Article

Adaptive decoding for brain-machine interfaces through Bayesian parameter updates.

Department of Neurobiology and Center for Neuroengineering, Duke University, Durham, NC 27710, U.S.A.
Neural Computation (Impact Factor: 1.69). 09/2011; 23(12):3162-204. DOI: 10.1162/NECO_a_00207
Source: PubMed

ABSTRACT Brain-machine interfaces (BMIs) transform the activity of neurons recorded in motor areas of the brain into movements of external actuators. Representation of movements by neuronal populations varies over time, during both voluntary limb movements and movements controlled through BMIs, due to motor learning, neuronal plasticity, and instability in recordings. To ensure accurate BMI performance over long time spans, BMI decoders must adapt to these changes. We propose the Bayesian regression self-training method for updating the parameters of an unscented Kalman filter decoder. This novel paradigm uses the decoder's output to periodically update its neuronal tuning model in a Bayesian linear regression. We use two previously known statistical formulations of Bayesian linear regression: a joint formulation, which allows fast and exact inference, and a factorized formulation, which allows the addition and temporary omission of neurons from updates but requires approximate variational inference. To evaluate these methods, we performed offline reconstructions and closed-loop experiments with rhesus monkeys implanted cortically with microwire electrodes. Offline reconstructions used data recorded in areas M1, S1, PMd, SMA, and PP of three monkeys while they controlled a cursor using a handheld joystick. The Bayesian regression self-training updates significantly improved the accuracy of offline reconstructions compared to the same decoder without updates. We performed 11 sessions of real-time, closed-loop experiments with a monkey implanted in areas M1 and S1. These sessions spanned 29 days. The monkey controlled the cursor using the decoder with and without updates. The updates maintained control accuracy and did not require information about monkey hand movements, assumptions about desired movements, or knowledge of the intended movement goals as training signals. These results indicate that Bayesian regression self-training can maintain BMI control accuracy over long periods, making clinical neuroprosthetics more viable.

0 Followers
 · 
187 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: Reinforcement learning (RL) based brain machine interfaces (BMIs) enable user to learn from environment through interactions to complete the task without desired signals, which is promising for clinical applications. Previous studies exploit Q-learning techniques to discriminate neural states into simple directional actions providing the trial initial timing. However, the movements in BMI applications could be quite complicated, and the action timing explicitly shows the intention when to move. The rich actions and the corresponding neural states form a large state-action space, imposing generalization difficulty on Q-learning. In this paper, we propose to adopt attention-gated reinforcement learning (AGREL) as a new learning scheme for BMIs to adaptively decode high-dimensional neural activities into 7 distinct movements (directional moves, holdings and resting) due to the efficient weight-updating. We apply AGREL on neural data recorded from M1 of a monkey to directly predict a seven-action set in a time sequence to reconstruct the trajectory of a center-out task. Compared to Q-learning techniques, AGREL could improve the target acquisition rate to 90.16% in average with faster convergence and more stability to follow neural activity over multiple days, indicating the potential to achieve better online decoding performance for more complicated BMI tasks.
    IEEE Transactions on Neural Systems and Rehabilitation Engineering 07/2014; 23(3). DOI:10.1109/TNSRE.2014.2341275 · 2.82 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Successful neurological rehabilitation depends on accurate diagnosis, effective treatment, and quantitative evaluation. Neural coding, a technology for interpretation of functional and structural information of the nervous system, has contributed to the advancements in neuroimaging, brain-machine interface (BMI), and design of training devices for rehabilitation purposes. In this review, we summarized the latest breakthroughs in neuroimaging from microscale to macroscale levels with potential diagnostic applications for rehabilitation. We also reviewed the achievements in electrocorticography (ECoG) coding with both animal models and human beings for BMI design, electromyography (EMG) interpretation for interaction with external robotic systems, and robot-assisted quantitative evaluation on the progress of rehabilitation programs. Future rehabilitation would be more home-based, automatic, and self-served by patients. Further investigations and breakthroughs are mainly needed in aspects of improving the computational efficiency in neuroimaging and multichannel ECoG by selection of localized neuroinformatics, validation of the effectiveness in BMI guided rehabilitation programs, and simplification of the system operation in training devices.
    BioMed Research International 09/2014; 2014. DOI:10.1155/2014/286505 · 2.71 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Development of neural prostheses over the past few decades has produced a number of clinically relevant brain-machine interfaces (BMIs), such as the cochlear prostheses and deep brain stimulators. Current research pursues the restoration of communication or motor function to individuals with neurological disorders. Efforts in the field, such as the BrainGate trials, have already demonstrated that such interfaces can enable humans to effectively control external devices with neural signals. However, a number of significant issues regarding BMI performance, device capabilities, and surgery must be resolved before clinical use of BMI technology can become widespread. This chapter reviews challenges to clinical translation and discusses potential solutions that have been reported in recent literature, with focuses on hardware reliability, state-of-the-art decoding algorithms, and surgical considerations during implantation.
    International Review of Neurobiology 01/2012; 107:137-60. DOI:10.1016/B978-0-12-404706-8.00008-5 · 2.46 Impact Factor