Towards Cardiac Output Estimation Using Earbud Photoplethysmography Sensor.

Area of Science:

• Biomedical Engineering
• Cardiovascular Physiology

Background:

• Cardiac Output (CO) monitoring is crucial for cardiovascular health assessment.
• Current invasive methods for CO monitoring present limitations, including physical constraints and lack of continuous tracking.
• Non-invasive techniques are sought to overcome these challenges in clinical practice.

Purpose of the Study:

• To develop and evaluate a non-invasive method for continuous Cardiac Output monitoring using photoplethysmography (PPG) from an earbud device.
• To assess the feasibility of improving CO prediction accuracy by incorporating PPG signal quality and morphology features.

Main Methods:

• Utilized an earbud-based PPG sensor for continuous data acquisition.
• Developed a predictive model for CO using PPG signals from 11 subjects.
• Incorporated analysis of PPG signal quality and morphology into the CO prediction model.

Main Results:

• Achieved a Mean Absolute Error (MAE) of 1.53 L/min for CO prediction.
• Attained a Mean Absolute Percentage Error (MAPE) of 18.82% in CO estimation.
• Demonstrated that incorporating signal quality and morphology enhances CO prediction accuracy.

Conclusions:

• Earbud-based PPG provides a viable non-invasive approach for continuous CO monitoring.
• Signal quality and morphological analysis are key factors for improving the accuracy of PPG-based CO estimation.
• This technology has the potential to significantly reduce barriers to non-invasive CO monitoring in various settings.