Brain-state classification and a dual-state decoder dramatically improve the control of cursor movement through a brain-machine interface

OBJECTIVE It is quite remarkable that brain machine interfaces (BMIs) can be used to control complex movements with fewer than 100 neurons. Success may be due in part to the limited range of dynamical conditions under which most BMIs are tested. Achieving high-quality control that spans these conditions with a single linear mapping will be more challenging. Even for simple reaching movements, existing BMIs must reduce the stochastic noise of neurons by averaging the control signals over time, instead of over the many neurons that normally control movement. This forces a compromise between a decoder with dynamics allowing rapid movement and one that allows postures to be maintained with little jitter. Our current work presents a method for addressing this compromise, which may also generalize to more highly varied dynamical situations, including movements with more greatly varying speed. APPROACH We have developed a system that uses two independent Wiener filters as individual components in a single decoder, one optimized for movement, and the other for postural control. We computed an LDA classifier using the same neural inputs. The decoder combined the outputs of the two filters in proportion to the likelihood assigned by the classifier to each state. MAIN RESULTS We have performed online experiments with two monkeys using this neural-classifier, dual-state decoder, comparing it to a standard, single-state decoder as well as to a dual-state decoder that switched states automatically based on the cursor's proximity to a target. The performance of both monkeys using the classifier decoder was markedly better than that of the single-state decoder and comparable to the proximity decoder. SIGNIFICANCE We have demonstrated a novel strategy for dealing with the need to make rapid movements while also maintaining precise cursor control when approaching and stabilizing within targets. Further gains can undoubtedly be realized by optimizing the performance of the individual movement and posture decoders.

[1]  S. Solla,et al.  Toward the Restoration of Hand Use to a Paralyzed Monkey: Brain-Controlled Functional Electrical Stimulation of Forearm Muscles , 2009, PloS one.

[2]  Claude Ghez,et al.  Separate adaptive mechanisms for controlling trajectory and final position in reaching. , 2007, Journal of neurophysiology.

[3]  A. Georgopoulos,et al.  The motor cortex and the coding of force. , 1992, Science.

[4]  J D Simeral,et al.  Continuous neuronal ensemble control of simulated arm reaching by a human with tetraplegia , 2011, Journal of neural engineering.

[5]  Michael J. Black,et al.  Point-and-Click Cursor Control With an Intracortical Neural Interface System by Humans With Tetraplegia , 2011, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[6]  E. Fetz,et al.  Direct control of paralyzed muscles by cortical neurons , 2008, Nature.

[7]  Deniz Erdogmus,et al.  Divide-and-conquer approach for brain machine interfaces: nonlinear mixture of competitive linear models , 2003, Neural Networks.

[8]  Nicolas Y. Masse,et al.  Reach and grasp by people with tetraplegia using a neurally controlled robotic arm , 2012, Nature.

[9]  David M. Santucci,et al.  Learning to Control a Brain–Machine Interface for Reaching and Grasping by Primates , 2003, PLoS biology.

[10]  Richard A Andersen,et al.  Decoding Trajectories from Posterior Parietal Cortex Ensembles , 2008, The Journal of Neuroscience.

[11]  J. M. Carmena,et al.  Closed-Loop Decoder Adaptation on Intermediate Time-Scales Facilitates Rapid BMI Performance Improvements Independent of Decoder Initialization Conditions , 2012, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[12]  Sung-Phil Kim,et al.  Independently Coupled HMM Switching Classifier for a Bimodel Brain-Machine Interface , 2006, 2006 16th IEEE Signal Processing Society Workshop on Machine Learning for Signal Processing.

[13]  J. Carmena,et al.  Emergence of a Stable Cortical Map for Neuroprosthetic Control , 2009, PLoS biology.

[14]  Browne,et al.  Cross-Validation Methods. , 2000, Journal of mathematical psychology.

[15]  Zachary Danziger,et al.  Learning Algorithms for Human–Machine Interfaces , 2009, IEEE Transactions on Biomedical Engineering.

[16]  L. Miller,et al.  Restoring sensorimotor function through intracortical interfaces: progress and looming challenges , 2014, Nature Reviews Neuroscience.

[17]  David T. Westwick,et al.  Identification of Multiple-Input Systems with Highly Coupled Inputs: Application to EMG Prediction from Multiple Intracortical Electrodes , 2006, Neural Computation.

[18]  Miguel A. L. Nicolelis,et al.  Adaptive Decoding for Brain-Machine Interfaces Through Bayesian Parameter Updates , 2011, Neural Computation.

[19]  Byron M. Yu,et al.  Motor cortical control of movement speed with implications for brain-machine interface control. , 2014, Journal of neurophysiology.

[20]  D. Rosenbaum,et al.  Limb position drift: implications for control of posture and movement. , 2003, Journal of neurophysiology.

