Researchers led by Imon Banerjee, PhD, of Emory University and colleagues trained a machine-learning algorithm to predict a patient's need for hospitalization within seven days of a positive reverse transcription polymerase chain reaction (RT-PCR) test based on patient demographics, medication, past medical procedures, comorbidities, and laboratory results. The multimodal "fusion" AI model yielded a promising level of prediction performance -- an 84% overall F1 score.
"We conclude that fusion modeling using medical history and current treatment data can forecast the need for hospitalization for patients infected with COVID-19 at the time of the RT-PCR test," the authors wrote.
What else did they find? You'll have to attend this Wednesday afternoon presentation to learn more.
This paper received a Roadie 2021 award for the most popular abstract by page views in this Road to RSNA section.
