Smartwatch PPG Peak Detection Method for Sinus Rhythm and Cardiac Arrhythmia

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Photoplethysmography (PPG) is a non-invasive optical technique used for vital signs monitoring.
• Accurate heart rate (HR) estimation from PPG signals is challenging during cardiac arrhythmias.
• Existing PPG peak detection algorithms are often limited to normal sinus rhythms (NSR).

Purpose of the Study:

• To develop an advanced PPG peak detection algorithm for smartwatches capable of discriminating various cardiac rhythms.
• To enhance the accuracy of atrial fibrillation (AF) screening by enabling reliable HR estimation during arrhythmias.
• To improve beat-to-beat and mean HR estimation compared to traditional algorithms.

Main Methods:

• A sequential series of algorithms was designed to detect and discriminate between normal sinus rhythm (NSR), premature atrial contraction (PAC), premature ventricle contraction (PVC), and atrial fibrillation (AF).
• A novel Poincaré plot scheme was incorporated to differentiate AF with Rapid Ventricular Response (RVR) from AF with normal basal heart rate.
• The algorithm was validated for its ability to differentiate PAC/PVC from NSR and AF.

Main Results:

• The proposed algorithm demonstrated significantly lower average beat-to-beat estimation error (over 40% reduction) compared to a traditional NSR-focused algorithm.
• Mean heart rate estimation error was also substantially reduced (over 50% reduction) using the new approach.
• The algorithm successfully accounted for various arrhythmias, outperforming algorithms designed solely for NSR.

Conclusions:

• The developed PPG peak detection algorithm offers superior accuracy in HR estimation across diverse cardiac rhythms, including arrhythmias.
• This advancement is crucial for enhancing the diagnostic capabilities of smartwatch-based AF screening tools.
• The novel method provides a more robust and reliable approach to wearable cardiovascular monitoring.