[21]  Sherwin S Chan,et al.  Motor cortical representation of position and velocity during reaching. , 2007, Journal of neurophysiology.

[22]  A. P. Georgopoulos,et al.  Movement parameters and neural activity in motor cortex and area 5. , 1994, Cerebral cortex.

[23]  Robert E. Kass,et al.  Comparison of brain–computer interface decoding algorithms in open-loop and closed-loop control , 2010, Journal of Computational Neuroscience.

[24]  A. Schwartz,et al.  High-performance neuroprosthetic control by an individual with tetraplegia , 2013, The Lancet.

[25]  Robin C. Ashmore,et al.  An Electrocorticographic Brain Interface in an Individual with Tetraplegia , 2013, PloS one.

[26]  Byron M. Yu,et al.  A high-performance brain–computer interface , 2006, Nature.

[27]  Justin C. Sanchez,et al.  Using Reinforcement Learning to Provide Stable Brain-Machine Interface Control Despite Neural Input Reorganization , 2014, PloS one.

[28]  Teresa H. Y. Meng,et al.  Model-based neural decoding of reaching movements: a maximum likelihood approach , 2004, IEEE Transactions on Biomedical Engineering.

[29]  A B Schwartz,et al.  Motor cortical representation of speed and direction during reaching. , 1999, Journal of neurophysiology.

[30]  Michael J. Black,et al.  Modeling and decoding motor cortical activity using a switching Kalman filter , 2004, IEEE Transactions on Biomedical Engineering.

[31]  Sagi Perel,et al.  Single-unit activity, threshold crossings, and local field potentials in motor cortex differentially encode reach kinematics. , 2015, Journal of neurophysiology.

[32]  C. Ethier,et al.  Continuous state-dependent decoders for brain machine interfaces , 2011, 2011 5th International IEEE/EMBS Conference on Neural Engineering.

[33]  L. Miller,et al.  Restoration of grasp following paralysis through brain-controlled stimulation of muscles , 2012, Nature.

[34]  Nicolas Y. Masse,et al.  Advantages of closed-loop calibration in intracortical brain–computer interfaces for people with tetraplegia , 2013, Journal of neural engineering.

[35]  Meel Velliste,et al.  Motor Cortical Correlates of Arm Resting in the Context of a Reaching Task and Implications for Prosthetic Control , 2014, The Journal of Neuroscience.

[36]  L. Miller,et al.  Motor cortical prediction of EMG: evidence that a kinetic brain-machine interface may be robust across altered movement dynamics. , 2011, Journal of neurophysiology.

[37]  John P. Cunningham,et al.  A High-Performance Neural Prosthesis Enabled by Control Algorithm Design , 2012, Nature Neuroscience.

[38]  S. Scott,et al.  Random change in cortical load representation suggests distinct control of posture and movement , 2005, Nature Neuroscience.

[39]  Emery N. Brown,et al.  A State-Space Analysis for Reconstruction of Goal-Directed Movements Using Neural Signals , 2006, Neural Computation.

[40]  J. Lackner,et al.  Motor adaptation to Coriolis force perturbations of reaching movements: endpoint but not trajectory adaptation transfers to the nonexposed arm. , 1995, Journal of neurophysiology.

[41]  Andrew S. Whitford,et al.  Cortical control of a prosthetic arm for self-feeding , 2008, Nature.

[42]  R. Andersen,et al.  Decoding motor imagery from the posterior parietal cortex of a tetraplegic human , 2015, Science.

[43]  P. Viviani,et al.  The law relating the kinematic and figural aspects of drawing movements. , 1983, Acta psychologica.

[44]  D R Humphrey,et al.  Predicting Measures of Motor Performance from Multiple Cortical Spike Trains , 1970, Science.

[45]  Nicholas G Hatsopoulos,et al.  Incorporating Feedback from Multiple Sensory Modalities Enhances Brain–Machine Interface Control , 2010, The Journal of Neuroscience.

[46]  S. Grossberg,et al.  Cortical networks for control of voluntary arm movements under variable force conditions. , 1998, Cerebral cortex.

[47]  Nicholas G. Hatsopoulos,et al.  Brain-machine interface: Instant neural control of a movement signal , 2002, Nature.

[48]  Jon A. Mukand,et al.  Neuronal ensemble control of prosthetic devices by a human with tetraplegia , 2006, Nature.

[49]  Byron M. Yu,et al.  Mixture of Trajectory Models for Neural Decoding of Goal-directed Movements a Computational Model of Craving and Obsession Decoding Visual Inputs from Multiple Neurons in the Human Temporal Lobe Encoding Contribution of Individual Retinal Ganglion Cell Responses to Velocity and Acceleration , 2008 .