Robust PPG-based Ambulatory Heart Rate Tracking Algorithm.

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Signal Processing

Background:

• Wearable devices offer convenient Heart Rate (HR) monitoring via Photoplethysmography (PPG) and actigraphy.
• PPG signals are prone to motion artifacts, challenging accurate HR estimation during physical activity.
• Existing algorithms struggle with reliable HR monitoring in real-world, dynamic conditions.

Purpose of the Study:

• To develop a lightweight and accurate HR algorithm for wearable devices.
• To address the limitations of PPG-based HR monitoring during motion and intense activities.
• To validate the proposed algorithm's performance against contemporary methods.

Main Methods:

• A novel Time-domain based method involving Adaptive filtering, Peak detection, Interval tracking, and Refinement (TAPIR) was developed.
• The algorithm utilizes simultaneously acquired PPG and accelerometer signals from wrist-wearable devices.
• TAPIR was evaluated on four diverse, publicly available datasets encompassing various activities and emotional states.

Main Results:

• TAPIR demonstrated significantly higher accuracy (P<0.01) in HR prediction during intense activity compared to existing algorithms.
• The algorithm outperformed methods based on Wiener filtering, time-frequency analysis, and deep learning.
• Validation confirmed TAPIR's clinical-grade performance and suitability for low-power embedded systems.

Conclusions:

• TAPIR offers a robust and accurate solution for continuous HR monitoring in ambulatory settings.
• The algorithm overcomes motion-related challenges inherent in PPG-based HR sensing.
• TAPIR is a valuable tool for real-world health monitoring applications using wearable technology.