Motion Artifact Reduction in Wearable Photoplethysmography Based on Multi-Channel Sensors with Multiple Wavelengths

Abstract

Photoplethysmography (PPG) is an easy and convenient method by which to measure heart rate (HR). However, PPG signals that optically measure volumetric changes in blood are not robust to motion artifacts. In this paper, we develop a PPG measuring system based on multi-channel sensors with multiple wavelengths and propose a motion artifact reduction algorithm using independent component analysis (ICA). We also propose a truncated singular value decomposition for 12-channel PPG signals, which contain direction and depth information measured using the developed multi-channel PPG measurement system. The performance of the proposed method is evaluated against the R-peaks of an electrocardiogram in terms of sensitivity (Se), positive predictive value (PPV), and failed detection rate (FDR). The experimental results show that Se, PPV, and FDR were 99%, 99.55%, and 0.45% for walking, 96.28%, 99.24%, and 0.77% for fast walking, and 82.49%, 99.83%, and 0.17% for running, respectively. The evaluation shows that the proposed method is effective in reducing errors in HR estimation from PPG signals with motion artifacts in intensive motion situations such as fast walking and running.

Potential Applications of Mobile and Wearable Devices for Psychological Support During the COVID-19 Pandemic: A Review

“…The result showed a reduction of error in active state HR estimation [123]. Fitbit's PurePulse 2.0 introduces a PPG sensor with multi-wavelength LEDs, which should provide better accuracy and robustness compared to their previous generation PurePulse R , equipped with only green LEDs [123], [124]. The accuracy validation of these newly debut PPG technologies is expected to stimulate the interest for further research.…”

“…A study has demonstrated the use of a similar LED configuration as Apple Watch Series 6 (4 cluster LEDs (green, red, and infrared)). The result showed a reduction of error in active state HR estimation [123]. Fitbit's PurePulse 2.0 introduces a PPG sensor with multi-wavelength LEDs, which should provide better accuracy and robustness compared to their previous generation PurePulse R , equipped with only green LEDs [123], [124].…”

“…By utilizing multiple light wavelengths in the PPG sensor, the blood pulsation information from different blood vessels can be measured. This result in better working performance and a wider range of application (e.g., blood pressure approximation) [123], [124]. Therefore, the later generation of PPG sensors in the smartwatch is equipped with multiple LEDs with different wavelength (generally green, red, and infrared LEDs) [109]- [112].…”

Seamlessly Integrable Optoelectronics for Clinical Grade Wearables

“…The location freedom of PPG allows designs that can be worn as a ring or bracelet to minimize the motion artefacts; however, due to the usage of rigid components, consumer wearables are shown to be only accurate only in limited conditions such as resting and slow walking. [ 29 ] Therefore, PPG measurements of wafer‐based wearables still require post processing by software algorithms to eliminate present motion artifacts. Algorithm based approaches generally offers case specific solutions that remain insufficient for realistic cases including randomly interfered multiple noise sources.…”

On-Device Filter Design for Self-Identifying Inaccurate Heart Rate Readings on Wrist-Worn PPG Sensors

“…Zhang et al suggest the motion artifact reduction algorithm using green PPG signal and IR PPG signal using wavelet transform and signal reconstruction. Lee et al suggest combining 12 channel PPG signals generated by four PPG sensors each containing red, green and infrared LEDs [26].…”