使善嚴重殘疾患者透過腦機介面與外界溝通，因此能有效改善該患者的生活方式。
　　在本論文中所開發的模組化受限玻爾茲曼機(modular restricted Boltzmann machine , MRBM)，能藉由多個識別層用於提取多種不同統計與頻譜計算方法之間的特性，最後連接一層決策層將多個參數特性整合並決策，因此該方法能有效判別腦部反射之訊號。首先使用典型相關分析(canonical correlation analysis, CCA)來計算誘發電位之時域相關性，接下來使用快速傅立葉轉換(fast Fourier transform, FFT)將誘發電位轉至頻域，透過設置窗函數的方式能有效提取目標頻率之特徵，最後使用同調性分析(Magnitude squared coherence, MSC)來計算誘發電位之頻域相關性，接下來將三種特徵提取方法分別經由三個做為識別層之受限玻爾茲曼機進行參數特性之提取，再連接一層決策層之受限玻爾茲曼機整合三個識別層之輸入進行決策與整合，最終用於判別穩態視覺誘發電位所提取之腦電訊號。
　　實驗結果表明本研究所開發之方法能夠有效提升穩態視覺誘發電位之辨識率與減少運算複雜度，未來期望將該系統應用於控制輔具或拼寫器來幫助與改善嚴重殘疾患者的生活。

Many patients with severe disabilities have many problems in their lives, such as inconvenience during expression and action, and it is quite difficult to use traditional assistive devices. Although there are many science and technology applications in the analysis of the human brain’s biological signals such as the Brain-Computer Interface (BCI)that there has been considerable development in related research, the accuracy of identifying brain signals is not ideal. This paper uses different statistical and spectral calculation methods combined with modules to improve the accuracy of recognizing brain signals, and ultimately apply to the relevant auxiliary equipment for patients with severe disabilities, thereby improving the quality of life of related patients.
The modular restricted Boltzmann machine (MRBM) designed in this dissertation is designed to extract the characteristics of many different input parameters through multiple identification layers and a layer of decision-making layer is connected to integrate the multiple parameter features in the end. Firstly, canonical correlation analysis (CCA) was used to calculate the temporal correlation of steady-state visual evoked potentials (SSVEP). Second, the fast Fourier transform (FFT) was used to transfer the steady-state visual evoked potentials to the frequency domain. The window function is used to effectively extract the characteristics of the target frequency. Thirdly, the frequency domain correlation of the steady-state visual evoked potentials is calculated to use the magnitude squared coherence (MSC). According the each types of features, the corresponding RBM is constructed and used to identify the decision results. With these results, the decision RBM is adopted to fuse the decision detected by using different types of features and then a fused decision result can be obtained.
The Restricted Boltzmann machine of the identification layer extracts the parameter characteristics. Finally, a constrained Boltzmann machine connected to a decision-making layer integrates the input of the three identification layers for decision-making and integration, and is ultimately used to determine the steady state visual evoked potentials.
Keyword: modular, restricted Boltzmann machine, Brain-Computer Interface, steady-state visual evoked potentials, EEG

中文摘要 iii
目錄 vii
表目錄 ix
圖目錄 x
Chapter 1 簡介 1
1.1 研究動機 1
1.2 文獻探討 2
1.2.1 疾病文獻探討 2
1.2.2 腦機介面方法探討 3
1.2.3 特徵提取方法探討 5
1.2.4 類神經網路方法探討 6
1.3 問題與目標 8
1.4 論文架構 9
Chapter 2 研究方法 10
2.1 穩態視覺誘發電位資料收集 10
2.2 特徵參數提取 12
2.1.1 典型相關分析 12
2.1.2 快速傅立葉轉換 13
2.1.3 同調性分析 13
2.3 模組化受限波爾茲曼機 14
Chapter 3 實驗結果與討論 19
3.1 輸入參數評估 19
3.2 受限波爾茲曼機效能評估 20
3.2.1 單一參數之評估 21
3.2.2 複數參數之評估 22
3.3 模組化受限波爾茲曼機效能評估 25
3.3.1 模組化效能評估 25
3.3.2 運算效能評估 27
3.4 其他識別模組比較 29
Chapter 4 結論 30
參考文獻 31

