A non-contact oxygen saturation detection method based on dynamic spectrum

Area of Science:

• Biomedical Engineering
• Optical Sensing
• Respiratory Medicine

Background:

• Blood oxygen saturation (SpO2) is crucial for monitoring respiratory conditions.
• Non-contact oximetry presents advantages for pandemic and sleep monitoring but faces technological and environmental challenges.
• Existing methods require physical contact, limiting widespread application.

Purpose of the Study:

• To develop and validate a novel non-contact method for SpO2 measurement.
• To utilize dynamic spectrum (DS) principles for enhanced accuracy in SpO2 detection.
• To address the limitations of current non-contact oximetry technologies.

Main Methods:

• A multispectral camera (660-950 nm) captured facial videos to derive temporal photoplethysmography (PPG) signals.
• Dynamic spectrum (DS) values were extracted from 24 wavelengths, followed by wavelength screening.
• A partial least squares (PLS) model was built using pulse oximetry as the reference standard.

Main Results:

• The developed method demonstrated a significant response to SpO2 variations.
• Regression coefficient (R) was 0.6366 and root mean square error (RMSE) was 0.9906.
• Prediction accuracy approached the ±2% range of commercial pulse oximeters.

Conclusions:

• The DS-based non-contact SpO2 measurement method effectively mitigates interferences from static tissue, individual differences, and environmental factors.
• This technique offers a viable new approach for non-contact oximetry, meeting significant market demand.
• The findings provide a foundation for improved remote respiratory monitoring solutions.