Cardiac arrhythmia detection using photoplethysmography

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Cardiovascular Diseases (CVDs) are a leading cause of mortality globally.
• Cardiac arrhythmias significantly contribute to CVD-related deaths.
• Current electrocardiogram (ECG) monitoring faces limitations in wearable applications and generates false alarms.

Purpose of the Study:

• To develop and evaluate a technique for detecting life-threatening cardiac arrhythmias using photoplethysmography (PPG) signals.
• To assess the viability of PPG as a wearable alternative to ECG for arrhythmia detection.
• To identify five specific arrhythmias: tachycardia, bradycardia, asystole, ventricular tachycardia, and ventricular fibrillation.

Main Methods:

• Utilized PhysioNet Challenge 2015 data for arrhythmia detection.
• Developed a novel technique to assign a pulse quality index to each PPG pulse.
• Focused analysis on the highest quality segments of the PPG signal for improved detection accuracy.

Main Results:

• Achieved an overall true positive rate (TPR) of 93% for arrhythmia detection.
• Obtained a true negative rate (TNR) of 53.78%.
• Demonstrated that PPG signals can be effectively used for identifying various arrhythmias.

Conclusions:

• Photoplethysmography (PPG) presents a viable and wearable alternative to traditional ECG for detecting critical cardiac arrhythmias.
• The proposed pulse quality indexing method enhances the reliability of PPG-based arrhythmia detection.
• PPG technology holds promise for improved remote and continuous cardiac monitoring.