Modeling Energy-Aware Photoplethysmography Hardware for Personalized Health Care Applications Across Skin Phototypes

ABSTRACT

Photoplethysmography (PPG) sensing is a popular optical method used to determine blood oxygen saturation and heart rate based on light reflected from a patient's skin. Both health metrics are useful in detecting COVID-19 in asymptomatic patients but remain impacted by physiological differences in individuals. In the context of wearable self-powered devices, PPG sensing is relatively high-power compared to available on-body energy. This paper presents a PPG sensing model, with a transimpedance amplifier (TIA) to demonstrate power, signaling, and design tradeoffs, and a photodiode model that includes the impact of a patient's skin phototype on reflected light and PPG sensing accuracy. It also presents preliminary measured results from on-body testing with existing hardware to verify the power and ability to extract a PPG signal at those power levels. This model demonstrates the need to first identify the minimum allowable photodiode current that can produce accurate results for each skin phototype and then determine user-specific circuit knobs to achieve personalized PPG sensing with optimized power consumption. © 2021 IEEE.