This project will conduct research to develop a rapid and powerful electronic probe that can detect biochemical attributes of COVID-19 from exhaled breath in real-time and on-demand. Our approach is an alternative and complementary solution to rRT-PCR for large-scale COVID-19 testing. The proposed sensor is a novel biohazard aerosol analyzer probe, which is based on phase interrogated ultra-sensitive microwave resonance. The probe can detect physicochemical attributes of human lung capacities and compositions of breath aerosols. We hypothesize that the compositions of the breath aerosols (water, virus, bacterial) will correlate to permittivity signatures of specific pulmonary diseases, which can be extracted using machine learning algorithms. The probe will separate sick from healthy individuals through a rapid and definitive test of an individual's breath. The proposal focuses on defining the theoretical and experimental sensitivity and selectivity of the probe and addresses the following: Is it possible to detect signatures from COVID-19 and other diseases from exhaled breath in real-time? Computer simulations will be employed to investigate the fundamental electromagnetic parameters of a prototype probe to determine the theoretical limit of detection. Probes will be fabricated and used to identify size distributions and chemical compositions of innocuous aerosols and those containing materials simulating viruses and respiratory tract secretions to detect COVID-19. Machine learning will be employed to analyze aerosol data and identify diseased individuals.
Intellectual Merit: This work will advance the state-of-the-art of non-invasive point-of-care probes in the health care arena. The integration of machine learning data analysis with real-time and on-demand medical diagnostics is a novel contribution which will permit real-time evaluation of large-scale probe data and the concomitant detection of vectors of disease propagation.
Broader Impacts: This work will inspire engineers to quickly advance the proposed strategy for identifying COVID-19 and other pulmonary disease signatures from human breath in real-time so that testing of human breath will soon become standard medical practice worldwide. The work is multidisciplinary, involving optics, electronics, chemistry, physics, virology, and machine learning. Testing has been identified as a scarce and essential resource. This research will permanently and dramatically enhance the containment of pandemics.
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.
