Shaping the Dynamics of a Bidirectional Neural Interface

Vato, Alessandro; Semprini, Marianna; Maggiolini, Emma; Szymanski, Francois D.; Fadiga, Luciano; Panzeri, Stefano; Mussa-Ivaldi, Ferdinando A.

doi:10.1371/journal.pcbi.1002578

Cited by 24 publications

(37 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Parallel development of components could also accelerate the ultimate realization of a device compact and powerful enough to be used as clinical tool able to transfer data between the brain and external devices wirelessly through an implanted interface (Azin et al, 2011; Fan et al, 2011; Borton et al, 2013; Angotzi et al, 2014). In this work, we also demonstrated that the modular architecture does not affect BMI performances, showing results comparable with the ones achieved in Vato et al (2012); this result suggests that BMI systems developed in other labs could also be re-implemented in a modular manner. To help the interested scientist in doing this, most of the material used in this project is freely available on Si-Code website : http://www.sicode.eu/results/software.…”

Section: Discussionsupporting

confidence: 89%

“…We extended the Dynamic Neural Interface described in Szymanski et al (2011) and Vato et al (2012, 2014) with the inclusion of a neuromorphic decoder module. This system uses the neural signals collected from a rat's brain to control the movement of an external object by means of a sensory and motor interface.…”

Section: Methodsmentioning

confidence: 99%

“…In order to assess the performance of this system, we took the following steps: first we developed suitable spike-based decoding methods that could be implemented by the neuromorphic processor chip, then we configured the chip to implement these methods in real-time and adapted the bidirectional BMI designed and tested in our lab (Vato et al, 2012) to include in the processing chain this neuromorphic component. Finally, we tested this neuromorphic bidirectional BMI in a closed-loop real-time experimental setup that involved the control of the motion of an external device by the decoded neural signals recorded from the brain of an anesthetized rat.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Bidirectional Brain-Machine Interface Featuring a Neuromorphic Hardware Decoder

et al. 2016

Self Cite

View full text Add to dashboard Cite

Bidirectional brain-machine interfaces (BMIs) establish a two-way direct communication link between the brain and the external world. A decoder translates recorded neural activity into motor commands and an encoder delivers sensory information collected from the environment directly to the brain creating a closed-loop system. These two modules are typically integrated in bulky external devices. However, the clinical support of patients with severe motor and sensory deficits requires compact, low-power, and fully implantable systems that can decode neural signals to control external devices. As a first step toward this goal, we developed a modular bidirectional BMI setup that uses a compact neuromorphic processor as a decoder. On this chip we implemented a network of spiking neurons built using its ultra-low-power mixed-signal analog/digital circuits. On-chip on-line spike-timing-dependent plasticity synapse circuits enabled the network to learn to decode neural signals recorded from the brain into motor outputs controlling the movements of an external device. The modularity of the BMI allowed us to tune the individual components of the setup without modifying the whole system. In this paper, we present the features of this modular BMI and describe how we configured the network of spiking neuron circuits to implement the decoder and to coordinate it with the encoder in an experimental BMI paradigm that connects bidirectionally the brain of an anesthetized rat with an external object. We show that the chip learned the decoding task correctly, allowing the interfaced brain to control the object's trajectories robustly. Based on our demonstration, we propose that neuromorphic technology is mature enough for the development of BMI modules that are sufficiently low-power and compact, while being highly computationally powerful and adaptive.

show abstract

Section: Discussionsupporting

confidence: 89%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Bidirectional Brain-Machine Interface Featuring a Neuromorphic Hardware Decoder

et al. 2016

Self Cite

View full text Add to dashboard Cite

show abstract

“…Reaching prosthetic hand to the target is important aspect in BMI application. Force should be set so that, the artificial hand moves towards the target according to the determined force field [23]. Here we considered the Gaussian force field that, this force was designed to drive the external machine toward an equilibrium state (Eq.…”

Section: Simulation Resultsmentioning

confidence: 99%

Function approximation by hardware spiking neural network

2015

View full text Add to dashboard Cite

Spiking neural networks (SNN) represent a special class of artificial neural networks, where neu-ron models communicate by sequences of spikes. SNNs are often referred to as the third generation of neural networks that highly inspired from natural computing in the brain and recent advances in neuroscience. In this paper we implement biologically-inspired, hardware-realizable SNN architecture using integrate-and-fire units, which is capable of approximating a real-valued function. Based on the results of MATLAB simulations, hardware synthesis and FPGA implementation, it is demonstrated that the implemented hardware can approximate linear and nonlinear functions with low minimum relative error. This framework may represent a fundamental computational unit for the development of artificial SNN, opening new perspectives in pattern recognition tasks.

show abstract

“…As has been confirmed with behavioral experiments (Talwar et al, 2002; Fitzsimmons et al, 2007; London et al, 2008; Do et al, 2011, 2012; Semprini et al, 2012), MiSt has the potential to be used as sensory feedback for closed-loop brain-machine-interface (BMI) devices (Vato et al, 2012; Liao et al, 2013). However, the short- and long-term effects of such artificial input to the brain have not been fully described.…”

Section: Introductionmentioning

confidence: 82%

Short-term dynamics of causal information transfer in thalamocortical networks during natural inputs and microstimulation for somatosensory neuroprosthesis

Semework

DiStasio

2014

Front. Neuroeng.

View full text Add to dashboard Cite

Recording the activity of large populations of neurons requires new methods to analyze and use the large volumes of time series data thus created. Fast and clear methods for finding functional connectivity are an important step toward the goal of understanding neural processing. This problem presents itself readily in somatosensory neuroprosthesis (SSNP) research, which uses microstimulation (MiSt) to activate neural tissue to mimic natural stimuli, and has the capacity to potentiate, depotentiate, or even destroy functional connections. As the aim of SSNP engineering is artificially creating neural responses that resemble those observed during natural inputs, a central goal is describing the influence of MiSt on activity structure among groups of neurons, and how this structure may be altered to affect perception or behavior. In this paper, we demonstrate the concept of Granger causality, combined with maximum likelihood methods, applied to neural signals recorded before, during, and after natural and electrical stimulation. We show how these analyses can be used to evaluate the changing interactions in the thalamocortical somatosensory system in response to repeated perturbation. Using LFPs recorded from the ventral posterolateral thalamus (VPL) and somatosensory cortex (S1) in anesthetized rats, we estimated pair-wise functional interactions between functional microdomains. The preliminary results demonstrate input-dependent modulations in the direction and strength of information flow during and after application of MiSt. Cortico-cortical interactions during cortical MiSt and baseline conditions showed the largest causal influence differences, while there was no statistically significant difference between pre- and post-stimulation baseline causal activities. These functional connectivity changes agree with physiologically accepted communication patterns through the network, and their particular parameters have implications for both rehabilitation and brain—machine interface SSNP applications.

show abstract

Shaping the Dynamics of a Bidirectional Neural Interface

Cited by 24 publications

References 60 publications

A Bidirectional Brain-Machine Interface Featuring a Neuromorphic Hardware Decoder

A Bidirectional Brain-Machine Interface Featuring a Neuromorphic Hardware Decoder

Function approximation by hardware spiking neural network

Short-term dynamics of causal information transfer in thalamocortical networks during natural inputs and microstimulation for somatosensory neuroprosthesis

Contact Info

Product

Resources

About