CDS&E: Novel Computational Models for Smart Wearable Blood Gas Monitor for Infants

Despite recent developments in medical wearable technology, monitoring of respiration parameters, compared to other vital signs, has yet to be sufficiently investigated. The majority of respiration research has focused heavily on respiration rate and blood oxygen saturation. Despite being well-known, these are only a subset of respiration parameters.

Beyond that, blood oxygen and carbon dioxide partial pressures - called blood gases - have medical significance and the potential of being measured noninvasively. This research specifically addresses a serious and global problem, respiration disorders of infants. Respiratory distress is one of the most common reasons for infant admission to the neonatal intensive care unit.

Studies have shown that after successful medical treatment in the neonatal intensive care unit, earlier discharge to a home life with remote monitoring infrastructures could improve the survival chances of these at-risk infants. Monitoring patients with respiratory disorders from a home setting could enhance the health and well-being of individuals of all ages, from neonatal patients with developing lungs to adults with chronic obstructive pulmonary disease (COPD), while creating and bringing new large data sets to the attention of clinicians.

Currently, conventional blood gas monitors employed in corded-bedside equipment can find place in clinical settings. However, a wireless miniaturized blood gas monitoring system supported with computational models can extend the use beyond the hospital for long-term care. Besides these potential benefits to society, this project advances hardware-software co-design for wireless wearable sensors.

Miniaturized blood gas research offers unique research experiences for undergraduate and graduate students to understand technical details of different disciplines and develop translational biomedical skills. The interdisciplinary team provides opportunities for underrepresented students to encourage them to pursue careers in research and STEM. The curriculum development allows to introduce interdisciplinary topics and hands-on experimental experiences to students.

The goal of this project is to develop a hardware-software integration framework for a miniaturized, noninvasive, wireless, luminescence-based oxygen sensing wearable device with an open-source software toolkit. The toolkit accurately translates the blood oxygen levels by utilizing measured transcutaneous oxygen diffused through the skin. Within this framework, computational models of oxygen transport is used to iteratively design an estimation algorithm for the intelligent sensor.

This sensor is resilient to noise arising from intra- and interpersonal variations, environmental factors, and sensor hardware operations. The core scientific contributions include 1) the presentation of computational models of oxygen transport incorporating factors affecting sensor readings such as weight, age, gender, sensor degradation, for building an estimation engine, for the first time in literature; 2) the creation of a novel miniaturized custom-designed wearable that measures transcutaneous oxygen; 3) the integration of hardware and software on a flexible miniaturized wearable device for long-term comfortable wear; 4) the determination of the computational models’ accuracy, the sensitivity of overall system, and the specificity of the sensor by capturing the dynamic respiratory physiological status of individuals with longitudinal data.

The developed hardware-software platform is validated on humans. Having the ability of sensing vital blood oxygen parameter with a wearable supported with computational models is unique, filling an important gap in the miniaturization of the transcutaneous blood gas sensor for noninvasive wearable device applications.

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.