Our study is significant as it focuses on developing a de-identified linked database system via REDCap, for exploring COVID-19 surveillance, clinical, multi-omics and geospatial data to help understand and monitor transmission dynamics, natural history, virology and clinical outcomes.
The study is also highly feasible as a quick response to the ongoing COVID-19 pandemic. As a result of improved computational architecture and improved capabilities in data management and analytics software, we can now quickly build integrated multitudinal and multimodal datasets within secure Health Insurance Portability and Accountability Act (HIPAA) compliant analytic environments to support data annotation, reproducible analytics, and controlled access archiving and sharing. This will enable us learn quickly and share important emerging COVID-19 patterns and trends across South Carolina.
To better understand the short-term and long-term clinical outcomes for COVID-19 patients, this project proposes do this innovatively by deploying active learning as a central component of the Big Data science approach to this problem. To achieve this, this project focuses on the following Specific Aims:
Create a de-identified linked database system via REDCap and a mobile application (app) for collating surveillance, clinical, multi-omics and geospatial data on both COVID-19 patients and health workers (HW) treating COVID-19 patients in South Carolina.
Examine the natural history of COVID-19 including transmission dynamics, disease progression, and geospatial visualization.
Identify important predictors of short- and long-term clinical outcomes of COVID-19 patients in South Carolina using machine learning algorithms.
Xiaoming Li (Arnold School of Public Health) and Bankole Olatosi (Arnold School of Public Health), co-principal investigators of the UofSC Big Data Health Science Center – have been awarded a $1,252,550 grant from the National Institute of Allergy and Infectious Diseases to develop a state-wide data-driven system to fight COVID-19 in South Carolina. The multidisciplinary investigation team for this project also includes BDHSC faculty members; Neset Hikmet (College of Engineering and Computing), Jianjun Hu (College of Engineering and Computing), Zhenlong Li (College of Arts and Sciences), Sharon Weissman (School of Medicine-Columbia) and Jiajia Zhang (Arnold School of Public Health). Additionally, the project includes the collaborators: Caroline Rudisill from the Arnold School of Public Health and Michael Shtutman from the College of Pharmacy.