Term of Award

Summer 2023

Degree Name

Master of Science, Kinesiology - Exercise Science Concentration

Document Type and Release Option

Thesis (open access)

Copyright Statement / License for Reuse

Digital Commons@Georgia Southern License

Department

Department of Health and Kinesiology

Committee Chair

Li Li

Committee Member 1

Barry Munkasy

Committee Member 2

Barry Joyner

Abstract

Background: Markerless (ML) motion capture systems have recently become available for biomechanics applications. Evidence has indicated the potential feasibility of using an ML system to analyze lower extremity kinematics. However, no research examined ML systems’ estimation of the lower extremity joint moments and powers. Objectives: This study primarily aimed to compare lower extremity joint moments and powers estimated by marker-based (MB) and ML motion capture systems during treadmill running. The secondary purpose was to investigate if movement’s speed would affect the ML’s performance. Methods: Sixteen volunteers ran on a treadmill for 120 s for each trial at the speed of 2.24, 2.91, and 3.58 m/s, respectively. The kinematic data were simultaneously recorded by 8 infrared cameras and 8 high-resolution video cameras. The force data were recorded via an instrumented treadmill. Results: Compared to the MB system, the ML system estimated greater increased hip and knee joint kinetics with faster speeds during the swing phase. Additionally, increased greater ankle joint moments with speed estimated by the ML system were observed at the early swing phase. In contrast, the greater ankle joint powers occurred at the initial stance phase. Conclusions: These observations indicated that inconsistent segment pose estimations (mainly the center of mass estimated by ML was farther away from the relevant distal joint center) might lead to systematic differences in joint moments and powers estimated by MB and ML systems. Despite the promising applications of the ML system in clinical settings, systematic ML overestimation requires extra attention.

OCLC Number

1411060737

Research Data and Supplementary Material

Yes

Files over 10MB may be slow to open. For best results, right-click and select "Save as..."

Share

COinS