Dexterous finger movements: decoding neuro- and myoelectric signals and properties of finger-related neurons in motor cortex

Update Item Information
Title Dexterous finger movements: decoding neuro- and myoelectric signals and properties of finger-related neurons in motor cortex
Publication Type dissertation
School or College College of Engineering
Department Biomedical Engineering
Author Baker, Justin Jeffrey
Date 2010-08
Description The state of the art in prosthetic arm/hand technology has not changed much since the 1950s. The desire to bring more intuitive and less cumbersome control of prosthetic arms/hands has caused neuroprosthetics researchers to investigate the use of myoelectric and neural signal recording devices to gain access to biological signals for the purpose of controlling a prosthetic limb. We hypothesized that the use of Implantable MyoElectric Sensors (IMES), wireless electromyogram (EMG) signal sensors, would allow safe and effective access to signals sufficient to control a dexterous prosthetic hand while avoiding the risks and disadvantages of surface and wired implantable EMG sensors. Therefore, we trained a macaque monkey to perform cued flexions, extensions, and combined flexions of the thumb, index finger, and middle finger using a manipulandum that used microswitches to monitor subtle, low-force finger flexions and extensions. We evaluated the ability to decode the monkey's finger flexion intent based on the recorded EMG that occurred during the behavioral finger movement task with two different algorithms. We concluded that the IMES can provide stable, independent access to EMG from extrinsic finger muscles that could be used to successfully control a dexterous prosthetic hand. We also hypothesized that a Utah Electrode Array (UEA), with only 1 mm long electrodes, implanted in the hand representation of the motor cortex would have enough access to neurons relating to finger movements for an accurate real-time decode of finger movements. Using a maximum likelihood decode, we successfully classified the monkey's finger movements in real-time. With the UEA in the hand representation of the motor cortex, we observed that the microscale organization of the neurons remained highly distributed and overlapping down to the scale of a single electrode on the array (several hundred microns). The finger movement selectivity (FMS) of neurons within the array footprint was seen to change across recording sessions. Some electrodes recorded from neurons of the same FMS more consistently than others. Finally, we observed that one third of the neurons either increased or decreased their firing rate with a small increase in force, while the majority of neurons exhibited no change in firing rate with the force increase.
Type Text
Publisher University of Utah
Subject Brain-computer interface; Decode; Finger; Motor cortex; Neuroprosthetics; Myoelectric prosthesis; Myoelectric signals; Implantable MyoElectric Sensors; IMES
Dissertation Institution University of Utah
Dissertation Name PhD
Language eng
Rights Management ©Justin Jeffrey Baker
Format application/pdf
Format Medium application/pdf
Format Extent 3,456,846 bytes
Source Original in Marriott Library Special Collections, QP6.5 2010 .B34
ARK ark:/87278/s6x92rtm
Setname ir_etd
ID 193021
Reference URL https://collections.lib.utah.edu/ark:/87278/s6x92rtm
Back to Search Results