Cardio-Pulmonary Imaging for COVID-19 Patients

We (the named investigators all potential PIs) and others (about 30 all together) have assembled around the theme of imaging COVID-19 patients from diagnosis through recovery to better understand the disease, progression and its outcomes. To approach this we want to start with building a cloud-based imaging-driven omic database / repository for COVID-19 patients and survivors. This will allow us to acquire, archive, curate, pool and share large data sets to which we can apply modern machine learning and artificial intelligence approaches to correlating and interpreting the many unknown and often subtle parameters of this disease. While we plan to ultimately collect all data, phenotypic and genomic, and all other metadata from large cohorts of diverse patients that we have at Columbia-Cornell-Presbyterian and in NYC, we can start immediately to collect preliminary results on our state-of-the-art MR systems at the Zuckerman Institute where we are already fully integrated with the Google Cloud Platform and are linked through the cloud with other collaborators in NYC and beyond. 

To fund this work, we are beginning to formulate a number of research grant applications covering the bases of data acquisition and operations for COVID-19 diagnoses, pathologies, research, and therapy development. A more clinically applied proposal is now being assembled for the NIHLBI with the approximate aims:
 1) Cardiac structure and clinical outcomes
 2) Cardiac function and clinical outcomes
 3) Cardiac/cardiopulmonary imaging data acquisition and data curation

Another more engineering oriented application answering a call from the NIH-NIBIB will build a COVID data repository along the lines of Andrew Laine's suggestion below:
 “Imaging-driven omics database / repository on COVID survivors aggregating (1) admission, (2) in-care, (3) discharge and (4) follow-up data for retrospective understanding of COVID disease and planning for future care.” Sairam Geethanath has already applied for a Fast Grant from private sponsors to build an AI toolkit for operating on and analyzing COVID patient data for biomarkers. 

Finally, As PI I am working with this group to pull all of this interest and these multiple applications together under an NIBIB P41 Bioimaging Center Grant which will focus on three themes: Neuro, Cardio-pulmonary, and Cancer. Clearly, COVID-19 research will play a major role all of these areas.