نوع مقاله : مقاله کامل پژوهشی

نویسندگان

1 دانشجوی دکتری مهندسی پزشکی، گروه مهندسی پزشکی، دانشکده‌ی مهندسی برق، دانشگاه علم و صنعت ایران، تهران، ایران

2 استاد، گروه مهندسی پزشکی، دانشکده‌ی مهندسی برق، دانشگاه علم و صنعت ایران، تهران، ایران

چکیده

کنترل پروتز عصبی به منظور بازیابی عمل‌کرد دست افراد مبتلا به فلج اندام­های فوقانی، یکی از کاربردهای مهم سیستم­های BCI می‌باشد. توانایی گرفتن اجسام، از ابتدایی­ترین نیازها برای انجام کارهای روزانه بوده و از این رو برای عمل‌کرد صحیح پروتز عصبی، لازم است تا کاربر بتواند مقدار نیروی لازم برای گرفتن اجسام را کنترل کند. به همین دلیل افزایش دقت رمزگشایی پیوسته‌ی نیرو موضوعی مهم برای عمل‌کرد صحیح این نوع سیستم­های BCI می­باشد. در اغلب پژوهش­های صورت گرفته در زمینه‌ی رمزگشایی نیرو از مدل­های خطی مانند فیلتر وینر، فیلتر کالمن و PLS استفاده شده و تا کنون تاثیر استفاده از مدل­های غیرخطی بر دقت رمزگشایی نیرو مورد بررسی قرار نگرفته است. هدف این پژوهش، بررسی تاثیر استفاده از مدل­های رگرسیون غیرخطی مبتنی بر توابع کرنل بر دقت رمزگشایی نیروی دست موش صحرایی با استفاده از سیگنال­های پتانسیل میدانی محلی می­باشد. بدین منظور روش­های رگرسیون ستیغی، PCR و PLS در نظر گرفته شده و با استفاده از تابع کرنل گوسی، از تعمیم­یافته‌ی غیرخطی آن‌ها برای تخمین پیوسته‌ی نیرو بهره گرفته شده است. ارزیابی و مقایسه‌ی روش­های رگرسیون ستیغی کرنلی، PCR کرنلی و PLS کرنلی نشان می­دهد که در نظر گرفتن ارتباطات غیرخطی بین ویژگی­های سیگنال مغزی، دقت رمزگشایی نیرو را نسبت به مدل­های خطی بهبود می‌بخشد. درصد بهبود میانگین ضریب R2 برابر با 7/12% برای روش رگرسیون ستیغی کرنلی نسبت به روش ستیغی، 5/25% برای روش PCR کرنلی نسبت به PCR و 1/19% برای روش PLS کرنلی نسبت به PLS به دست آمده است. بهترین دقت رمزگشایی نیرو نیز به ازای روش رگرسیون ستیغی کرنلی، با میانگین ضریب همبستگی 72% و مقدار R2 برابر با 62/0 به دست آمده است.

کلیدواژه‌ها

عنوان مقاله [English]

Evaluation of Nonlinear Models based on Kernel Functions for Force Decoding using Local Field Potential Signals

نویسندگان [English]

  • Maryam Fatemi 1
  • Mohammad Reza Daliri 2

1 Ph.D. Student, Biomedical Engineering Department, School of Electrical Engineering, Iran University of Science and Technology (IUST), Tehran, Iran

2 Professor, Biomedical Engineering Department, School of Electrical Engineering, Iran University of Science and Technology (IUST), Tehran, Iran

چکیده [English]

Controlling of neuroprostheses to restore grasping ability in patients with paralyzed or amputated upper limbs is one of the important applications of BCI systems. The ability to get objects is necessary for daily works so, for a reliable function of the neuroprostheses, it is necessary for the user to control the amount of force needed for grasping. For this reason, increasing the accuracy of continuous force decoding is an important issue for the convenient function of these BCI systems. In most studies in the field of force decoding, linear models such as wiener filter, Kalman filter, PLS, etc. are used to decode force. So far, the effect of using nonlinear models is not investigated on force decoding. The goal of this study is to investigate the effect of using nonlinear regression models based on kernel functions on the accuracy of force decoding in Vistar rats using local field potential signals. To do this, we choose ridge regression, PCR and PLS methods and use the Gaussian kernel function to construct a generalized nonlinear model for the force decoding. Evaluating kernel ridge, kernel PCR and kernel PLS methods shows that considering nonlinear relations between brain signal’s features improves decoding accuracy. The mean coefficient of determination (R2) improves 12.7% in kernel ridge toward ridge regression, 25.5% in kernel PCR toward PCR and 19.1% in kernel PLS toward PLS method. The best decoding accuracy has been achieved by the kernel ridge regression method and the mean correlation coefficient between the estimated and measured force is 0.72 and R2 is 0.62.

