Mixture of trajectory models for neural decoding of goal-directed movements

被引:112
作者
Yu, Byron M.
Kemere, Caleb
Santhanam, Gopal
Afshar, Afsheen
Ryu, Stephen I.
Meng, Teresa H.
Sahani, Maneesh
Shenoy, Krishna V.
机构
[1] Stanford Univ, Dept Elect Engn, Stanford, CA 94305 USA
[2] Stanford Univ, Med Scientist Training Program, Stanford, CA 94305 USA
[3] Stanford Univ, Dept Neurosurg, Stanford, CA 94305 USA
[4] Stanford Univ, Neurosci Program, Stanford, CA 94305 USA
[5] UCL, Gatsby Computat Neurosci Unit, London, England
关键词
D O I
10.1152/jn.00482.2006
中图分类号
Q189 [神经科学];
学科分类号
071006 ;
摘要
Probabilistic decoding techniques have been used successfully to infer time-evolving physical state, such as arm trajectory or the path of a foraging rat, from neural data. A vital element of such decoders is the trajectory model, expressing knowledge about the statistical regularities of the movements. Unfortunately, trajectory models that both 1) accurately describe the movement statistics and 2) admit decoders with relatively low computational demands can be hard to construct. Simple models are computationally inexpensive, but often inaccurate. More complex models may gain accuracy, but at the expense of higher computational cost, hindering their use for real-time decoding. Here, we present a new general approach to defining trajectory models that simultaneously meets both requirements. The core idea is to combine simple trajectory models, each accurate within a limited regime of movement, in a probabilistic mixture of trajectory models (MTM). We demonstrate the utility of the approach by using an MTM decoder to infer goal-directed reaching movements to multiple discrete goals from multi-electrode neural data recorded in monkey motor and premotor cortex. Compared with decoders using simpler trajectory models, the MTM decoder reduced the decoding error by 38 (48) percent in two monkeys using 98 (99) units, without a necessary increase in running time. When available, prior information about the identity of the upcoming reach goal can be incorporated in a principled way, further reducing the decoding error by 20 (11) percent. Taken together, these advances should allow prosthetic cursors or limbs to be moved more accurately toward intended reach goals.
引用
收藏
页码:3763 / 3780
页数:18
相关论文
共 68 条
[1]  
AFSHAR A, 2005, SOC NEUROSCI ABSTR
[2]   MOVEMENT PARAMETERS AND NEURAL ACTIVITY IN MOTOR CORTEX AND AREA-5 [J].
ASHE, J ;
GEORGOPOULOS, AP .
CEREBRAL CORTEX, 1994, 4 (06) :590-600
[3]   Reach plans in eye-centered coordinates [J].
Batista, AP ;
Buneo, CA ;
Snyder, LH ;
Andersen, RA .
SCIENCE, 1999, 285 (5425) :257-260
[4]   Recursive Bayesian decoding of motor cortical signals by particle filtering [J].
Brockwell, AE ;
Rojas, AL ;
Kass, RE .
JOURNAL OF NEUROPHYSIOLOGY, 2004, 91 (04) :1899-1907
[5]  
Brown EN, 1998, J NEUROSCI, V18, P7411
[6]  
CAMINITI R, 1991, J NEUROSCI, V11, P1182
[7]   Learning to control a brain-machine interface for reaching and grasping by primates [J].
Carmena, JM ;
Lebedev, MA ;
Crist, RE ;
O'Doherty, JE ;
Santucci, DM ;
Dimitrov, DF ;
Patil, PG ;
Henriquez, CS ;
Nicolelis, MAL .
PLOS BIOLOGY, 2003, 1 (02) :193-208
[8]   Computational model of a primate arm: from hand position to joint angles, joint torques and muscle forces [J].
Chan, Sherwin S. ;
Moran, Daniel W. .
JOURNAL OF NEURAL ENGINEERING, 2006, 3 (04) :327-337
[9]   Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex [J].
Chapin, JK ;
Moxon, KA ;
Markowitz, RS ;
Nicolelis, MAL .
NATURE NEUROSCIENCE, 1999, 2 (07) :664-670
[10]   Preparatory activity in premotor and motor cortex reflects the speed of the upcoming reach [J].
Churchland, Mark M. ;
Santhanam, Gopal ;
Shenoy, Krishna V. .
JOURNAL OF NEUROPHYSIOLOGY, 2006, 96 (06) :3130-3146