Non-Invasive Camera-Based Cardiac Output Monitoring via Facial Photoplethysmography: Advancing Hemodynamic Assessment
Summary
This study introduces a non-contact method for monitoring cardiac output (CO) using facial photoplethysmography (PPG). This camera-based system shows promise for hemodynamic assessment in critical care settings.
Area of Science:
- Biomedical Engineering
- Medical Imaging
Background:
- Hemodynamic monitoring is crucial in emergency medicine, particularly during fluid bolus therapy.
- Current methods for cardiac output assessment can be invasive or limited in real-time application.
Purpose of the Study:
- To develop and validate a non-contact method for monitoring cardiac output (CO) trends using facial photoplethysmography (PPG).
- To assess the system's efficacy in hemodynamic evaluation during intravenous fluid bolus therapy in an emergency setting.
Main Methods:
- A MATLAB-based pipeline was developed for CO trend monitoring using facial PPG.
- A deep learning framework was used for dynamic region-of-interest (ROI) segmentation on the forehead.
- Green (G) and Green-Red (G-R) PPG signal processing methods were applied to mitigate nonlinearity.
Main Results:
- The system demonstrated strong correlations between PPG-derived CO trends and ground truth values (average r=0.64 for G-R, r=0.59 for G).
- Peak correlation coefficients reached up to 0.87 in optimal ROIs and subjects.
- The non-contact approach showed potential for accurate hemodynamic assessment.
Conclusions:
- Camera-based CO monitoring using facial PPG is a viable, non-invasive tool for hemodynamic assessment.
- This technology offers a scalable solution for critical care and emergency applications, especially in resource-limited scenarios.
