A PPG Signal Dataset Collected in Semi-Naturalistic Settings Using Galaxy Watch

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Health Informatics

Background:

• Consumer wearables offer continuous, non-invasive health monitoring via photoplethysmography (PPG).
• Existing datasets often use research-grade devices, limiting real-world applicability of consumer-grade wearables.
• There is a need for datasets reflecting real-world usage of consumer wearables.

Purpose of the Study:

• To present GalaxyPPG, a novel dataset for assessing consumer-grade wearable sensors.
• To evaluate the sensing performance of wrist-worn PPG devices during diverse activities.
• To facilitate research in PPG-based heart rate and heart rate variability monitoring.

Main Methods:

• Collected data from 24 participants using Galaxy Watch 5, Empatica E4, and Polar H10 (ECG).
• Acquired wrist-worn PPG and chest-worn ECG signals during various motion- and stress-inducing activities.
• Developed an open-source toolkit for data collection and analysis with Samsung Galaxy Watch.

Main Results:

• The GalaxyPPG dataset captures PPG and ECG data in semi-naturalistic settings.
• The data provides insights into consumer wearable performance under dynamic conditions.
• The open-source toolkit supports reproducible research using Samsung Galaxy Watch.

Conclusions:

• The GalaxyPPG dataset addresses the gap in real-world consumer wearable data.
• This resource can advance PPG-based applications like heart rate tracking and stress detection.
• The open-source toolkit promotes further research and development in wearable health monitoring.