Date of Award
Master of Science (Computer Science)
School of Science
Professor Kamal Alameh
Dr Srenten Askraba
Photoplethysmography is an optical technique for measuring the perfusion of blood in skin and tissue arterial vessels. Due to its simplicity, accessibility and abundance of information on an individual’s cardiovascular system, it has been a pervasive topic of research within recent years. With these benefits however there are many challenges concerning the processing and conditioning of the signal in order to allow information to be extracted. One such challenge is removing the baseline drift of the signal, which is caused by respiratory rate, muscle tremor and physiological changes within the body as a response to various stimuli.
Over the years there have been many methods developed in order to condition the signal such as Wavelet Transform, Cubic Spline Interpolation, Morphological Operators and Fourier-Based filtering techniques. All have their own individual benefits and drawbacks. These drawbacks are that they are unsuitable for real-time usage due to the computation power needed, or have the trade-off of being real-time at the cost of deforming the signal which is unideal for accurate analysis. This thesis aims to explore these techniques in order to develop an algorithm that can be used to condition the signal against the baseline drift in real-time, while being able to achieve good computational efficiency and the preservation of the signal form.
Nguyen, T. N. (2016). An algorithm for extracting the PPG Baseline Drift in real-time. Retrieved from http://ro.ecu.edu.au/theses/1801