Skip to main content

Boise State researchers use wastewater and machine learning to predict deaths and hospitalizations due to COVID

The Boise State University Wastewater Epidemiology Laboratory led by Greg Hampikian, professor in the Department of Biology, recently published the first machine learning model to predict deaths and hospitalizations due to COVID—by counting the viruses in wastewater. The model showed that peaks in sewage treatment virus particles matched peaks in hospitalizations 14 days later, and peaks in deaths 23 days later. The peer-reviewed paper appears in the latest issue of Science of the Total Environment, and can be accessed for free on the Science Direct website.

Authors of the paper include recent Boise State graduates Ernie Ogden and Swarna Kanchan who both worked in the lab, along with four undergraduate biology majors, a master of science student and a postdoctoral researcher from the lab.

“This is really a dream come true,” Hampikian said. “We started pitching the idea that sewage was a valuable health asset more than a decade ago. No one believed us, now it has become a standard epidemiological resource, and we have shown it can predict both hospitalizations and deaths. This gives hospitals and care givers more time to prepare and allocate lifesaving resources. We believe that our model can be used for a host of diseases.”

For additional information, contact Greg Hampikian at (208) 781-0438 or greghampikian@boisestate.edu.