کلیدواژه‌ها [English]

  • Brain Computer Interface (BCI)
  • Local Field Potential
  • Continuous Force Decoding
  • Nonlinear Regression
[1]   J. Wolpaw, E.W. Wolpaw, “Brain-computer interfaces: principles and practice,” OUP; USA: 2012.
[2]   R. D. Flint, P.T Wang, Z. A. Wright. ‘Extracting kinetic information from human motor cortical signals’ Neuroimage, vol. 101, pp. 695-703, 2014.
[3]   J.M. Carmena, M.A. Lebedev, R.E. Crist, J.E. O'Doherty, D.M. Santucci, D.F. Dimitrov, P.G. Patil, C.S. Henriquez, M.A. Nicolelis, “Learning to control a brain–machine interface for reaching and grasping by primates,” PLoS Biology, vol. 1, pp. 1–16,2003.
[4]   R. Gupta , J. Ashe, “Offline Decoding of End-Point Forces Using Neural Ensembles: Application to a Brain–Machine Interface,” IEEE Transaction on Neural System and Rehabilitation Engineering , vol. 17, no. 3 ,2009
[5]   A.K. Bansal, W. Truccolo, C.E. Vargas-Irwin, J.P. Donoghue, “Decoding 3D reach and grasp from hybrid signals in motor and premotor cortices: spikes, multiunit activity, and local field potentials,” Journal of Neurophysiology, vol. 107, pp. 1337–1355, 2012.
[6]   M. Saleh, K. Takahashi, N.G. Hatsopoulos, “Encoding of coordinated reach and grasp trajectories in primary motor cortex,” Journal of Neuroscience, vol.32, pp. 1220–1232, 2012.
[7]   C.E. Vargas-Irwin, G. Shakhnarovich,, P. Yadollahpour, J.M. Mislow, M.J. Black, J.P. Donoghue, “Decoding complete reach and grasp actions from local primary motor cortex populations,” Journal of  Neuroscience, vol.30, pp. 9659–9669, 2010.
[8]   M. Mollazadeh, V. Aggarwal, A.G. Davidson, A.J. Law, N.V. Thakor, M.H. Schieber, “Spatiotemporal variation of multiple neurophysiological signals in the primary motor cortex during dexterous reach-to-grasp movements,” Journal of Neuroscience, vol. 31, pp. 15531–15543, 2011
[9]   R.D. Flint, E.W. Lindberg, L.R. Jordan, L.E. Miller, M.W. Slutzky, ‘Accurate decoding of reaching movements from field potentials in the absence of spikes,’ Journal of Neural Engineering, vol 9, 2012.
[10]R.D. Flint, C. Ethier, E.R. Oby, L.E. Miller, M.W. Slutzky, “Local field potentials allow accurate decoding of muscle activity,”  Journal of  Neurophysiology, vol. 108, pp. 18–24 2012.
[11]C. Mehring, J. Rickert, E. Vaadia, S. Cardoso de Oliveira, A. Aertsen, S. Rotter, “Inference of hand movements from local field potentials in monkey motor cortex,” Nature Neuroscience, vol. 6, pp. 1253–1254, 2003.
[12]D.A. Markowitz, Y.T. Wong, C.M. Gray, B. Pesaran, “Optimizing the decoding of movement goals from local field potentials in macaque cortex,” Journal of Neuroscience, vol. 31, pp. 18412–18422, 2011.
[13]A.K. Bansal, C.E. Vargas-Irwin, W. Truccolo, J.P.  Donoghue, “Relationships among low-frequency local field potentials, spiking activity, and three-dimensional reach and grasp kinematics in primary motor and ventral premotor cortices,” Journal of Neurophysiology, vol. 105, pp. 1603–1619, 2011.
[14]J. Rickert, SC. Oliveira, E. Vaadia, A. Aertsen, S. Rotter, C. Mehring, “Encoding of movement direction in different frequency ranges of motor cortical local field potentials,” Journal of Neuroscience, vol. 25, pp.8815–8824, 2005.
[15]C. Mehring, MP. Nawrot, SC. de Oliveira, E. Vaadia, A. Schulze-Bonhage, A. Aertsen, T. Ball, “Comparing information about arm movement direction in single channels of local and epicortical field potentials from monkey and human motor cortex,” Physiology Paris, vol. 98, pp. 498–506, 2004.
[16]J. Zhuang, W. Truccolo, C. Vargas-Irwin, J.P. Donoghue, “Decoding 3-D reach and grasp kinematics from high frequency local field potentials in primate primary motor cortex,” IEEE Transaction on Biomedical Engineering, vol. 57, pp. 1774–1784, 2010.
[17]T. Milecovic, W. Trocolo, “Local field potentials in primate motor cortex encode grasp kinetic parameters,” Neuroimage, vol. 114, pp. 338-355, 2015.
[18]A. Khorasani, N. Heydari Beni, V. Shalchyan, M.R. Daliri, “Continuous Force Decoding from Local Field Potentials of the Primary Motor Cortex in Freely Moving Rat,” Scientific Reports, 2016.
[19]S. A. Shah, H. Tan, P. Brown. “Continuous Force Decoding from Deep Brain Local Field Potentials for Brain Computer Interfacing,” International IEEE EMBS Conference in Neural Engineering, pp. 371–374, 2017.
[20]R. D. Flint, P.T Wang, Z. A. Wright, “Extracting kinetic information from human motor cortical signals,” Neuroimage, vol. 101, pp.695-703, 2014.
[21]Ch. Chen, D shin. “Decoding grasp force profile from electrocorticography signals in non-human primate sensorimotor cortex,” Neuroscience Research, vol. 83, 1-7, 2014.
[22]J. Wessberg, C. R. Stambaugh, J. D. Kralik, P. D. Beck, M. Laubach, J. K. Chapin, et al., "Real-time prediction of hand trajectory by ensembles of cortical neurons in primates," Nature, vol. 408, pp. 361-365, 2000.
[23]R. D. Flint, E. W. Lindberg, L. R. Jordan, L. E. Miller, M. W. Slutzky, "Accurate decoding of reaching movements from field potentials in the absence of spikes," Journal of neural engineering, vol 9,  2012.
[24]J. H.Kim, F.Biebmann, S.W. Lee, “Decoding Three-Dimensional Trajectory of Executed and Imagined Arm Movements from Electroencephalogram Signals”, IEEE Transactions on Neural Systems and Rehabilitation Engineering,Vol.23,num. 5,2014.
[25]J. M. H, F. Bießmann, N. Jiang, H. Rehbaum,D. Farina, F. C. Meinecke, K. R. Müller, L. C. Parra,” Linear and Nonlinear Regression Techniques for Simultaneous and Proportional Myoelectric Control”,  IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol.23, num.2, 2014.
[26]S. A. Shah, .Tan, G.Tinkhauser ,P.Brown, “Towards Real-Time, Continuous Decoding of Gripping Force From Deep Brain Local Field Potentials”, IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol. 26, issue 7, 2018.
[27]B. Scholkopf, A. Smola, K. R. Muller. ‘Nonlinear Component Analysis as a Kernel Eigenvalue Problem’. Neural Computation, vol 10, num 5, pp 1299-1319, 1998.
[28]R. Rosipal and L. J. Trejo. ‘Kernel Partial Least Squares Regression in Reproducing Kernel Hilbert Space’. Journal of Machine Learning Research2, pp 97-123. 2001.
[29]B.V. Srinivasan, Y.C. Luo, D. G. Romero. ‘A Symmetric Kernel Partial Least Squares Framework for Speaker RECoGnition’, IEEE Transactions on Audio, Speech, and Language Processing ,vol  21, num 7, pp 1415-1423, 2013.
[30]Khorasani, N. Heydari Beni, V. Shalchyan M.R. Daliri. “Continuous Force Decoding from Local Field Potentials of the Primary Motor Cortex in Freely Moving Rat”. Scientific Reports, 2016.