Imagine having an inexpensive (free?), 15-second screen for COVID-19 -- what a powerful weapon against this virus! It is possible that a neural network applied to the ECG might be that weapon. We hope to find out, but need help. @EricTopol @SurajKapa @MayoClinic
Animal data suggest that days after coronavirus infection (RbCV) ECG changes are visible. ncbi.nlm.nih.gov/pubmed/10037086 Human experience in COVID-10 patients includes ECG and rhythm changes, and troponin elevation. While non-specific to the human observer, #AI may help.
We have previously used #AI applied to the ECG to identify silent and subtle disease nature.com/articles/s4159… thelancet.com/journals/lance… . Using smartphone electrodes makes this approach massively scalable. @MayoClinicCV @noseworthypeter @zachia5 @preventingCVD
We propose to collect ECGs from patients and controls from around the world to see whether we can use our existing framework to create an ECG-based screen for COVID-19, feasible if cardiac involvement results in a detectable signal (see CCA-ACC, by Dr. Cao) 
The goal is a rapid screen and prognostic test for patients with COVID-19. A few basic principles:
1. If a test is developed, it will be made available globally at no charge
2. All contributors will be acknowledged
1. If a test is developed, it will be made available globally at no charge
2. All contributors will be acknowledged
If you are a medical provider who cares for patients with COVID-19 and would like to participate in this project, send us an email note at: covidECG@mayo.edu
Thank you for considering - working together we can hopefully limit human suffering from this scourge.
Thank you for considering - working together we can hopefully limit human suffering from this scourge.
