AI platform predicts COVID's potential spread in rural Tennessee

Apr 30, 2020 at 09:03 am by Staff


April 29, 2020 - Decode Health, a Nashville healthcare tech company that is tracking the spread and prevalence of the COVID-19 virus using its proprietary AI platform, predicts that rural counties near major population centers are likely to be where the second wave of the disease emerges

Because of travel between these counties and nearby large cities (i.e. for work, medical care, entertainment and shopping), the disease could re-emerge in large cities in the summer or fall, potentially at an accelerated rate, according to Decode's models. Counties statewide, and especially those around large cities, need to be vigilant now and in the coming weeks about adhering to social distancing and other preventative measures as businesses reopen in accordance with the Tennessee Pledge.

The good news is that Decode Health's overall findings show that Tennessee is outperforming contiguous states in managing the spread of the disease -- with excellent progress in counties with large cities, including Davidson County.

Decode's AI software generates predictive models based on publicly released daily statistics and proprietary social determinants of health data. The models provide insights at the national, state and county levels to predict risk for future infection and poor outcomes. Their goal is to: (1) spotlight counties where infections are likely to emerge, (2) define characteristics for counties where the worst outcomes, including death, are likely to occur, and (3) monitor infection rates to understand positive or negative trends weeks in advance.

"Predictive analytics is the key to tracking re-emergence and swiftly identifying trends that enable proactive patient engagement at the national, state and community levels," said Decode Health CEO Chase Spurlock. "As we think about reopening our economy, it's important to identify disease trends quickly to avoid costly shutdowns. We can't afford to repeat this process."

Decode's modeling platform, developed over the last five years, has previously been used to identify and monitor high-risk patients for chronic diseases including multiple sclerosis, Crohn's disease, ulcerative colitis and chronic kidney disease. This platform now provides crucial insight into COVID-19 trends that indicate growing infection rates and probabilities for poor outcomes, including hospitalization and death.

The Tennessee map above, created by Decode, breaks down the level of COVID-19 vulnerability by county, with those in red representing the highest risk of poor outcomes.

Sections: COVID