Back to search

University of Leeds

Smart neural bridge: Algorithms to drive a brain machine interface for control of the paralyzed limb

Abstract

dc:description.abstract

Treating damage to the nervous system is limited to targeting natural recovery, with limited recourse if these mechanisms fail. The level of disruption between descending systems and motor effectors is the major factor determining recovery in SCI. Attempts have been made to reconnect descending motor signals with the correct muscles but have struggled to restore motor control to pre-injury states. This thesis has tried to address limitations in current devices by designing algorithms to control a brain machine interface at the level of the spinal cord that provides functional electrical stimulation in a closed loop manner incorporating synergy information extracted from muscle activity. Synergy information is identified using dimensionality reduction algorithms. To determine which method was suitable for online analysis the accuracy of commonly used algorithms for synergy extraction and activity onset detection were compared. Findings from this comparison challenge assumptions regarding the utility of various methods. The most accurate algorithm, non-negative matrix factorization, was implemented online and applied to isometric knee extensions at different angles. It was shown that in contrast to the accepted view, proprioceptive feedback plays a significant role in synergy recruitment. Using an interneuron population model, the experimentally observed synergies were reproduced using only changes in afferent feedback. Using muscle synergies as a target for motor control requires a method for generating specific electromyography waveforms. An artificial neural network successfully learned the relationship between stimulation parameters and electromyography for stimulation of the rat hind limb. These algorithms were combined in a simulated injury environment using the same interneuron model described previously with connections removed or reduced. The combined algorithms were able to successfully restore muscle synergies to normal levels in some injury conditions. These algorithms represent a system that uses closed loop control of muscle synergy recruitment that could be implemented in a variety of devices.

Degree

thesis:*
Name dc:type.qualificationname
Ph.D
Level dc:type.qualificationlevel
doctoral
Grantor dc:publisher.institution
University of Leeds
Year dc:date.issued
2020

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • York, Gareth James Richard
Advisors dc:contributor.advisor
  • Chakrabarty, Samit
  • Steenson, Paul
  • Kim, Jongrae

Rights

Language dc:language.dcterms_iso639_2
eng

Identifiers

dc:identifier.*
Identifier
uk.bl.ethos.811281
Dc Identifier Grantnumber
1724438

Chain of custody

source
Harvested from
White Rose University Consortium
Base URL
etheses.whiterose.ac.uk/cgi/oai2
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
related terms
citation

York, Gareth James Richard. Smart neural bridge: Algorithms to drive a brain machine interface for control of the paralyzed limb. doctoral thesis, University of Leeds, 2020.