Assessing the Effects of Filtering Processing on Pulse Wave Transit Time Measured by Photoplethysmography from Earlobe

The waveform of a photoplethysmography (PPG) signal depends on individual physiological conditions. Pulse transition time (PTT), derived from electrocardiogram (ECG) and PPG waveform features, is an important physiological parameter for blood pressure estimation. Filtering can change PPG signal waveform and the timing of PPG feature points. This work aimed to quantitatively investigate the filtering-induced PTT changes at earlobe in healthy subjects of different ages and with different PTT definitions. The ECG and earlobe PPG signals were recorded simultaneously at a sample rate of 2500 Hz for 120 s from 100 healthy adults in 5 age groups (20-29, 30-39, 40-49, 50-59, and over 60 years). The PPG signals were preprocessed using band-pass filtering (pass and stop bands: >0.5 Hz and <0.2 Hz for high-pass filter, <20 Hz and >30 Hz for low-pass filter) and then low-pass filtered (pass and stop bands: <3 Hz and >5 Hz). The PTT difference between the preprocessed and filtered PPG signals was calculated and analyzed. Four pulse feature points were defined and extracted (i.e., maximum peak (MP), valley point (VP), maximal first derivative (M1D), and maximal second derivative (M2D)). The results showed significant effect of PTT definition on filtering-induced PTT relative difference (p<0.001). It was found that the age group (50-59) and the MP definition had the minimum filtering-induce PTT changes. This work suggested that the physiological factors including PTT definition and age should be considered in the PTT-based blood pressure estimation methods.