I-Corps: Translation Potential of a Skin-Tone-Aware Pulse Oximeter

This I-Corps project is based on the development of a pulse oximeter that corrects for skin tone. Pulse oximeters are used in hospitals, ambulances, homes, and wearable health products. However, inaccurate oxygen readings can miss or delay detection of low blood oxygen, a problem that affects millions of patients who rely on non-invasive monitoring and has been shown to vary with skin tone. Blood oxygen levels are one of the first metrics by which medical professionals evaluate patient health, making accurate measurements using the standard of care - a pulse oximeter - critical for treatment and intervention. This technology combines additional light-based measurements with a software correction method to account for skin pigmentation at the measurement site, improving oxygen estimates without requiring a complete replacement of existing monitoring systems. It is designed to improve the sensitivity of the pulse oximeter to varying oxygen levels by improving accuracy across patient populations at lower oxygen levels, where clinical decision-making is most urgent. This technology may be used in critical care, anesthesia, emergency medicine, home health, remote patient monitoring, and consumer wearables, which may improve clinical and patient outcomes. This I-Corps project utilizes experiential learning coupled with first-hand investigation of the industry ecosystem to assess the translation potential of a multiwavelength, hardware–algorithm pulse oximetry correction platform. The technology uses optical sensing at selected wavelengths to estimate local skin pigmentation, a nonlinear regression model to correct peripheral oxygen saturation estimates, wavelength-specific operating regions to preserve signal quality at low oxygen saturation, and narrow-spectral-width light sources to improve sensitivity to oxygenation changes. The key scientific advance is a correction layer that can estimate pigmentation at the measurement site and adjust oxygen readings in routine non-invasive monitoring, enabling users to obtain more dependable data while preserving familiar workflows and existing device infrastructure. The technical results include bench testing, peer-reviewed characterization of skin-tone-dependent error sources, prototype development, and early clinical evaluation. This technology may improve the accuracy of pulse oximeters across patient populations as well as improve patient outcomes and mortality. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.