Recurrent neural networks (RNNs) are useful tools for learning nonlinear relationships in time series data with complex temporal dependences. In this paper, we explore the ability of a simplified type of RNN, one with limited modifications to the internal weights called an echostate network (ESN), to effectively and continuously decode monkey reaches during a standard center-out reach task using a cortical brain–machine interface (BMI) in a closed loop. We demonstrate that the RNN, an ESN implementation termed a FORCE decoder (from first order reduced and controlled error learning), learns the task quickly and significantly outperforms the current state-of-the-art method, the velocity Kalman filter (VKF), using the measure of target acquire time. We also demonstrate that the FORCE decoder generalizes to a more difficult task by successfully operating the BMI in a randomized point-to-point task. The FORCE decoder is also robust as measured by the success rate over extended sessions. Finally, we show that decoded cursor dynamics are more like naturalistic hand movements than those of the VKF. Taken together, these results suggest that RNNs in general, and the FORCE decoder in particular, are powerful tools for BMI decoder applications.

/pdf/a-recurrent-neural-network-for-closed-loop-intracortical-2r07uxqc9p.pdf

A recurrent neural network for closed-loop intracortical brain-machine interface decoders.

Objective. Cortically-controlled motor prostheses aim to restore functions lost to neurological disease and injury. Several proof of concept demonstrations have shown encouraging results, but barriers to clinical translation still remain. In particular, intracortical prostheses must satisfy stringent power dissipation constraints so as not to damage cortex. Approach. One possible solution is to use ultra-low power neuromorphic chips to decode neural signals for these intracortical implants. The first step is to explore in simulation the feasibility of translating decoding algorithms for brain?machine interface (BMI) applications into spiking neural networks (SNNs). Main results. Here we demonstrate the validity of the approach by implementing an existing Kalman-filter-based decoder in a simulated SNN using the Neural Engineering Framework (NEF), a general method for mapping control algorithms onto SNNs. To measure this system?s robustness and generalization, we tested it online in closed-loop BMI experiments with two rhesus monkeys. Across both monkeys, a Kalman filter implemented using a 2000-neuron SNN has comparable performance to that of a Kalman filter implemented using standard floating point techniques. Significance. These results demonstrate the tractability of SNN implementations of statistical signal processing algorithms on different monkeys and for several tasks, suggesting that a SNN decoder, implemented on a neuromorphic chip, may be a feasible computational platform for low-power fully-implanted prostheses. The validation of this closed-loop decoder system and the demonstration of its robustness and generalization hold promise for SNN implementations on an ultra-low power neuromorphic chip using the NEF.

/pdf/design-and-validation-of-a-real-time-spiking-neural-network-14ok4d5g9r.pdf

Design and validation of a real-time spiking-neural-network decoder for brain–machine interfaces

Objective. Support vector machines (SVM) have developed into a gold standard for accurate classification in brain–computer interfaces (BCI). The choice of the most appropriate classifier for a particular application depends on several characteristics in addition to decoding accuracy. Here we investigate the implementation of hidden Markov models (HMM) for online BCIs and discuss strategies to improve their performance. Approach. We compare the SVM, serving as a reference, and HMMs for classifying discrete finger movements obtained from electrocorticograms of four subjects performing a finger tapping experiment. The classifier decisions are based on a subset of low-frequency time domain and high gamma oscillation features. Main results. We show that decoding optimization between the two approaches is due to the way features are extracted and selected and less dependent on the classifier. An additional gain in HMM performance of up to 6% was obtained by introducing model constraints. Comparable accuracies of up to 90% were achieved with both SVM and HMM with the high gamma cortical response providing the most important decoding information for both techniques. Significance. We discuss technical HMM characteristics and adaptations in

http://iopscience.iop.org/article/10.1088/1741-2560/10/5/056020/pdf

Hidden Markov model and support vector machine based decoding of finger movements using electrocorticography

Simulation of large-scale networks of spiking neurons has become appealing for understanding the computational principles of the nervous system by producing models based on biological evidence. In particular, networks that can assume a variety of (dynamically) stable states have been proposed as the basis for different behavioural and cognitive functions.

