Camera-SpO₂ Measurement Using Time-Division Multiplexed Near-Infrared Imaging: A Feasibility Study on DC-based...

Area of Science:

• Biomedical Optics
• Medical Imaging
• Physiological Monitoring

Background:

• Tissue oxygen saturation (StO₂) relates to blood oxygen saturation (SpO₂).
• Noncontact SpO₂ measurement using camera-captured optical signals is promising.
• DC signal variations due to lighting and skin tone hinder SpO₂ calibration.

Purpose of the Study:

• To compare AC- and DC-based SpO₂ calibration models using a single camera.
• To evaluate the Ratio-of-Ratios (RoR) principle for SpO₂ estimation.
• To analyze the strengths and limitations of different calibration approaches.

Main Methods:

• Employed time-division multiplexing for AC- and DC-based SpO₂ calibration.
• Conducted experiments on 11 adult participants using a breath-holding protocol.
• Compared RoR-DC 750/940 and RoR-AC 750/940 models.

Main Results:

• The RoR-DC 750/940 model showed superior SpO₂ estimation compared to RoR-AC 750/940.
• DC-based model achieved lower mean absolute error (1.63% vs. 2.55%).
• DC-based model demonstrated higher correlation (r = 0.75 vs. 0.27) with reference SpO₂.

Conclusions:

• Multi-wavelength time-division multiplexing enables effective SpO₂ tracking with a single camera.
• DC-based calibration is more sensitive for detecting SpO₂ drops.
• This technique offers a viable noncontact approach for SpO₂ monitoring.