Multi-frame event dependent locomotion mode classification with FIRNNs

Simple item page
dc.contributor.advisorStevenson, Maryhelen
dc.contributor.advisorEnglehart, Kevin
dc.contributor.authorArsenault, Norman
dc.date.accessioned2023-03-01T16:15:56Z
dc.date.available2023-03-01T16:15:56Z
dc.date.issued2015
dc.date.updated2020-06-02T00:00:00Z
dc.description.abstractUsing electromyography (EMG) data obtained from four transfemoral amputees, Finite Impulse Response Neural Networks (FIRNNs) were investigated for the task of locomotion mode identification. Classification accuracy is found to improve as the duration of the observation interval, presented to the FIRNN, increases. Improvement in classifier accuracy is found to depend on the associated gait event; the more locomotion-mode transitions associated with a gait event, the higher the improvement in the classifier accuracy. Overall, the average classification accuracy on transitions improves by 15.9%. FIRNNs prove much more tolerant to increasing input dimensionality when compared with Linear Discriminant Analysis (LDA) classifiers. When Principal Component Analysis (PCA) is used to reduce input dimensionality, LDA performance is nearly equivalent to FIRNN performance without PCA. A confidence based rejection system is implemented from scaled FIRNN outputs and found to increase classification accuracy and average confidence for the nonrejected patterns. Increasing the observation interval also leads to improved confidence and reduced rejection ratios for fixed decision thresholds.
dc.description.copyright© Norman Arsenault, 2015
dc.description.noteElectronic Only.
dc.description.noteM.Sc. University of New Brunswick, Department of Electrical and Computer Engineering, 2015
dc.formattext/xml
dc.format.extentviii, 77 pages : illustrations
dc.format.mediumelectronic
dc.identifier.oclc(OLoLC)1153170245
dc.identifier.otherThesis 9640
dc.identifier.urihttps://unbscholar.lib.unb.ca/handle/1882/13128
dc.language.isoen_CA
dc.publisherUniversity of New Brunswick
dc.rightshttp://purl.org/coar/access_right/c_abf2
dc.subject.disciplineElectrical and Computer Engineering
dc.subject.lcshNeural networks (Computer science)
dc.subject.lcshElectromyography.
dc.subject.lcshNeural circuitry.
dc.subject.lcshArtificial legs.
dc.subject.lcshGait in humans.
dc.subject.lcshIntelligent control systems.
dc.subject.lcshPattern perception.
dc.subject.lcshMyoelectric prosthesis.
dc.titleMulti-frame event dependent locomotion mode classification with FIRNNs
dc.typemaster thesis
thesis.degree.disciplineElectrical and Computer Engineering
thesis.degree.fullnameMaster of Science in Engineering
thesis.degree.grantorUniversity of New Brunswick
thesis.degree.levelmasters
thesis.degree.nameM.Sc.E.

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
item.pdf
Size:
1.68 MB
Format:
Adobe Portable Document Format
Download

Collections

Open Theses & Dissertations