公开(公告)号：US20230137595A1

公开(公告)日：2023-05-04

申请号：US17978764

申请日：2022-11-01

Applicant: University of Washington

Inventor： Rajesh P.N. Rao

IPC: A61N1/36 ,  A61N1/02 ,  G16H40/63 ,  G06N20/00

Abstract: Restoring and/or augmenting neural function may be achieved in some examples by receiving signals associated with a first region of a nervous system, and generating a stimulation pattern based on the signals associated with the first region of the nervous system and a first artificial network. The stimulation pattern may be provided to a second region of the nervous system to induce a behavioral output. In some examples, a second artificial network may be used to train the first artificial network. The second artificial network may be configured to predict the behavioral output from the individual based on the stimulation provided by the first artificial network. Parameters of the first artificial network can be adjusted using the output of the second artificial network to optimize the output signals of the first artificial network to achieve restoration and/or augmentation goals.

公开(公告)号：US20240316310A1

公开(公告)日：2024-09-26

申请号：US18608428

申请日：2024-03-18

Applicant: University of Washington ,  SEATTLE CHILDREN'S HOSPITAL DBA SEATTLE CHILDREN'S RESEARCH INSTITUTE

Inventor： Courtnie Paschall ,  Jeffrey Andrew Herron ,  Emmanuel Tanumihardja ,  Rajesh P.N. Rao ,  Jason Scott Hauptman ,  Jeffrey G. Ojemann

IPC: A61M21/02

CPC classification number: A61M21/02 ,  A61M2205/04 ,  A61M2205/33 ,  A61M2210/0693 ,  A61M2230/63

Abstract: An extended reality (XR) neural perceptual feedback (NPF) system includes an XR system and a neural interface. The XR system presents a virtual environment to the user and monitors the user's interactions with the virtual environment. Based on the interaction with the virtual environment, the neural interface applies stimulation to the nervous system of the user. For example, a stimulation pattern may be selected based on the type of interaction, what type of virtual object the interaction was with, or combinations thereof. The stimulation may cause a percept in the user, allowing them to have perceptual feedback from interacting with the virtual environment. The neural interface may also be used to direct interactions within the virtual environment, for example using neural signal decoding to determine how to update the virtual environment.

公开(公告)号：US20240232644A1

公开(公告)日：2024-07-11

申请号：US18408173

申请日：2024-01-09

Applicant: University of Washington

Inventor： Rajesh P.N. Rao ,  Dimitrios C. Gklezakos ,  Vishwas Sathish

IPC: G06N3/096 ,  G06N3/045

CPC classification number: G06N3/096 ,  G06N3/045

Abstract: An active predictive coding network (APCN) is a multi-level network which may break down larger problems into constituent parts. APCNs solve problems compositionally by composing solutions using sequences of previously learned solutions to sub-problems. Considering a pair of adjacent levels, the higher level network includes a state vector and an action vector as well as a state network and action network. At a higher-level ‘macrostep’ the state network updates the state vector and the action network updates the action vector and lower-level state and action networks are generated (or updated) based on the updated state and action vectors respectively. For a number of lower-level ‘microsteps’ the lower-level state network updates a lower-level state vector and the lower-level action network updates a lower-level action vector. The higher-level network may be updated based on the operation of the lower-level network.

公开(公告)号：US20190321639A1

公开(公告)日：2019-10-24

申请号：US16389823

申请日：2019-04-19

Applicant: University of Washington

Inventor： Rajesh P.N. Rao ,  Jeffrey G. Ojemann ,  Eberhard E. Fetz ,  Chet T. Moritz ,  Jeffrey Herron

IPC: A61N1/36 ,  A61N1/05

Abstract: Systems and methods for promoting neuroplasticity in a brain of a subject to improve and/or restore neural function are disclosed herein. One such method includes detecting residual movement and/or muscular activity in a limb of the subject, such as a paretic limb. The method further includes generating a stimulation pattern based on the detected movement and/or muscular activity, and stimulating the brain of the subject with the stimulation pattern. It is expected that delivering stimulation based on the detected residual movement and/or muscular activity of the limb will induce neuroplasticity for restoring neural function, such as control of the limb. A second method involves detecting brain signals and delivering contingent stimulation. A third method involves delivering pairs of successive stimulus patterns to two brain sites, controlled either by preprogrammed sequences or contingent on neural or muscular activity or movement.