Heart rate estimation using on-nail wearable photoplethysmography

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Physiological Monitoring

Background:

• Current wearable heart rate (HR) monitoring primarily relies on photoplethysmography (PPG) sensors placed on the skin.
• Limited research has explored PPG signal acquisition from rigid nail surfaces, such as fingernails and toenails.
• Rigid surfaces may offer enhanced motion robustness compared to flexible skin-based sensor placements.

Purpose of the Study:

• To introduce and evaluate an on-nail wearable photoplethysmography (PPG) sensor for estimating heart rate (HR).
• To assess the feasibility of HR monitoring from fingernail and toenail locations in both motion-free and motion-present conditions.
• To compare the accuracy of the on-nail PPG sensor against commercial electrocardiogram (ECG) and pulse oximeter (PO) devices.

Main Methods:

• Development of an on-nail wearable PPG sensor designed for reflective mode signal acquisition.
• Recruitment of 20 participants for data collection in both static and dynamic (motion) scenarios.
• Comparative analysis of HR estimations from the on-nail PPG sensor against gold-standard ECG and PO measurements.

Main Results:

• The on-nail PPG sensor demonstrated strong correlation with ECG-derived HR, with low root mean square errors (RMSE) of 1.6 bpm (fingernail) and 2.2 bpm (toenail) in motion-free conditions.
• During motion, RMSE increased to 5.6 bpm for fingernails and 12.8 bpm for toenails, indicating reduced accuracy but continued signal acquisition.
• Sensor accuracy was not significantly affected by participant skin tone, suggesting broad applicability.

Conclusions:

• On-nail wearable PPG sensors show significant potential for accurate heart rate monitoring.
• Nail-based PPG offers a feasible alternative for discrete, long-term HR monitoring, potentially integrated into footwear or accessories.
• Further research could optimize sensor design and algorithms to improve motion artifact resilience for continuous monitoring applications.