Nearly everyone in public health will tell you that taming the coronavirus outbreak is going to require something not yet developed — the ability to quickly identify hot spots for the illness, then deploy resources for contact tracing and containment, over and over until the illness fades from its own lack of transmission.

Thanks to a partnership created last weekend bridging Mayo Clinic data analytic professionals with the Minnesota Department of Health, the statistical programming tools developed by the clinic to find hot zones of greater viral transmission in southern Minnesota will soon go statewide as well.

The method uses AI (artificial intelligence) to calculate and compare rates of positive to negative cases in a given region, stratifying the data by county, then determining if a greater portion of those tested in the region are turning up positive than elsewhere.

Newsletter signup for email alerts

"It really pops out," says Dr. Andrew Badley of the differential rates of coronavirus Mayo's AI has already discovered statewide. "We have a seen a wide difference between geographic areas."

The initiative is the extension of an partnership struck several years ago between Mayo Clinic and the Cambridge, Mass.-based data analytics firm Nference.

The objective of the new platform, Badley says, is to monitor where cases are arising most intensively, use the calculations to predict where more cases will occur, "then use that to make decisions about where to allocate testing resources and protective equipment."

"If you have an area with a rate of positive tests of 1%, that means you're doing a pretty good job. if you see an area with a positive test rate of 10%, that means you should focus more tests in that area. In statistical terms we're calculating a rate of infections, as opposed to a total number of cases."

The roll-out of the new tool focused on counties Mayo serves and quickly found a hot spot.

"Using this technology we found an area where there was a higher rate of positives for number of tests done," Badley says. "Because of that we've deployed more testing and personal protective equipment in that area. If we are able to diagnose more cases, we then could either quarantine or treat them as needed. Our hope is that already has impacted the spread of disease in that area."

The health department is sharing its data for use with the Mayo tool. It's one of many concerted efforts by state officials to build out homegrown ways of tracking and addressing the virus in the absence of federal assistance or direction and even federal obstruction.

"It is our intention to make these determinations as comprehensive within the state of Minnesota as possible," Badley says. After that, the clinic plans to offer it free to any state that wants to use it.

The platform faces the same narrowness of sample selection created by the need — due to testing shortages nationally — to test just select types of people, but does its best to work around them.

"This does not expand the testing pipeline," Badley says, "but it deploys it in a perhaps a more thoughtful way."

As a public service, we’ve opened this article to everyone regardless of subscription status. If this coverage is important to you, please consider supporting local journalism by clicking on the subscribe button in the upper right-hand corner of the homepage.

Minnesota Department of Health COVID-19 hotline: 651-201-3920.

COVID-19 discrimination hotline: 833-454-0148

Minnesota Department of Health COVID-19 website:  Coronavirus Disease (COVID-19) website.