Abstract

Development and Deployment of a 'Smart Diagnostic Ecosystem' for Covid-19 and Beyond: Near Real-Time Customization of Programmable Medical Microdevices

John T McDevitt, Chair, Department Biomaterials & Biomimetics, New York University College of Dentistry

SARS-CoV-2, the virus that causes coronavirus disease (COVID-19), has reached pandemic levels and resulted in significant morbidity and mortality that has affected every inhabited continent. Delays in diagnostic testing and the inability to track infections on a local, international and global basis has hampered the management of the COVID19 pandemic. This pathogenic coronavirus has had a disproportionate impact on the underserved and impoverished. While COVID-19 has yielded devastating consequences, the pandemic has also opened the door for acceleration of core diagnostic capabilities that have the potential to lead to lasting impact for our society. With this vantage point in mid, in the recent past we have launched a series of efforts that target the development and deployment of a ‘smart diagnostics ecosystem’. Here we integrate programmable chip-based diagnostic systems capable of multiplexed measurements alongside clinical decision support tools that utilize strategically chosen nonclinical data elements that elicit signatures that can be used to capture diseases before they spiral out of control, including COVID-19. Using newly established diagnostic models that leverage large COVID-19 databases we have developed, validated and scaled a clinical decision support system and mobile app to assist in COVID-19 severity assessment, management, and care. Model training data from 701 patients with COVID-19 were collected across practices within the Family Health Centers network at New York University Langone Health. A two-tiered model was developed. Tier 1 uses easily available, nonlaboratory data to help determine whether biomarker-based testing and/or hospitalization is necessary. Tier 2 predicts probability of mortality using biomarker measurements (CRP, PCT, D-dimer) and age. Both Tier 1 and Tier 2 models were validated using two external datasets from hospitals in Wuhan, China comprising 160 and 375 patients, respectively. These capabilities have been extended recently to screening of dental patients as well as to the developments of on integrated immunity scoring system. The ecosystem now includes cloud connected databases and AI linked clinical decision support tools that were developed initially for COVID-19, but are now helping to accelerate the arrival of new oncology tools.