Date of Award

9-24-2019

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

First Advisor

Maev, R.

Second Advisor

Balasingam, B.

Rights

info:eu-repo/semantics/openAccess

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Abstract

Accurate heart rate monitoring during intense physical exercise is a challenging problem due to the high levels of motion artifacts (MA) in photoplethysmography (PPG) sensors. PPG is a non-invasive optical sensor that is being used in wearable devices to measure blood flow changes using the property of light reflection and absorption, allowing the extraction of vital signals such as the heart rate (HR). However, the sensor is susceptible to MA which increases during physical activity. This occurs since the frequency range of movement and HR overlaps, difficulting correct HR estimation. For this reason, MA removal has remained an active topic under research. Several approaches have been developed in the recent past and among these, a Kalman filter (KF) based approach showed promising results for an accurate estimation and tracking using PPG sensors. However, this previous tracker was demonstrated for a particular dataset, with manually tuned parameters. Moreover, such trackers do not account for the correct method for fusing data. Such a custom approach might not perform accurately in practical scenarios, where the amount of MA and the heart rate variability (HRV) depend on numerous, unpredictable factors. Thus, an approach to automatically tune the KF based on the Expectation-Maximization (EM) algorithm, with a measurement fusion approach is developed. The applicability of such a method is demonstrated using an open-source PPG database, as well as a developed synthetic generation tool that models PPG and accelerometer (ACC) signals during predetermined physical activities.

Share

COinS