Back to search

University of British Columbia

A system for the monitoring of sleep-related parameters based on inertial measurement units

Abstract

dc:description

Sleep is an essential process needed by the body; as a result, the effects of sleep deprivation are severe and include increased risk of heart diseases and dementia. It is estimated that 40% of Canadians suffer from sleep disorders; nevertheless, most of them are unaware of their condition. Approximately 75–80% of cases of sleep apnea (SA), the most common sleep disorder, are still undiagnosed and thus lacking treatment. However, demand for sleep studies is unmet by the available spaces—high operating costs and expensive equipment limits their availability and the access to diagnosis. Thus, the necessity of inexpensive, yet adequate alternatives for the monitoring of sleep parameters is evident. We present the development of a respiratory effort (RE), body position (BP), and heart rate (HR) monitoring systems for sleep using inertial measurement units (IMU) comprising a 3D accelerometer, gyroscope, and magnetometer. These parameters are required by the American Academy of Sleep Medicine (AASM), and necessary for diagnosis of SA. The chest’s angle variations due to breathing are tracked using all three sensors in the IMU and an extended Kalman filter (EKF) as a data fusion method, thus obtaining the RE in supine and lateral recumbent positions, as well as monitoring BP by tracking gravity vector orientation relative to the body. The system is self-contained, wireless, battery operated and real time. Thus, we improve over previous works that only measure RE limited to a supine position of the body. The HR is estimated by detecting cardiac induced vibrations from a single axis angular rate measuring channel. Using a simple algorithm with an adaptive threshold for peak detection, we obtain good instantaneous HR readings. Although its simplicity sacrifices some accuracy, it is amenable for real-time implementation in a low cost, wireless microcontroller, such as the one utilized for RE/BP. Both tests show promising results, with a high correlation value for RE (e.g. 96.26%) and low error values for HR (e.g. mean 0.35bpm) with respect the reference signals. Furthermore, preliminary tests show the possibility of obtaining all three signals using a single IMU device, suggesting a viable alternative to current technologies.

Degree

thesis:*
Name thesis:degree_name
Master of Applied Science - MASc
Level thesis:degree_level
master's
Discipline thesis:degree_discipline
Electrical and Computer Engineering
Grantor dc:publisher
University of British Columbia
Year dc:date
2018

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Hernández Baladés, Joel Ezequiel

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivatives 4.0 International
Language dc:language
eng

Identifiers

dc:identifier.*
Handle dc:identifier
http://hdl.handle.net/2429/67971

Chain of custody

source
Harvested from
University of British Columbia
Base URL
circle.library.ubc.ca/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Hernández Baladés, Joel Ezequiel. A system for the monitoring of sleep-related parameters based on inertial measurement units. master's thesis, University of British Columbia, 2018. http://hdl.handle.net/2429/67971