A NEURAL PROSTHESIS FOR RESTORING WALKING AFTER SPINAL CORD INJURY

Simple item page
dc.contributor.advisorEtienne-Cummings, Ralphen_US
dc.contributor.committeeMemberAndreou, Andreas G.en_US
dc.contributor.committeeMemberFoster, Amy C.en_US
dc.creatorMazurek, Kevin Andrewen_US
dc.date.accessioned2014-12-23T04:37:37Z
dc.date.available2014-12-23T04:37:37Z
dc.date.created2014-05en_US
dc.date.issued2013-12-12en_US
dc.date.submittedMay 2014en_US
dc.description.abstractThe goal of this doctoral work was to develop a hardware neural prosthesis capable of producing temporally changing current stimulation patterns for restoring walking in individuals with spinal cord injury (SCI). To solve this problem, a suitable control strategy must be developed which can adapt to environmental and internal perturbations (e.g. change in surface friction or muscle fatigue). The biological central pattern generator (CPG) for locomotion is a neural circuit capable of producing alternating patterns of flexion and extension. This acts as the nervous system's controller for producing walking movements as it integrates descending neural commands from the brain as well as feedback from the dorsal columns (afferent feedback). This combination of afferent and efferent communication can be modeled as a system which integrates open and closed loop control to produce walking movements. A control strategy was first designed in software using a state-space approach where the states represented functional movements observed during healthy walking (e.g. swing, stance). The state transitions took the form of IF-THEN rules which could monitor transitions such as swing-to-stance, stance-to-swing, and muscle fatigue. This strategy was successfully tested in vivo in an anesthetized cat model of SCI using a functional electrical stimulation (FES) technique called intramuscular stimulation (IMS) where the muscles in the legs were electrically stimulated to ex or extend the hip, knee, and ankle joints. With a control strategy in place, a hardware solution in the form of an application specific integrated circuit was developed called a Locomotion Processing Unit (LPU). The LPU was programmable and realized several configurations of the state-based control rather than the specific strategy tested. The states temporally shaped the stimulation waveform and transitioned based on external sensory signals or internal timing rules. The output of the LPU contained current stimulating circuits for intraspinal microstimulation (ISMS) which activates neural circuits in the spinal cord to produce biofidelic movements. This hardware solution was successfully tested in vivo to produce over-ground walking in another anesthetized cat model of SCI to the form of 121.8m of cumulative walking. It produced comparable results to a desktop system which generated 698.9m of walking during the same experiment. This neural prosthesis represented the first time that ISMS produced walking without the need for external hardware such as a workstation PC or data acquisition boards. This prototype can lead towards an implantable neural prosthesis capable of restoring walking in individuals with SCI, potentially translating this technology from bench to bedside for clinical application.en_US
dc.format.mimetypeapplication/pdfen_US
dc.identifier.urihttp://jhir.library.jhu.edu/handle/1774.2/36990
dc.languageen
dc.publisherJohns Hopkins University
dc.subjectfunctional electrical stimulationen_US
dc.subjectintraspinal microstimulationen_US
dc.subjectlocomotionen_US
dc.subjectneural prosthesisen_US
dc.titleA NEURAL PROSTHESIS FOR RESTORING WALKING AFTER SPINAL CORD INJURYen_US
dc.typeThesisen_US
dc.type.materialtexten_US
thesis.degree.departmentElectrical and Computer Engineeringen_US
thesis.degree.disciplineElectrical Engineeringen_US
thesis.degree.grantorJohns Hopkins Universityen_US
thesis.degree.grantorWhiting School of Engineeringen_US
thesis.degree.levelDoctoralen_US
thesis.degree.namePh.D.en_US
Files
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
LICENSE.txt
Size:
2.68 KB
Format:
Plain Text
Description:
Download
Collections
ETD -- Doctoral Dissertations