Machine learning-enabled classification and monitoring of COVID-19 immunity levels using a paper-based multiplexed sensor (Conference Presentation)

We report a paper-based multiplexed vertical flow assay (xVFA) that uses five structural proteins of SARS-CoV-2 to detect IgG and IgM antibodies in serum samples; this helped us not only define the COVID-19 immunity status of the individuals but also to track the temporal dynamics of the immune responses of individuals at different time points. Designed for improved portability, cost-effectiveness, accessibility, and accuracy in COVID-19 serology testing, our method achieved a blind testing accuracy of 89.5% in determining the immunity levels of individuals using a neural-network-based serodiagnostic algorithm, all performed in <20 min by requiring only 40 µL of serum for each test. Our xVFA platform is potentially scalable to different SARS-CoV-2 variants and COVID-19 vaccines and offers a rapid and accessible technology for monitoring the immune level of individuals against SARS-CoV-2.