Researchers from Chapel Hill, N.C.-based UNC School of Medicine's machine learning models helped identify long-COVID patients, patients who had the virus for more than 28 days, by analyzing patterns in electronic health record data, according to a May 16 study published in The Lancet Digital Health.
Researchers developed and trained three machine learning models using demographics, healthcare utilization, diagnosis and medication data for 97,995 adults with COVID-19 and 597 patients from a long-COVID clinic, to identify potential long-COVID among all patients with COVID-19, patients hospitalized with COVID-19 and patients who had COVID-19 but were not hospitalized.
The results found that the machine learning models were able to accurately identify patients who potentially have long-COVID.
According to the study, there is an urgent need to understand long-COVID, identify treatments and accurately identify who has it. These machine learning models could help correctly classify and manage patients with long-COVID.