(3.235.25.169) 您好!臺灣時間:2021/04/20 03:37
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:李世章
研究生(外文):Shih-Zhang Li
論文名稱:基於核化切片逆回歸對神經元動作電位進行解碼
論文名稱(外文):Effective Dimension Reduction for Neural Decoding with Kernel Sliced Inverse Regression
指導教授:楊世宏楊世宏引用關係陳右穎陳右穎引用關係
指導教授(外文):Shih-Hung YangYou-Yin Chen
學位類別:碩士
校院名稱:國立陽明大學
系所名稱:生物醫學工程學系
學門:工程學門
學類:生醫工程學類
論文種類:學術論文
論文出版年:2016
畢業學年度:104
語文別:英文
論文頁數:38
中文關鍵詞:腦機介面核化切片逆回歸非線性降維神經訊號解碼
外文關鍵詞:brain machine interfacekernel sliced inverse regressionnonlinear effective dimension reductionneural decoding
相關次數:
  • 被引用被引用:0
  • 點閱點閱:71
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
近年來,神經科學的研究領域一直持續的熱絡發展,其中最重要的研究之一就是腦機介面。在腦機介面中,運動皮質區的神經群動作電位之解碼演算法更是其中的關鍵,它的作用是解析運動皮質區的神經群放電並且預測出生物體的運動軌跡進而去控制外在的神經義肢,而目前在這個領域中已經有許多的演算法被開發出來,甚至運用在臨床上。然而在這類的研究中,隨著電極的數量越來越多,資料的維度也越來越大。因此,我們想要找出一個方法,用小維度的神經訊號,也能夠成功解碼老鼠的前肢軌跡。
在本研究裡,我們設計了一個有關老鼠的行為實驗去驗證我們所選用的演算法,核化切片逆回歸(Kernel Sliced Inverse Regression)。核化切片逆回歸是利用腦部的神經訊號以及從影像中所擷取的軌跡來建立模型,並且跟傳統的線性切片逆回歸(Sliced Inverse Regression)以及主成分分析法 (Principal components analysis)做比較。從實驗結果看來,核化切片逆迴歸在預測軌跡的正確率,比切片逆迴歸以及主成分分析要好很多,用少量的神經訊號就能實現高正確率的預測效果。最後我們也會比較這三種方法的計算時間,核化切片逆迴歸的計算時間在我們系統的取樣時間以內,這點使得核化切片逆迴歸可以在即時系統的情況下運作。

Brain Machine Interface (BMI) was one of the most important research topics of neuroscience in recent years. Neural decoding algorithm was the key technique of BMI which decode neural activity into control signal for external device, such as computer cursors and prosthetic limbs. Many linear decoding algorithms have been developed, such as population vector algorithm and ordinary linear estimator, which have showed their ability in decoding movement. However, these algorithms required a large number of neurons to achieve desirable performance which may not feasible for the application with small number of neurons. Therefore, a decoder which could select important features and reduce dimension of inputs was desired. This study adopted kernel sliced inverse regression (kSIR) as the decoder due to its ability of nonlinear dimension reduction. The kSIR was employed to predict forelimb movement of a rat based on sorted single-unit activity from primary motor cortex in a water-reward lever pressing task. The proposed BMI was compared to the other feature selection algorithms and linear decoders. The results demonstrated that the kSIR performed higher prediction accuracy than the other methods where only a small number of neurons was available. Furthermore, the computation time of kSIR was less than sampling time which made the real-time BMI feasible.
Table of Contents
Acknowledgments i
摘要 ii
Abstract iii
Table of Contents iv
List of Tables vi
List of Figures vii
Chapter 1. Introduction 1
1.1 Brain Machine Interface 1
1.2 Literature Review 3
1.3 Motivations and Objectives 5
1.4 Thesis Organization 6
Chapter 2. Materials and Methods 7
2.1 Architecture of neural decoding of forelimb movement 7
2.2 In Vivo Animal Studies for the Neural Signal Acquisition 8
2.2.1 Animals 8
2.2.2 Surgical Techniques 9
2.2.3 Behavior Training 10
2.3 Data Preprocessing for Feature Extraction 12
2.4 Sliced Inverse Regression 15
2.4.1 Forward and Inverse Regression 15
2.4.2 An algorithm of SIR 17
2.4.3 PCA and SIR 20
2.5 Kernel Sliced Invers Regression 22
2.5.1 Kernel function 22
2.5.2 An algorithm of kSIR 23
2.6 Real-time interface Design 25
Chapter 3. Results 26
3.1 Effects of temporal order 26
3.2 Effects of nonlinear effective dimension reduction 28
3.3 Effects of training data size 31
3.4 Experimental Results 32
Chapter 4. Discussion 33
4.1 Influence of Temporal Order 33
4.2 Influence of Training Data Size and computing time 34
Chapter 5. Conclusion 35
References 36

