Graduation Semester and Year

2020

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Computer Engineering

Department

Computer Science and Engineering

First Advisor

Gergely Zaruba

Second Advisor

Manfred Huber

Abstract

Spatial and temporal gait analysis can provide useful measures for determining a person’s state of health while also identifying deviations in day-to-day activity. The SmartCare project is a multi-discipline health technologies project that aims to provide an unobtrusive and pervasive system that provides in-home health monitoring for the elderly. This research work focuses on the pressure-sensitive smart floor of the SmartCare project by using an experimental floor to develop methods for future use on a floor deployed within a home. This work presents a procedure to automatically calibrate a smart floor’s pressure sensors without specialized physical effort. The calibration algorithm automatically filters out non-human static weight and only retains weight generated by human activity. This technique is designed to correctly translate sensor values to kg weight units even when direct independent access to the pressure sensors is prohibited and when a shared tile floor sits above the sensor grid. Using the calibrated sensor values, machine learning techniques are used to extract individual contact points on a smart floor generated by a person’s walking cycle. This work presents a three-step process of building and training neural network models of different architectures (feedforward, convolutional, and autoencoder) to learn the unique non-linear relationship between weight distribution, tile coupling, and physical floor variations. Finally, this research work presents a recursive Hierarchical Clustering Analysis algorithm that uses the individual contact points generated by the floor model to extract individual footfalls of a person during a walking cycle. The footfall clusters are further grouped and segmented into walking sequences. Spatial gait analysis is performed on the resulting footfall clusters within each walking sequence to measure a variety of gait parameters. The results of the gait analysis are compared to those generated by a high-resolution mat alternative showing comparable results for most of the computed gait metrics.

Keywords

Machine learning, Smart floor, In-home healthcare, Assistive health technologies, Neural networks, Clustering, Gait analysis, Sensor calibration, SmartCare

Disciplines

Computer Sciences | Physical Sciences and Mathematics

Comments

Degree granted by The University of Texas at Arlington

29365-2.zip (3637 kB)

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.