A Bayesian Multi-Scale Framework for Photoplethysmogram Imaging Waveform Processing

Abstract

Listen

Translate article

Photoplethysmography imaging (PPGI) is an increasingly populartechnique for remotely creating signals with a plethora of medicalinformation, referred to as PPGI waveforms. However, PPGI waveformsare often heavily affected by illumination variation and motionartefacts. Current PPGI waveform processing methods are usefulfor estimating heart rate, however, structural detail is not preserved,rendering the signal incapable of providing additional medical information.For this reason, we propose a multi-scale framework basedon the Bayesian residual transform which aims to suppress noiseand preserve structural details necessary for extracting cardiovascularinformation beyond the scope of heart rate. Experiments conductedon a dataset consisting of 24 different PPGI waveforms andcorresponding PPG waveforms captured via a finger pulse oximetersuggests a high level of noise and ambient illumination variationsuppression is achieved while signal fidelity is largely retained.