List of Tables
Table 1. Original data 17
Table 2. Sorting by G and slicing 18
Table 3. G and SIR variates 19
Table 4. Computation time on temporal order 27
Table 5. Comparison of kSIR, SIR and PCA Computing time on data size 31

List of Figures
Figure 1. Schematic BMI setup in a person with a spinal-cord injury. 1
Figure 2. Schematic diagram depicts the BMI. 3
Figure 3. Schematic of the skilled hindlimb task. 4
Figure 4. System architecture of the brain machine interface system. 8
Figure 5. The flowchart of neural decoding process. 8
Figure 6. The locations of the electrodes in the primary motor cortex (red arrows). 10
Figure 7. Dimensions of the behavioral task box. 11
Figure 8. Two individual spikes were detected from the raw data. 12
Figure 9. A light-colored band(red ring) was tied around the right forelimb to be recognized by the video tracking system to measure the trajectory. 13
Figure 10. Multi-channel Acquisition Processor. 14
Figure 11. Illustration of the forelimb trajectory. 15
Figure 12. The general model. 16
Figure 13. SIR model 20
Figure 14. The basic concept of PCA 21
Figure 15. kernel of SIR model 23
Figure 16. User interface of brain machine interface system. 25
Figure 17. The effect of temporal order. 27
Figure 18. The effect after dimension reduction on the X and Y axis. 28
Figure 19. One trial amount of displacement, left and right individually represent the X and Y component. 29
Figure 20. One trial trajectory. Here, d is the dimension after reduction. 30
Figure 21. The experimental setup of the brain machine interface included: (A) Client of Real Time Interface (B) Arduino Mega 2560 (C) Five-axis Robotic Arm 32

References

Aflalo, T., Kellis, S., Klaes, C., Lee, B., Shi, Y., Pejsa, K., . . . Heck, C. (2015). Decoding motor imagery from the posterior parietal cortex of a tetraplegic human. Science, 348(6237), 906-910.
Atallah, H. E., Lopez-Paniagua, D., Rudy, J. W., & O'Reilly, R. C. (2007). Separate neural substrates for skill learning and performance in the ventral and dorsal striatum. Nature neuroscience, 10(1), 126-131.
Ball, T., Kern, M., Mutschler, I., Aertsen, A., & Schulze-Bonhage, A. (2009). Signal quality of simultaneously recorded invasive and non-invasive EEG. Neuroimage, 46(3), 708-716.
Balleine, B. W., Killcross, A. S., & Dickinson, A. (2003). The effect of lesions of the basolateral amygdala on instrumental conditioning. The Journal of neuroscience, 23(2), 666-675.
Berger, T. W., Chapin, J. K., Gerhardt, G. A., McFarland, D. J., Principe, J. C., Soussou, W. V., . . . Tresco, P. A. (2008). Brain-Computer Interfaces: An international assessment of research and development trends: Springer Science & Business Media.
Chang, Y.-W., Hsieh, C.-J., Chang, K.-W., Ringgaard, M., & Lin, C.-J. (2010). Training and testing low-degree polynomial data mappings via linear SVM. The Journal of Machine Learning Research, 11, 1471-1490.
Chapin, J. K., Moxon, K. A., Markowitz, R. S., & Nicolelis, M. A. (1999). Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex. Nature neuroscience, 2(7), 664-670.
Collinger, J. L., Wodlinger, B., Downey, J. E., Wang, W., Tyler-Kabara, E. C., Weber, D. J., . . . Schwartz, A. B. (2013). High-performance neuroprosthetic control by an individual with tetraplegia. The Lancet, 381(9866), 557-564.
Dhamala, M., Jirsa, V. K., & Ding, M. (2004). Enhancement of neural synchrony by time delay. Physical Review Letters, 92(7), 074104.
Franke, F., Jäckel, D., Dragas, J., Müller, J., Radivojevic, M., Bakkum, D., & Hierlemann, A. (2014). High-density microelectrode array recordings and real-time spike sorting for closed-loop experiments: an emerging technology to study neural plasticity. Closing the Loop Around Neural Systems, 31.
Hochberg, L. R., Serruya, M. D., Friehs, G. M., Mukand, J. A., Saleh, M., Caplan, A. H., . . . Donoghue, J. P. (2006). Neuronal ensemble control of prosthetic devices by a human with tetraplegia. Nature, 442(7099), 164-171.
Jensen, W., & Rousche, P. J. (2005). Movement discrimination based on rat primary motor cortex responses. Paper presented at the Neural Engineering, 2005. Conference Proceedings. 2nd International IEEE EMBS Conference on.
König, H. (2013). Eigenvalue distribution of compact operators (Vol. 16): Birkhäuser.
Lahr, J., Schwartz, C., Heimbach, B., Aertsen, A., Rickert, J., & Ball, T. (2015). Invasive brain–machine interfaces: a survey of paralyzed patients’ attitudes, knowledge and methods of information retrieval. Journal of neural engineering, 12(4), 043001.
Lai, H.-Y., Chen, Y.-Y., Lin, S.-H., Lo, Y.-C., Tsang, S., Chen, S.-Y., . . . Wu, R. (2011). Automatic spike sorting for extracellular electrophysiological recording using unsupervised single linkage clustering based on grey relational analysis. Journal of neural engineering, 8(3), 036003.
Lebedev, M. A., & Nicolelis, M. A. (2006). Brain–machine interfaces: past, present and future. TRENDS in Neurosciences, 29(9), 536-546.
Li, K.-C. (1991). Sliced inverse regression for dimension reduction. Journal of the American Statistical Association, 86(414), 316-327.
Li, K.-C. (2000). High dimensional data analysis via the SIR/PHD approach.
Linderman, M. D., Santhanam, G., Kemere, C. T., Gilja, V., O'Driscoll, S., Yu, B. M., . . . Meng, T. H. (2008). Signal processing challenges for neural prostheses. Signal Processing Magazine, IEEE, 25(1), 18-28.
Mercer, J. (1909). Functions of positive and negative type, and their connection with the theory of integral equations. Philosophical transactions of the royal society of London. Series A, containing papers of a mathematical or physical character, 209, 415-446.
Narayanan, N. S., Kimchi, E. Y., & Laubach, M. (2005). Redundancy and synergy of neuronal ensembles in motor cortex. The Journal of neuroscience, 25(17), 4207-4216.
Schwartz, A. B., & Moran, D. W. (1999). Motor cortical activity during drawing movements: population representation during lemniscate tracing. Journal of Neurophysiology, 82(5), 2705-2718.
Shenoy, K. V., & Carmena, J. M. (2014). Combining decoder design and neural adaptation in brain-machine interfaces. Neuron, 84(4), 665-680.
Taylor, D. M., Tillery, S. I. H., & Schwartz, A. B. (2002). Direct cortical control of 3D neuroprosthetic devices. Science, 296(5574), 1829-1832.
Velliste, M., Perel, S., Spalding, M. C., Whitford, A. S., & Schwartz, A. B. (2008). Cortical control of a prosthetic arm for self-feeding. Nature, 453(7198), 1098-1101.
Wessberg, J., Stambaugh, C. R., Kralik, J. D., Beck, P. D., Laubach, M., Chapin, J. K., . . . Nicolelis, M. A. (2000). Real-time prediction of hand trajectory by ensembles of cortical neurons in primates. Nature, 408(6810), 361-365.
Wood, F., Fellows, M., Donoghue, J., & Black, M. (2004). Automatic spike sorting for neural decoding. Paper presented at the Engineering in Medicine and Biology Society, 2004. IEMBS'04. 26th Annual International Conference of the IEEE.
WU, Q. (2007). Regularized sliced inverse regression for kernel models. Duke University, Durham.
Wu, W., Gao, Y., Bienenstock, E., Donoghue, J. P., & Black, M. J. (2006). Bayesian population decoding of motor cortical activity using a Kalman filter. Neural computation, 18(1), 80-118.
Yu, Y., Zhang, S.-m., Zhang, H.-j., Liu, X.-c., Zhang, Q.-s., Zheng, X.-x., & Dai, J.-h. (2010). Neural decoding based on probabilistic neural network. Journal of Zhejiang University Science B, 11(4), 298-306.


連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