Researchers from New Haven, Conn.-based Yale University have developed an algorithm that can predict patients' disease severity and hospital lengths of stay.
The researchers, which also included scientists from foreign universities, based the tool on COVID-19 data but say it could be applied to other viral and disease outbreaks.
"Precision public health — that is the whole idea," senior author Vasilis Vasiliou, PhD, a Yale epidemiologist, told Politico in a Sept. 5 story. "The power that these algorithms are giving you is something amazing."
The study, published Aug. 29 in Human Genomics, compared 111 COVID-19 patients admitted to Yale New Haven Hospital in 2020 to a control group of 342 healthy medical professionals, finding biomarkers that aligned with disease severity. The researchers' artificial intelligence-assisted triage platform includes a clinical decision tree, hospitalization estimator, and disease severity predictor.
The researchers noted that the data was collected before COVID-19 vaccines and many treatments for the virus, which could reduce the biomarker changes, and that the hospitalized patients were largely Black while the control group was mostly white, not excluding race as a factor.
"Being able to predict which patients can be sent home and those possibly needing intensive care unit admission is critical for health officials seeking to optimize patient health outcomes and use hospital resources most efficiently during an outbreak," Dr. Vasiliou said in an Aug. 28 Yale news release.