Use this URL to cite or link to this record in EThOS:
Title: High dimensional surface electromyography and low dimensional muscle synergy in lower limb amputees during transient- and steady-state gait
Author: Mehryar, Pouyan
ISNI:       0000 0004 7425 7191
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Thesis embargoed until 01 Oct 2022
Access from Institution:
The prevalence of lower limb amputation has been rising rapidly with the primary causes associated with dysvascular disease and traumatic injuries. The knowledge of muscle coordination during walking could help in the rehabilitation of individuals with limb loss. The goal of this research was to investigate the neuromuscular differences between healthy subjects (HS) and lower limb amputees during the walking task at different states and speeds using different statistical approaches. High dimensional (HD) electromyography (EMG) data for ten muscles were collected from thirteen healthy subjects’ (HS) dominant leg and eleven transfemoral amputees’ (TFA) intact leg (IL) during transient-state walking at three self-selected speeds (slow, normal, and fast). This data were analyzed at two levels from two different approaches: the HD EMG/muscle activation pattern and low dimensional muscle synergy/modular motor control which were obtained using the linear envelope of EMG signals and concatenated non-negative matrix factorization (CNMF), respectively, from biomechanics and robotic control approach. While the biomechanics approach considers the covariance between the HD muscle activities and low dimensional temporal components of muscle synergy, robotic control accounts for individual muscle activities and temporal components of muscle synergy using statistical parametric mapping (SPM). HD EMG data for ten muscles were also collected from four HS and one transtibial amputee’s (TTA) IL and prosthetic leg (PL) during steady-state walking at a self-selected speed. The muscle synergy was analyzed using the developed CNMF algorithm among legs in pairwise comparisons. The effect of speeds on both HS and TFA muscle activities from biomechanics and robotic control perspectives showed statistically significant differences, suggesting neuromuscular adaptation mechanism in both groups to satisfy the kinematic and kinetic demands of increasing transient-state walking speed. Some differences in HD muscle activities related to the plantarflexors could be observed among the groups, indicating compensatory adjustment of TFA IL for the lack of push off from the PL. The effect of speeds on HS muscle synergy vectors showed reasonable correlations as opposed to those of TFA synergy vectors during transient-state walking. The high correlation suggests that the central nervous system (CNS) activates the same group of muscles synergistically. In comparison among HS dominant leg, TTA IL and TTA PL, the primary muscle(s) had a significant impact on the level of muscle synergy vectors correlation. The activation coefficient profiles suggested that amputees’ IL and PL were significantly different when compared together and to the HS. The same number of synergy groups (=4) found in HS, TFA and TTA indicate analogous complexity implemented by the CNS which does not depend on the state of the gait cycle (transient vs. steady), speed (slow, normal and fast), and level of amputation (below knee vs. above knee). These results have important clinical and robotic control implications. It could provide useful information to therapists to tailor rehabilitation strategy to focus on the muscles and the timing where significant differences occur in the gait cycle. As a result, this could decrease the risk of secondary physical conditions (e.g., osteoarthritis) and increase gait efficiency. The information may also be useful for the prosthetic manufacturers to design prostheses that incorporate information from the IL and/or PL to improve the myoelectric prostheses and develop synergy-based control frame.
Supervisor: Dehghani-Sanij, Abbas ; O'Connor, Rory Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available