Morphological Photoplethysmography Features Enhance Stress Detection in Earbud Sensors

Area of Science:

• Physiological sensing and wearable technology
• Biomedical engineering and signal processing
• Health informatics and stress management

Background:

• Long-term stress significantly impacts mental and physical health, driving interest in effective monitoring solutions.
• Earbud-based photoplethysmography (PPG) has shown potential for heart rate variability (HRV) stress detection.
• The utility of PPG morphological features from earbuds for stress detection remains largely unexplored.

Purpose of the Study:

• To evaluate the effectiveness of earbud PPG morphological features for stress detection.
• To compare the performance of PPG morphological features against PPG HRV and ECG HRV features.
• To assess the feasibility of using earbud sensors for stress monitoring.

Main Methods:

• Analysis of physiological data from 97 subjects during stressed and non-stressed periods.
• Training and evaluation of machine learning models using PPG morphological features, PPG HRV features, and ECG HRV features.
• Comparison of classification performance using F1 scores.

Main Results:

• Earbud PPG morphological features achieved a higher F1 score (0.879) than PPG HRV features (0.773) for stress classification.
• A combination of PPG morphological and HRV features (F1 score: 0.880) demonstrated performance comparable to ECG HRV features (F1 score: 0.887).
• Earbud PPG features show comparable stress detection fidelity to ECG, despite hardware limitations.

Conclusions:

• Earbud PPG morphological features offer a promising approach for stress detection.
• Earbud sensors are a viable and user-friendly option for physiological stress monitoring.
• The findings support the use of comfortable, everyday wearable devices for health sensing.