[1] Allan H. Ropper, Martin A. Samuels (2009). Adams and Victor’s Principles of Neurology, U.S.A, McGraw-Hill Medical.
[2] M.S. Hawley, S.P. Cunningham, P.D. Green, P. Enderby, R. Palmer, S. Sehgal, and P. O'Neill, (2013). A Voice-Input Voice-Output Communication Aid for People with Severe Speech Impairment.
[3] D. Beukelman, P. Mirenda,(2005). Augmentative and Alternative Communication.
[4] Omid Dehzangi, Muhamed Farooq. (2017). Subject-Dependent SSVEP Identification Using GMM Training and Adaptation, 16th IEEE International Conference on Machine Learning and Applications, Mexico Cancun.
[5] Jesu ́s Sagaho ́n-Azu ́a, Blanca Tovar-Corona, Jesu ́s A. Zu ́n ̃iga-Valladares. (2017). Comparison between the Canonical Correlation Analysis and the Support Vector Machines as classification algorithms in an SSVEP-based Brain-Computer Interface, 14th International Conference on Electrical Engineering, Computing Science and Automatic Control (CCE).
[6]边肇祺 (2000)，模式識別。
[7] Kelly, Evelyn B. (2013). Encyclopedia of human genetics and disease.
[8] Phukan J, Pender NP, Hardiman O. (2007). Cognitive impairment in amyotrophic lateral sclerosis.
[9] P N Leigh, K Ray-Chaudhuri .(1994). Motor neuron disease(p886~p896).
[10] National Institute of Neurological Disorders and Stroke (NINDS) .(2010). Motor Neuron Diseases Fact Sheet.
[11] Kiernan, MC; Vucic, S; Cheah, BC; Turner, MR; Eisen, A; Hardiman, O; Burrell, JR; Zoing, MC. (2011) Amyotrophic lateral sclerosis.
[12] Alan E.H. Emery (1991). Population frequencies of inherited neuromuscular diseases—A world survey
[13] World Health Organization. (2001).International classification of functioning, disability and health.
[14] Matthew W. Brault. (2012). Americans With Disabilities: 2010.
[15] Ryan Spicer*, Julia Anglin^†, David M. Krum^‡, and Sook-Lei Liew^§. REINVENT: A Low-Cost, Virtual Reality Brain-Computer Interface for Severe Stroke Upper Limb Motor Recovery, University of Southern California, U.S.A, State of California.
[16] KUBACKI, JAKUBOWSKI, RYBARCZYK. (2017). Research on possibilities of transporter movement using brain-computer interface based on Steady-State Visually Evoked Potential (SSVEP), Institute of Mechanical Technology Poznan University of Technology, Poland, Poznań.
[17] Gert Pfurtschellera , Christa Neuperb ,(1997). Motor imagery activates primary sensorimotor area in humans.
[18] G. Pfurtscheller ; C. Neuper ,(2001).Motor imagery and direct brain-computer communication.
[19] Picton, T. W., Bentin, S., Berg, P., Donchin, E., Hillyard, S. A., Johnson, R., et al. (2000). Guidelines for using human event-related potentials to study cognition: Recording standards and publication criteria. Psychophysiology,
[20] Jonathan Sherwood .(2000). Give it a thought - and make it so.
[21] Ding J, Sperling G, Srinivasan R ,(2006). Attentional modulation of SSVEP power depends on the network tagged by the flicker frequency.
[22] Danhua Zhu, Jordi Bieger, Gary Garcia Molina, Ronald M. Aarts.(2010). A Survey of Stimulation Methods Used in SSVEP-Based BCIs.
[23] Zhonglin Lin , Changshui Zhang , Wei Wu , Xiaorong Gao ,(2007). Frequency recognition based on canonical correlation analysis for SSVEP-based BCIs.
[24] Kuo-Kai Shyu , Po-Lei Lee , Yu-Ju Liu , Jyun-Jie Sie ,(2010). Dual-frequency steady-state visual evoked potential for brain computer interface.
[25] Zimu Zhang , Xiuquan Li , Zhidong Deng ,(2010). A CWT-based SSVEP classification method for brain-computer interface system.
[26] Dongsoo Kim , Vance Zemon , Alice Saperstein , Pamela D. Butler , Daniel C. Javitt ,(2004).Dysfunction of early-stage visual processing in schizophrenia: harmonic analysis.
[27] Xiu An, Deping Kuang , Xiaojiao Guo, Yilu Zhao , Lianghua He (2014).A Deep Learning Method for Classification of EEG Data Based on Motor Imagery.
[28] Yaguang Jia ,Jun Xie ; Guanghua Xu , Min Li , Sicong Zhang , Ailing Luo , Xingliang Han ,(2017). A separated feature learning based DBN structure for classification of SSMVEP signals.
[29] Vitoantonio Bevilacqua, Giacomo Tattoli, Domenico Buongiorno.(2014). A novel BCI-SSVEP based approach for control of walking in Virtual Environment using a Convolutional Neural Network.
[30] No-Sang Kwak, Klaus-Robert Müller, Seong-Whan Lee.(2017). A convolutional neural network for steady state visual evoked potential classification under ambulatory environment.
[31] Irshad Ahmad Ansari, Rajesh Singla, Munendra Singh,(2014). SSVEP and ANN based optimal speller design for Brain Computer Interface.
[32] R Singla, BA Haseena ,(2014).Comparison of ssvep signal classification techniques using svm and ann models for bci applications.
[33] Omid Dehzangi , Muhamed Farooq ,(2017). Subject-Dependent SSVEP Identification Using GMM Training and Adaptation.
[34] S. Pouryazdian and A. Erfanian ,(2009).Detection of steady-state visual evoked potentials for brain-computer interfaces using PCA and high-order statistics.