/pdf/real-time-on-chip-implementation-of-dynamical-systems-with-22ynko9qim.pdf

Real time on-chip implementation of dynamical systems with spiking neurons

Motor prostheses aim to restore function to disabled patients. Despite compelling proof of concept systems, barriers to clinical translation remain. One challenge is to develop a low-power, fully-implantable system that dissipates only minimal power so as not to damage tissue. To this end, we implemented a Kalman-filter based decoder via a spiking neural network (SNN) and tested it in brain-machine interface (BMI) experiments with a rhesus monkey. The Kalman filter was trained to predict the arm's velocity and mapped on to the SNN using the Neural Engineering Framework (NEF). A 2,000-neuron embedded Matlab SNN implementation runs in real-time and its closed-loop performance is quite comparable to that of the standard Kalman filter. The success of this closed-loop decoder holds promise for hardware SNN implementations of statistical signal processing algorithms on neuromorphic chips, which may offer power savings necessary to overcome a major obstacle to the successful clinical translation of neural motor prostheses.

/pdf/a-brain-machine-interface-operating-with-a-real-time-spiking-2k8y7wxjaf.pdf

A Brain-Machine Interface Operating with a Real-Time Spiking Neural Network Control Algorithm

We used a spiking neural network (SNN) to decode neural data recorded from a 96-electrode array in premotor/motor cortex while a rhesus monkey performed a point-to-point reaching arm movement task. We mapped a Kalman-filter neural prosthetic decode algorithm developed to predict the arm's velocity on to the SNN using the Neural Engineering Framework and simulated it using Nengo, a freely available software package. A 20,000-neuron network matched the standard decoder's prediction to within 0.03% (normalized by maximum arm velocity). A 1,600-neuron version of this network was within 0.27%, and run in real-time on a 3GHz PC. These results demonstrate that a SNN can implement a statistical signal processing algorithm widely used as the decoder in high-performance neural prostheses (Kalman filter), and achieve similar results with just a few thousand neurons. Hardware SNN implementations - neuromorphic chips - may offer power savings, essential for realizing fully-implantable cortically controlled prostheses.

/pdf/spiking-neural-network-decoder-for-brain-machine-interfaces-yt85fhmdca.pdf

Spiking neural network decoder for brain-machine interfaces

Miniature is massive when it comes to electronics. While there exists a continuous effort in industry to integrate more functionalities into the same area, the prohibitive scaling cost at 45nm and beyond makes it difficult to continue the trend. Towards this, More-than-Moore techniques have been proposed, which explore new dimensionality of integration by creating and integrating non-digital functionality to semiconductor products (Zhang and Roosmalen. More than Moore—creating hnanoelectronics systems/nanoelectronics systems. Springer, New York, 2009). They motivate new technological possibilities and unlimited application potential.

IC-Package-System Integration Design

The unquenched thirst for higher levels of electronic systems integration and higher performance goals has produced a plethora of design and business challenges that are threatening the success enjoyed so far as modeled by Moore's law. To tackle these challenges and meet the design needs of consumer electronics products such as those of cell phones, audio/video players, digital cameras that are composed of a number of different technologies, vertical system integration has emerged as a required technology to reduce the system board space and height in addition to the overall time-to-market and design cost. System-in-package (SiP) is a system integration technology that achieves the aforementioned needs in a scalable and cost-effective way, where multiple dies, passive components, and discrete devices are assembled, often vertically, in a package. This paper surveys the electrical and layout perspectives of SiP. It first introduces package technologies, and then presents SiP design flow and design exploration. Finally, the paper discusses details of beyond-die signal and power integrity and physical implementation such as I/O (input/output cell) placement and routing for redistribution layer, escape, and substrate.

Shauki Elassaad

Papers

Spiking neural network decoder for brain-machine interfaces

IC-Package-System Integration Design

System-In-Package: Electrical and Layout Perspectives

A Brain-Machine Interface with an Innovative Spiking Neural Network Decoder