Date of Award

5-2017

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Biomedical Engineering

Major Professor

Jeffrey A. Reinbolt

Committee Members

JAM Boulet, Eric Wade, Songning Zhang

Abstract

Roughly 47.5 million people in the US have a disability, with 8.6 million reporting arthritis as their main cause of disability, making arthritis the leading cause of physical disability. With decreased mortality rates and a large, aging baby boomer generation, there will be more adults living with chronic musculoskeletal conditions causing disabilities that limit walking. Since walking ability is directly related to an individual’s independence at home and in the community, losing this ability is a major setback for patients with arthritis. Knee osteoarthritis (OA) is the most prevalent form of arthritis affecting approximately 27 million adults and accounts for over 55% of all arthritis-related hospital admissions. OA is a highly painful disease with treatments limited to pain management. However, gait modification has recently shown promise as an early intervention treatment strategy to slow disease progression. Thus, the objective of this dissertation is to investigate subject-specific gait modifications to minimize joint loads for treating patients with knee OA.

The first study in this dissertation relies heavily on the development of subject-specific musculoskeletal models to analyze muscle forces and joint contact loads during toe-in gait modification for subjects with knee OA. This study will generate muscle-actuated, dynamic simulations to estimate muscle forces and internal joint contact loads during gait. The results of this study will aid in the advancement of gait modification as a treatment strategy for knee OA. The last two studies will employ machine learning and optimization techniques— specifically, forward sequential feature selection and surrogate-based optimization— to evaluate toe-in gait modification and improve its efficacy for use as a treatment strategy for knee OA. The goal will be to develop testable subject-specific gait modification patterns that reduce joint loads.

The use of both dynamic simulations and data mining techniques provides a unique approach to investigating the relationship between joint biomechanics and muscle function and joint contact loads with respect to gait modification. This approach has the potential to gain much needed insight into the underlying mechanism of gait modification and help advance research to create subject-specific gait modification patterns for treating knee OA in the future.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS