Exploring Arterial and Venous Oxygenation Separation with Multichannel PPG^*

State-of-the-art pulse oximeters capture only arterial blood oxygen <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\text{SaO}_{2})$</tex> but not venous blood oxygen <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(\text{SvO}_{2})$</tex>. <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{SvO}_{2}$</tex> is crucial for evaluating the body's oxygen utilization, providing essential insights into cardiovascular and respiratory function. <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{SvO}_{2}$</tex> is difficult to measure noninvasively due to the deeper locations and lower pulsatility of the veins compared to the arteries. To explore the possibility of noninvasive assessment of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{SvO}_{2}$</tex>, we hypothesize that reflective photoplethysmography (PPG) measured on the skin surface over a shallow vein, such as the radial vein, captures the pulsatile absorbance from the vein. However, the difficulty lies in the crosstalk in the PPG signals induced by the adjacent artery. To tackle this challenge, we built a multichannel wireless PPG array, applied Laplacian filtering, and proposed a physiologically inspired spatial filtering method to distinguish arterial vs. venous <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{R}$</tex> values, defined as the ratio of the pulsatile absorbance at two different wavelengths, typically red and infrared (IR). The efficacy of our approach is demonstrated through preliminary in vitro and in vivo experiments.