Date of Award

6-2025

Degree Name

MS in Mechanical Engineering

Department/Program

Mechanical Engineering

College

College of Engineering

Advisor

Stephen Klisch

Advisor Department

Mechanical Engineering

Advisor College

College of Engineering

Abstract

For many clinical populations, the measurement of balance is a critical task. Previous studies have agreed that smartphone IMUs provide a valid measure of balance that can be significantly correlated with force plate measurements and is sensitive enough to distinguish between various balance poses, but many balance assessment protocols require uncommon equipment or the use of hands during assessment. These limit accessibility of balance assessment. To eliminate the need for hands or any additional equipment, this study assessed the use of a smartphone recording the root mean square of 3D acceleration magnitude. The goal of the study was to determine if acceleration recordings taken from a pants pocket or shirt pocket were sufficiently sensitive to identify differences in balance poses of increasing difficulty, and if those measurements were correlated with previously validated methods. An iPhone app was also developed to allow future users to self-assess balance without any additional equipment. Results collected from a shirt pocket were found to be significantly different between all three balance poses tested (p < 0.0237) and results collected from a pants pocket were significantly different between two of the three poses (p < 0.0006). All differences matched expected order based on difficulty of poses. For each pocket, at least two of the three poses produced results that were significantly and either moderately or strongly correlated to the previously validated method (0.45 < ρ < 0.70, p < 0.0410. As an exploratory finding, there was also found to be no significant correlation between participant height and acceleration results. The iPhone app developed alongside this study produced results strongly correlated (r = 0.9991, p < 0.0001) with the measurement technique used for the rest of the study. Overall, the hands-free balance assessment protocol provided a valid measure of balance, able to identify small changes and mostly correlated with previous techniques, particularly when measured with the phone in a shirt pocket. The iPhone app developed alongside this study produced similar results and will allow future at-home assessment.

Download

Share

COinS