NSF/FDA SiR: Pulse Oximetry Measurement Errors Correlated with Patient Skin Pigmentation: Optical Mechanisms and Effect Multipliers

University Of Massachusetts Amherst, Amherst MA

Investigators

Abstract

Pulse oximeters are powerful and inexpensive devices commonly used for early detection of health problems and monitoring of treatment. However, studies have shown that patient skin pigmentation levels can affect the measurement accuracy. A better understanding of the underlying mechanisms will help the refinement of instrumentation and algorithms to improve the device performance. Computational modeling will be used to simulate light propagation in adult finger tissue and investigate the impact of different biological and device design factors on device performance. Laboratory measurements of tissue-mimicking models will be performed to validate the computational model under well-controlled conditions. Such effective, least-burdensome evaluation of device performance has the potential to lead to improvements in clinical performance that will impact patients undergoing surgical procedures, anesthesia, and other critical conditions. Furthermore, the planned research will be seamlessly integrated with educational, mentoring, and outreach activities to advance the cross-disciplinary program in the field of Biomedical Optics and Regulatory Sciences. This project aims to develop and validate a series of test methods for pulse oximetry. The components of this work include: (1) Fabricating physical models with biologically relevant properties of adult fingers (e.g., epidermal layers with different pigmentation levels, bio-mimetic microvascular networks, and mock cardiovascular flow loop with tunable oxygenation to generate real-time physiological flow/pressure waveforms); (2) Developing in silico models of adult fingers of different sizes and performing Monte Carlo simulations to elucidate light-tissue interactions in pulse oximetry; and (3) Validating computational models using spectroscopic measurements in finger-simulating physical models, and investigating effects of epidermal melanin content along with other biological factors (e.g., perfusion level, finger size) and device characteristics (e.g., source bandwidth, sensor reflectivity) on key light-tissue interaction parameters, as well as on estimated oxygen saturation measurement accuracy. This project will provide extensive insights into the factors influencing quantitative accuracy of pulse oximetry. 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.