Optimal Target Placement for Neural Communication Prostheses (original) (raw)

Toward Optimal Target Placement for Neural Prosthetic Devices

Journal of Neurophysiology - J NEUROPHYSIOL, 2008

Neural prosthetic systems have been designed to estimate continuous reach trajectories (motor prostheses) and to predict discrete reach targets (communication prostheses). In the latter case, reach targets are typically decoded from neural spiking activity during an instructed delay period, before the reach begins. Such systems use targets placed in radially symmetric geometries, independent of the tuning properties of the neurons available. Here we seek to automate the target placement process and increase decode accuracy in communication prostheses by selecting target locations based on the neural population at hand. Motor prostheses that incorporate intended target information could also benefit from this consideration. We present an optimal target placement algorithm that approximately maximizes decode accuracy with respect to target locations. In simulated neural spiking data fit from two monkeys, the optimal target placement algorithm yielded statistically significant improvements up to 8 and 9% for two and sixteen targets, respectively. For four and eight targets, gains were more modest, as the target layouts found by the algorithm closely resembled the canonical layouts. We trained a monkey in this paradigm and tested the algorithm with experimental neural data to confirm some of the results found in simulation. In all, the algorithm can serve not only to create new target layouts that outperform canonical layouts, but it can also confirm or help select among multiple canonical layouts. The optimal target placement algorithm developed here is the first algorithm of its kind, and it should both improve decode accuracy and help automate target placement for neural prostheses.

Decoding of plan and peri-movement neural signals in prosthetic systems

IEEE Workshop on Signal Processing Systems, 2002

One might anticipate that combining plan activity with peri-In this paper we introduce a theoretical framework for improved processing of peri-movement neural activity for neurally controlled prosthetic systems through maximum likelihood sequence estimation. This framework further suggests a computational method for integrating plan and peri-movement neural activity. We show that combining plan activity, usually associated with target specification, with peri-movement neural activity yields more accurate estimates of the trajectoryof an arm movement. The effectiveness of the method is demonstrated in simulation. Performance as a function of the specific number of plan and peri-movement neurons, as well as other system and design parameters is analyzed. The algorithm presented is also compared against previous, sample-based approaches, specifically a "point-process'' filter for plan activity and a standard linear filter framework in the perimovement regime.

Improving neural prosthetic system performance by combining plan and peri-movement activity

Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2004

While most neural prosthetic systems to date estimate arm movements based solely on the activity prior to reaching movements during a delay period (plan activity) or solely on the activity during reaching movements (perimovement activity), we show that decode classiÞcation can be improved by 56% and 71% respectively by using both types of activity together. We recorded from the pre-motor cortex of a rhesus monkey performing a delayed-reach task to one of seven targets. We found that taking into account the timevarying structure in peri-movement activity further improved performance by 15%, while doing the same for plan activity did not improve performance. We also found low correlations in activity between pairs of simultaneously-recorded units and across time periods within a given trial condition. These results show that decode performance can be signiÞcantly improved by combining information from the plan and peri-movement periods, and that there is nearly no loss in performance when assuming independence between units and across time periods within a given trial condition.

In search of more robust decoding algorithms for neural prostheses, a data driven approach

2010

In the past decade the field of neural interface systems has enjoyed an increase in attention from the scientific community and the general public, in part due to the enormous potential that such systems have to increase the quality of life for paralyzed patients. While significant progress has been made, serious challenges remain to be addressed from both biological and engineering perspectives. A key issue is how to optimize the decoding of neural information, such that neural signals are correctly mapped to effectors that interact with the outside world-like robotic hands and limbs or the patient's own muscles. Here we present some recent progress on tackling this problem by applying the latest developments in machine learning. Neural data was collected from macaque monkeys performing a real-time hand grasp decoding task. Signals were recorded via chronically implanted electrodes in the anterior intraparietal cortex (AIP) and ventral premotor cortex (F5), brain areas that are known to be involved in the transformation of visual signals into hand grasping instructions. We present a comparative study of different classical machine learning methods with an application of decoding of hand postures, as well as a new approach for more robust decoding. Results suggests that combining data-driven algorithmic approaches with well-known parametric methods could lead to better performing and more robust learners, which may have direct implications for future clinical devices.

