Leland Stanford Junior University

Cambridge Enterprise Limited

Jonathan C Kao

Krishna V Shenoy

Sergey Stavisky

David Sussillo

Paul Nuyujukian

John Patrick Cunningham

Nir Even-Chen

Mark M Churchland

Chethan Pandarinath

A brain machine interface (BMI) to control a device is provided. The BMI has a neural decoder, which is a neural to kinematic mapping function with neural signals as input to the neural decoder and kinematics to control the device as output of the neural decoder. The neural decoder is based on a continuous-time multiplicative recurrent neural network, which has been trained as a neural to kinematic mapping function. An advantage of the invention is the robustness of the decoder to perturbations in the neural data; its performance degrades less—or not at all in some circumstances—in comparison to the current state decoders. These perturbations make the current use of BMI in a clinical setting extremely challenging. This invention helps to ameliorate this problem. The robustness of the neural decoder does not come at the cost of some performance, in fact an improvement in performance is observed.

Systems and methods for intracortical brain machine interface decoding

Multiplicative recurrent neural network for fast and robust intracortical brain machine interface decoders

A brain machine interface (BMI) to control a device is provided. The BMI has a neural decoder, which is a neural to kinematic mapping function with neural signals as input to the neural decoder and kinematics to control the device as output of the neural decoder. The neural decoder is based on a continuous-time multiplicative recurrent neural network, which has been trained as a neural to kinematic mapping function. An advantage of the invention is the robustness of the decoder to perturbations in the neural data; its performance degrades less&#x2014;or not at all in some circumstances&#x2014;in comparison to the current state decoders. These perturbations make the current use of BMI in a clinical setting extremely challenging. This invention helps to ameliorate this problem. The robustness of the neural decoder does not come at the cost of some performance, in fact an improvement in performance is observed.

A brain machine interface (BMI) for improving a performance of a subject is provided. The BMI has two decoders that act in real-time and in parallel to each other. The first decoder is for intention execution of a subject's intention. The second decoder is for error detection in a closed-loop error fashion with the first detector and to improve the performance of the first detector. Embodiments of this invention may enable an entirely new way to substantially increase the performance and robustness, user experience, and ultimately the clinical viability of BMI systems.

Task-outcome error signals and their use in brain-machine interfaces

A brain machine interface (BMI) for restoring performance of poorly performing decoders is provided. The BMI has a decoder for decoding neural signals for controlling the brain machine interface. The decoder separates in part neural signals associated with a direction of movement and neural signals associated with a speed of movement of the brain machine interface. The decoder assigns relatively greater weight to the neural signals associated with a direction of movement.

Brain-machine interface utilizing interventions to emphasize aspects of neural variance and decode speed and angle using a kinematics feedback filter that applies a covariance matrix

A brain machine interface for control of prosthetic devices is provided. In its control, the interface utilizes parallel control of a continuous decoder and a discrete action state decoder. In the discrete decoding, we not only learn states affiliated with the task, but also states related to the velocity of the prosthetic device and the engagement of the user. Moreover, we not only learn the distributions of the neural signals in these states, but we also learn the interactions/transitions between the states, which is crucial to enabling a relatively higher level of performance of the prosthetic device. Embodiments according to this parallel control system enable us to reliably decode not just task-related states, but any 'discrete action state,' in parallel with a neural prosthetic 'continuous decoder,' to achieve new state-of-the-art levels of performance in brain-machine interfaces.

Brain machine interface utilizing a discrete action state decoder in parallel with a continuous decoder for a neural prosthetic device

A brain-machine interface is provided that incorporates a neural dynamical structure in the control of a prosthetic device to restore motor function and is able to significantly enhance the control performance compared to existing technologies. In one example, a neural dynamical state is inferred from neural observations, which are obtained from a neural implant. In another example, the neural dynamical state can be inferred from both the obtained neural observations and from the kinematics. A controller interfaced with the prosthetic device uses the inferred neural dynamical state as input to the controller to control kinematic variables of the prosthetic device.

Growing community of inventors

Stanford, CA, United States of America

Jonathan C Kao