TOLEDO, Ohio — Two University of Toledo students are using their talents to help predict how long people with COVID-19 will stay in a hospital.
Engineering and computer science doctoral student, Mohammadresa Nemati and mechanical engineering graduate student, Jamal Ansary have recently developed a machine learning model to predict the amount of time COVID-19 patients are hospitalized.
Not only are the two classmates and friends. They're roommates.
They say the idea for the model came from being in lockdown together at the beginning of COVID-19 and seeing what was going on in our area.
"We were stuck at home and we were watching the news. What was really concerning to us was that we saw a sudden surge in the number of people that are being admitted to hospitals," Jamal Ansary said.
According to the two graduate students, their predictive model can give hospitals better insight on how to control overloads.
"Knowing how long a patient is likely to be hospitalized in a hospital can help decision makers efficiently allocate equipment and facilities," Mohammadresa Nemati said.
Using clinical data from more than a thousand COVID-19 patients, their model can predict recovery time with 70 percent accuracy based solely on age and sex.
They say that wanted to offer their help in any way they could, and this research was their way.
"We are pretty happy about the research that we've done and there can be a lot of work in the future. There is much more data available. The fact that we have gave back to the community and give back to the people, it felt right for us," Ansary said.
You can read more about their predictive model here.