Neural Prostheses: Linking Brain Signals to Prosthetic Devices

This paper discusses Neuroprosthetic applications for potential use by paralyzed patients or amputees. These systems require advanced processing of neural signals to drive the prosthetic devices using decoding algorithms. The possibility of predicting motor commands from neural signals are the core of neural prosthetic devices and along this line we show how it is possible to predict movement intentions as well as what subjects are seeing from the firing of population of neurons.

Neural prosthetic control signals from plan activity

NeuroReport, 2003

The prospect of assisting disabled patients by translating neural activity from the brain into control signals for prosthetic devices, has £ourished in recent years. Current systems rely on neural activity present during natural arm movements. We propose here that neural activity present before or even without natural arm movements can provide an important, and potentially advantageous, source of control signals. To demonstrate how control signals can be derived from such plan activity we performed a computational study with neural activity previously recorded from the posterior parietal cortex of rhesus monkeys planning arm movements. We employed maximum likelihood decoders to estimate movement direction and to drive ¢nite state machines governing when to move. Performance exceeded 90% with as few as 40 neurons. NeuroRe-port12 :000^000

A factor-analysis decoder for high-performance neural prostheses

International Conference on Acoustics, Speech, and Signal Processing, 2008

Increasing the performance of neural prostheses is necessary for assuring their clinical viability. One performance limitation is the presence of correlated trial-to-trial variability that can cause neural responses to wax and wane in concert as the subject is, for example, more attentive or more fatigued. We report here the design and characterization of a Factor- Analysis-based decoding algorithm that is able to contend with this confound. We characterize the decoder (classifier) on a previously reported dataset where monkeys performed both a real reach task and a prosthetic cursor movement task while we recorded from 96 electrodes implanted in dorsal pre- motor cortex. In principle, the decoder infers the underlying factors that co-modulate the neurons' responses and can use this information to function with reduced error rates (1 of 8 reach target prediction) of up to ~75% (~20% total prediction error using independent Gaussian or Poisson models became ~5%). Such Factor-Analysis based methods appear to be effective when attempting to combat directly unobserved trial-by-trial neural variabiliy.

Model-based decoding of reaching movements for prosthetic systems

Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2004

Model-based decoding of neural activity for neuroprosthetic systems has been shown, in simulation, to provide significant gain over traditional linear filter approaches. We tested the model-based decoding approach with real neural and behavioral data and found a 18% reduction in trajectory reconstruction error compared with a linear filter. This corresponds to a 40% reduction in the number of neurons required for equivalent performance. The model-based approach further permits the combination of target-tuned plan activity with movement activity. The addition of plan activity reduced reconstruction error by 23% relative to the linear filter, corresponding to 55% reduction in the number of neurons required. Taken together, these results indicate that a decoding algorithm employing a prior model of reaching kinematics can substantially improve trajectory estimates, thereby improving prosthetic system performance.

A high-performance neural prosthesis enabled by control algorithm design

Nature Neuroscience, 2012

Neural prostheses translate neural activity from the brain into control signals for guiding prosthetic devices, such as computer cursors and robotic limbs, and thus offer disabled patients greater interaction with the world. However, relatively low performance remains a critical barrier to successful clinical translation; current neural prostheses are considerably slower with less accurate control than the native arm. Here we present a new control algorithm, the recalibrated feedback intention-trained Kalman filter (ReFIT-KF), that incorporates assumptions about the nature of closed loop neural prosthetic control. When tested with rhesus monkeys implanted with motor cortical electrode arrays, the ReFIT-KF algorithm outperforms existing neural prostheses in all measured domains and halves acquisition time. This control algorithm permits sustained uninterrupted use for hours and generalizes to more challenging tasks without retraining. Using