Methods of Extracting Feature from Photoplethysmogram Waveform for Non-Invasive Diagnostic Applications

Abstract

<p>This paper presents a bibliographical survey of recently-published research on different techniques to extract feature from photoplethysmogram (PPG). These techniques and approaches have been implemented for better accuracy in detecting diseases. Moreover, several aspects in analyzing PPG waveform are discussed on the techniques in feature extraction, parameters involved and performance comparisons. This review will serve as a comparative study and reference for researches working on PPG waveform in health care applications.</p>

Age-Related Changes in the Characteristics of the Elderly Females Using the Signal Features of an Earlobe Photoplethysmogram

“…These values correspond to the peak amplitudes, and the indicators for the ratio of amplitude inflection sections or duration were used as evaluation variables. In general, the a- and b-waves correspond to the early systolic component with minimal influence from the reflection wave, the c-wave corresponds to the mid-systolic component, the d-wave corresponds to the post-systolic component, and the e-wave corresponds to the position of the diastolic component after the dicrotic notch wave or the position of the diastolic component [ 18 ]. However, it is difficult to detect the dichroic notch and the subsequent diastolic components because the wave does not have an inflection point corresponding to the age or PPG measurement location [ 19 ].…”

“…The c-wave at the third derivative is defined as the systolic component (p1), and the d-wave is the late systolic or early diastolic component (p2) [ 17 , 18 ]. In the case of the third derivative, an advantage is that the inflection point of the second derivative can be identified more clearly.…”

PPGFeat: a novel MATLAB toolbox for extracting PPG fiducial points

“…The non-invasive technique is used to measure vital signs ( Allen, 2007 ) and has been described in studies as a tool for monitoring heart rate, blood pressure, and respiratory rate ( Fukushima et al, 2012 ; Karlen et al, 2013 ; He et al, 2014 ; Islam et al, 2017 ), and to identify persons with risk of diseases ( Mahri et al, 2017 ; Chakraborty et al, 2020 ; Charlton et al, 2022b ; Lee et al, 2022 ). However, extracting relevant features from the PPG signal is a challenging task ( Elgendi et al, 2018 ; Chakraborty et al, 2019 ; Ab Hamid and Nayan, 2020 ).…”

“…The PPG waveform has been widely studied for a long time. Various feature extraction methods have been proposed in the literature ( Elgendi et al, 2018 ; Chakraborty et al, 2019 ; Ab Hamid and Nayan, 2020 ). PPG is included in several wearable devices due to its non-complex setup and usefulness in the collection of a number of different and valuable physiological parameters ( Almarshad et al, 2022 ).…”