Fuzzy-genetic photoplethysmograph peak detection

Abstract

Photoplethysmography (PPG) promises noninvasive body metrics measurement, especially that of heart rate. However, this system is prone to noise due to motion artifacts. This paper presents a fuzzy inference system, with membership functions and rules tuned by a genetic algorithm that utilizes the principal components of the PPG data accelerometer data from the x, y, and z coordinates in order to recover the peaks from the distorted PPG signal. A comparative test demonstrated that a 56.66% peak-to-peak correspondence to a reference ECG signal is achievable with the fuzzy-genetic system in place.