Date of Award

Fall 2024

Project Type

Thesis

Program or Major

Electrical and Computer Engineering

Degree Name

Master of Science

First Advisor

Diliang Chen

Second Advisor

MD Shaaf Mahmud

Third Advisor

John LaCourse

Abstract

Gait analysis is vital in human biomechanicals, clinical diagnostics, and rehabilitation, providing essential insights into human locomotion. Ground reaction forces play a significant role in understanding gait dynamics, detecting abnormalities, and assessing treatment outcomes. Accurate measurement and analysis of ground reaction forces and other gait parameters are crucial for developing effective interventions and improving patient outcomes.

The first part of this thesis explores methods to optimize and standardize the number of pressure sensors and their optimum placements within the insole for accurate and reliable estimation of temporal gait parameters. Standardization of sensor position for temporal gait analysis is crucial for ensuring that data collected from various devices are comparable,enabling broader applicability of the technology in clinical and research settings. Consistent sensor placement allows for reliable replication of results, facilitating longitudinal studies and multi-center trials. Next, a customizable smart insole system was developed to validate the proposed sensor standardization method. Seven temporal gait parameters were analyzed across three different foot sizes and walking speeds.

The final section of the thesis explores advancements in estimating ground reaction forces using a novel center of pressed sensor sensing mechanisms, inertial sensing, and machine learning. The performances of three machine-learning models in estimating ground reaction estimation methods in relation to unseen data are analyzed. Finally, the accuracy and reliability enhancements of the ground reaction force estimation, fusing the inertial sensing data with the center of the pressed sensor measurements, are discussed.

Share

COinS