Skip to main content
Skip to "About government"
Language selection
Français
Government of Canada /
Gouvernement du Canada
Search
Search the website
Search
Menu
Main
Menu
Jobs and the workplace
Immigration and citizenship
Travel and tourism
Business and industry
Benefits
Health
Taxes
Environment and natural resources
National security and defence
Culture, history and sport
Policing, justice and emergencies
Transport and infrastructure
Canada and the world
Money and finances
Science and innovation
You are here:
Canada.ca
Library and Archives Canada
Services
Services for galleries, libraries, archives and museums (GLAMs)
Theses Canada
Item – Theses Canada
Page Content
Item – Theses Canada
OCLC number
953107294
Link(s) to full text
LAC copy
Author
Roshan Fekr, Atena,
Title
Exploring human respiratory information through a design of a wearable e-Health monitoring system
Degree
Ph. D. -- McGill University, 2016
Publisher
[Montreal] : McGill University Libraries, [2016]
Description
1 online resource
Notes
Thesis supervisor: Katarzyna Radecka (Supervisor2).
Thesis supervisor: Zeljko Zilic (Supervisor1).
Includes bibliographical references.
Abstract
"The measurements of human respiration signal caused by the actions of the chest wall respiratory muscles can help predict health crises. As technology matures, there exists a large potential for effective techniques that can develop the capabilities of health care systems in diagnostics and treatment of respiratory disorders. One recent area of interest is applying wearable Micro-Electro-Mechanical Systems (MEMS) to detect small movements of the body that occur during expansion and contraction of the lungs in each respiration cycle. This thesis presents a newly developed experimental system via wearable sensing technology and wireless communication. In this dissertation, we make use of accelerometer sensors to model the interior and posterior movements of the chest wall during breathing function at rest positions. These motions are analyzed in order to explore different respiratory parameters with high accuracy versus the medical references. To do so, first the problems of self-recalibration of multi-sensory systems as well as fault-tolerant multi-sensor data fusion are considered. Next, an accelerometer-based approach is developed to accurately estimate the breathing signal, respiratory timing variables and the phase shift between chest wall compartments, which is used for paradoxical breathing detection. Since it is essential to determine the critical events caused by sudden rise or fall in per breath tidal volume of the people, a technique is provided to automatically find accurate threshold values according to each individual's breath characteristics. Moreover, we integrate the use of inertial sensors with machine learning techniques to model a wide range of human respiratory patterns for the goal of cloud-based recognition of respiratory problems. Novel approaches are discussed for extracting information-rich features from the respiration signal to improve the performance of the classifiers. Furthermore, a hierarchical tree model is proposed based on multiobjective Evolutionary Algorithm (EA) to optimize two performance metrics of classification, simultaneously. Finally, an innovative biofeedback mechanism is introduced based on Dynamic Time Warping with a fast segmentation method to provide a real-time quantitative feedback during breathing therapy. So that, the proposed platform potentially lifts the people's motivation up towards treatment while precisely tracks their practice quality improvement at low cost."--
Other link(s)
digitool.Library.McGill.CA
escholarship.mcgill.ca
escholarship.mcgill.ca
Subject
Electrical and Computer Engineering
Date modified:
2022-09-01