Semester

Spring

Date of Graduation

2020

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Industrial and Managements Systems Engineering

Committee Chair

Ashish Nimbarte

Committee Member

Anna Allen

Committee Member

Hongwei Hsiao

Committee Member

Gary Winn

Committee Member

Feng Yang

Abstract

The impact of work-related musculoskeletal disorders (WMSDs) is enormous due to a combination of direct and indirect costs associated with healthcare, lost workdays and human suffering. Because of the established relationship between Localized Muscle Fatigue (LMF) development and WMSDs, and in order to reduce and/or prevent WMSDs in workplaces, different fatigue assessment methods have been developed. Surface Electromyography (SEMG) is a commonly used LMF assessment technique. The SEMG signals are typically analyzed in time and frequency domains to predict LMF based on a relative change with respect to initial, or under no-fatigue conditions. Quantifying such change, however, relies on the assumption that the SEMG measures without fatigue present, under different muscular demands, can serve as an appropriate reference within the joint range-of-motion. To our knowledge, the assumption that the electromyographic measures do not change/vary due to factors other than LMF has not been thoroughly tested. Therefore, the objective of this study was to quantify variability of various SEMG measures in non-fatigued shoulder muscles and its implication for assessing muscle fatigue.

In the first Specific Aim, an experiment was performed to quantify variability of six EMG measures (RMS, MAV, ZC, MnPF, MdPF, and PFB11-22 Hz) in seven non-fatigued shoulder muscles. Twelve human participants performed 120 occupationally relevant static holding tasks. The variability in SEMG data was quantified using Mean Square Error (√MSE) obtained from ANOVA models. The SEMG measures were found to vary between 5.32% to 12.25% due to factors other than muscle fatigue. The narrowest range of variability was observed for ZC (10.20% to 11.00%), and the largest range of variability was observed for MdPF (8.72% to 12.25%).

In the second Specific Aim, a relationship between SEMG variability and LMF based on perceived exertion ratings was studied. Twelve human participants performed 8 fatigue inducing exertions for 10-45 seconds. The data were analyzed to identify muscle fatigue onset based on the perceived exertion ratings and the corresponding relative changes in SEMG measures. A good agreement was observed between the definition of LMF based on perceived exertion ratings and the relative change in the SEMG measures (quantified in Aim 1) for ZC, MnPF, and MdPF. And the study concludes that for the shoulder muscles a change higher than 11.00%, 11.45%, and 12.25% in ZC, MnPF, and MdPF, respectively, can be an indication of LMF.

In conclusion, the study findings suggest that a change higher than 11.00%, 11.45%, and 12.25% in ZC, MnPF, and MdPF, respectively, can be an indication of LMF. These findings could be useful in improving real-time fatigue predication models and/or methods to curtail the incidence of LMF based WMSDs in workplaces.

Download

Share

COinS