Proof of Concept Low-Cost Arterial-Venous Oxygen Sensing
A real-time, non-invasive method for measuring venous oxygen saturation (SvO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>) would offer promising outcomes for ICU care, especially in diagnosing sepsis and optimizing care, for mechanical ventilation and blood transfusions. This research proposes a low-cost (US$170), proof-of-concept, reflectance optical sensor capable of measuring arterial-venous oxygen saturations at the finger. By introducing a digit cuff actuator system, pulsations at low frequency and pressure were induced in the venous system, allowing separate processing of arterial and venous signals. Three different methods were utilized for signal processing in MATLAB: Peak and Trough Detection, Fast Fourier Transforms, and Power Spectral Density. An evaluation trial with five healthy adult subjects demonstrated the sensor's ability to estimate SpaO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> and SpvO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>, with results comparable to literature values. While SpaO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> values trended lower than expected, with median SpaO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> ranging from 89.27% to 93.78%, SpvO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> values remain within the normal range of 60-90% with subject trial medians between 76.99% and 84.30%. Small standard deviations between tests and estimation methods confirmed the reliability of the system, with consistent arterial estimations between normal and induced pulsation tests, proving both arterial and venous saturations can be measured and estimated near simultaneously. SpaO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> values are likely lower due to a lack of calibration with a blood gas analyzer and from using an empirically derived model equation determined for a transmittance sensor. However, the sensor with calibration and validation on a large test cohort shows potential for low-cost, non-invasive estimations of both SpaO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> and SpvO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> and thus oxygen extraction monitoring, providing potential for improving patient outcomes while reducing risks and costs.
Authors
- J. Geoffrey Chase (ORCID: https://orcid.org/0000-0001-9989-4849)
- Jake Campbell
- Josephine A. Dixion
- Jordan F. Hill (ORCID: https://orcid.org/0009-0001-2206-7615)
- Christopher G. Pretty
Institutions
- University of Canterbury (NZ)
Publication Details
- Published
- 2024-09-02
- DOI
- https://doi.org/10.1109/mesa61532.2024.10704852
- Primary Topic
- Hemodynamic Monitoring and Therapy
- Type
- article
- Field-Weighted Citation Impact
- 0.00