The New Jersey Institute of Technology's
Electronic Theses & Dissertations Project

Title:	Comparison of muscle synergies elicited from transcranial meganetic stimulation (tms) and voluntary movements
Author:	Wei, Yifei
View Online:	njit-etd2016-085 (x, 54 pages ~ 1.5 MB pdf)
Department:	Department of Biomedical Engineering
Degree:	Master of Science
Program:	Biomedical Engineering
Document Type:	Thesis
Advisory Committee:	Adamovich, Sergei (Committee chair) Sahin, Mesut (Committee member) Roman, Max (Committee member)
Date:	2016-01
Keywords:	Muscle synergy hypothesis Muscle synergies Transcranial magnetic stimulation
Availability:	Unrestricted
Abstract:	A key question in motor control is the redundancy of musculoskeletal elements involved. This problem refers to as the degree of freedom problem. The Muscle Synergy Hypothesis is one of the hypotheses that aim to resolve the problem which defines that a muscle synergy is a combination of a small set of muscles activated at different levels, serving as a building block that constructs motor behaviors. A recent study (Overduin et al. 2012) demonstrated that muscle synergies decomposed by Nonnegative Matrix Factorization (NMF) from EMG patterns evoked by intra-cortical microsimulation (ICMS) in the monkey remarkably matched ones observed in naturalistic reach-and-grasp behaviors. Another study (Ajiboye et al. 2009) showed that synergies elicited from a small number of hand postures can allow prediction of hand postures in general. Inspired by aforementioned studies, the aim of this study was to investigate whether Transcranial Magnetic Stimulation (TMS) can elicit muscle synergies matching ones observed in voluntary movements in healthy human subjects and whether these synergies can serve as frameworks to predict EMG patterns evoked by either TMS or voluntary movements. Five healthy right-handed subjects participated in the study. 8 hand muscles were recorded to capture either TMS-evoked motor evoked potential (MEP) and electromyography (EMG) resulted from subjects’ shaping American Sign Language (ASL) letters and numbers. NMF was utilized to extract synergies from both MEP and EMG data. We observed 5 or 6 synergies can capture 90% of variance of original and matched synergies of two classes. The reconstructions of the original datasets (VTMS: MEP data; Vvol: EMG data; Vrand: Random data as control) from synergies (Hvol synergies elicited from ASL tasks; HTMS synergies elicited from TMS) was done by the nonnegative least-square algorithm, and Proportion of Variance Accounted for (PAV) served as a measure to quantify the quality of the estimation, giving results Hvol -> Vvol: 0.92±0.02; HTMS -> VTMS: 0.94±0.02; Hvol -> Vrand: 0.53±0.03; HTMS -> Vrand: 0.53±0.07; HTMS -> Vvol: 0.70±0.06; Hvol -> VTMS: 0.79±0.06. In conclusion, we argue that cortical components may involve in encoding synergies and we also demonstrate the possibility of synergies serving as frameworks in predicting and explaining human hand postures in general.